________________________________________________________________________________________________________________

HBS Professor Michael Beer and Sunru Yong prepared this case solely as a basis for class discussion and not as an endorsement, a source of
primary data, or an illustration of effective or ineffective management.

This case, though based on real events, is fictionalized, and any resemblance to actual persons or entities is coincidental. There are occasional
references to actual companies in the narration.

Copyright © 2008 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685,
write Harvard Business Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. This publication may not be digitized,
photocopied, or otherwise reproduced, posted, or transmitted, without the permission of Harvard Business School.

M I C H A E L B E E R

S U N R U Y O N G

TerraCog Global Positioning Systems:
Conflict and Communication on Project Aerial

Emma Richardson squinted at the TerraCog GPS (Global Positioning System) prototype in her
hand. She zoomed in until the display showed a clearer satellite photo of the lake 200 feet in front of
her and into which her Labrador had already happily bounded. Most weekends, Richardson made
the hike to the lake to clear her mind and, on occasion, to test new GPS models from her employer,
TerraCog, Inc. Unfortunately, with the “Project Aerial” launch meeting scheduled for the next day, it
was difficult to enjoy this particular hike. Emma wondered how to get all parties to reach an
agreement on the price point for Aerial. TerraCog had started losing share to a competitor, Posthaste,
and it was imperative to get the new product to market.

Arriving at the lake, Richardson gave in to the urge to check her phone and grimaced as she
noticed two new voicemails. The first message was from Allen Roth, the director of design &
development (see Exhibits 1 and 2 for an organizational chart and brief biographies of key
managers):

“Emma, it’s Allen. Listen, Tony and I have been over these cost numbers on Aerial.
We cut all that we could and we ended up with only a 7% or 8% reduction to cost.
Unfortunately, I don’t think this will get us to the price point that Sales is looking for.
But I don’t need to remind you that we gave Sales the features and functionality they
wanted in Aerial, so I’m not going back now to ask my team to do the impossible. We’ll
hash it out tomorrow, but I figured it best you hear it from me.”

The second message was from her boss, Richard Fiero, the company president:

“Emma, I wanted to check on Aerial. I heard grumbling from Ed and the sales team
on Friday. They seemed frustrated with Tony Barren’s production team. Make sure
Production has its act together. Tony should know he’s on thin ice after the recent

2 1 8 4
A P R I L 1 1 , 2 0 0 8

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

2184 | TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial

2 BRIEFCASES | HARVARD BUSINESS SCHOOL

production fiasco on that sonar project—he’s got to succeed on Aerial. We need to have
Aerial on shelves at the start of Q3. Some board members are worried, so Aerial will be
near the top of the agenda at the board meeting next month.”

Neither message was encouraging. The Aerial meeting the next day, involving the sales, design &
development, and production departments, was now guaranteed to be contentious.

It was March 2008—only two months since Richardson had been promoted to executive vice
president. Fiero had tasked her with moving TerraCog toward greater operational alignment and
increasing cross-departmental cooperation. Richardson had already been tested by both inventory
problems and quality issues, which had led to significant tension between the U.S. headquarters in
Chicago and the production team in Shenzhen, China. Now, disagreement over the proposed price
point for Aerial threatened to derail the launch of the prototype in her hand.

Company and Industry History
TerraCog was a privately held company specializing in high-quality Global Positioning System

(GPS) and fishing sonar equipment. Founded in 1977, TerraCog got its start manufacturing high-end
sonar equipment for serious sport fishermen and boaters. In the late 1990s, the company had
introduced its first GPS products, marketed specifically to hunters, hikers, and campers.

Management believed that it was the company’s skill at translating retailer and user feedback into
exceptional product design and functionality that fueled the growth of its GPS business. Through
attentive channel management and, as Fiero put it, “a deep understanding of what specialty retailers
needed,” TerraCog had developed strong relationships with its key accounts. Fiero also believed that
TerraCog’s grasp of its consumers’ preferences and usage had given it an edge over GPS
manufacturers whose core business was in automotive applications. The firm had built its GPS line
for the serious outdoor enthusiasts’ market, and the products had won plaudits for durability and
value-added features like the integrated compass and barometric altimeter. Moreover, industry
reports indicated that the TerraCog GPS outperformed competing products on navigation.
TerraCog’s proprietary firmware—a custom computer program embedded into hardware that “ran”
functions—optimized the GPS chipset’s Wide Area Augmentation System capability, which provided
more precise navigation.

The company was not always first to market. In fact, TerraCog had found it was free to lag in
technological innovation with little risk because, when the company finally introduced new products,
they surpassed those of competitors in addressing customer needs. Customer word-of-mouth
recommendations had given TerraCog strong momentum with its handheld GPS. In early 2007,
TerraCog prepared to enter new, underserved GPS sub-markets, including cycling and fitness
applications.

“Google Earth™ for your GPS”
At the Summer 2006 Outdoor Retailer Show—the biggest trade show for vendors of outdoor

goods—a competitor, Posthaste, had unveiled a GPS prototype called “BirdsI” that displayed satellite
imagery. The imagery was not live, but rather static satellite photographs that had been “stitched”
into a single view. This was a marked improvement on the simple, vector-based graphics used by the
rest of the industry (see Exhibit 3 for a comparison). This did not impress the TerraCog team. The
imagery was crisp and had a certain visual appeal, but TerraCog’s research showed that BirdsI

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial | 2184

HARVARD BUSINESS SCHOOL | BRIEFCASES 3

technology did not offer substantive performance improvement over the standard maps in
TerraCog’s GPS system. Furthermore, the TerraCog team was convinced that Posthaste’s receiver
lagged TerraCog’s product in both accuracy and reception quality.

While the TerraCog team dismissed the Posthaste concept, a number of key buyers and product
reviewers found it an exciting innovation. One magazine reviewer observed, “Imagine having
Google Earth™ built into your GPS—it’s much more compelling to look at an actual satellite image
than to have yellow for land, blue blobs for water, and grey squiggles for roads.” Based on the buzz,
TerraCog’s executives debated whether to upgrade to satellite imagery. However, they realized that
adding the feature to the existing GPS platform required upgrades to processor speed and memory,
as well as new firmware. After some deliberation, the company dropped the idea as a non-essential
fad. TerraCog’s management remained confident that the company’s core customers were discerning
purchasers who would value durability and performance over dressed-up graphics.

In October 2006, with much fanfare, Posthaste introduced BirdsI as “the only handheld GPS with
satellite imagery.” BirdsI had an exclusive launch at two major national outdoor retailers, both of
which were key accounts for TerraCog. Within two months, TerraCog’s sales representatives in the
field reported impressive sell-through rates for BirdsI nationwide. While the product’s success
surprised TerraCog, management attributed it to the ebullience of the holiday shopping season. The
TerraCog team was confident that the popularity of BirdsI wouldn’t last.

Project Aerial

However, by spring 2007 TerraCog’s sales reps were noticing increasing customer demand for a
GPS with satellite imagery like BirdsI. Ed Pryor, vice-president of sales, began pressing for a reversal
of the decision not to develop the product. “It’s embarrassing to have no answers for our retailers
when they ask for our version of this,” he said. “Look at it from our perspective. We’ve changed the
compensation plan for the whole Sales team—including me—so we take a real hit if we don’t reach
our sales targets. Customers now want something different, and I can’t tell my reps we have no
plans to develop the product they need to hit those targets.” In response to these repeated requests,
TerraCog’s president, Richard Fiero, changed his mind on satellite imagery, if only to satisfy the
“gadget” appeal of such an innovation. The initiative was dubbed Project Aerial. In order to speed
development and avoid the costs of new moldings and major reconfiguration, the team decided to
redesign within the existing GPS platform.

Shortly after making the decision to proceed with Aerial, Fiero and Pryor met with Allen Roth,
director of design & development. Roth brought his key managers to the meeting: Cory Wu, who
oversaw software and firmware, and Alice Gorga, who managed hardware design.

RICHARD FIERO: Allen, we’re obviously in a hurry to get to market. But we don’t want something
slapped together – let’s make sure we get this product completely right the first time. Our reputation
for quality is paramount.

ALLEN ROTH: Understood. Are we including all the same features that we have in our current
GPS line?

ED PRYOR: Yes. We plan to offer Aerial at approximately a $50 retail premium to the current top-
of-the-line GPS, so it’s important to maintain the same high-end functionality.

ALLEN ROTH: What about speed? Satellite imagery requires a lot of processing power, so without
some serious juicing, Aerial might run slower than you’d like.

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

2184 | TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial

4 BRIEFCASES | HARVARD BUSINESS SCHOOL

ED PRYOR: I think we’ll be okay there, Allen. Our consumers are tech-savvy—they know there’s an
inherent trade-off to get more sophisticated graphics.

As the meeting ended, Roth indicated that they would have to do some careful planning to keep
costs as low as possible, but he was sure the product design could be completed by year’s end. At
that point, they could hand it off to production to develop detailed cost estimates, which would allow
the sales team, in consultation with finance, to determine pricing and develop a go-to-market plan.
Given the manufacturing lead-time, TerraCog expected to get Aerial to stores by the 2008 holiday
season (see Exhibit 4 for timeline of events).

The product development team members did not greet the Aerial decision with enthusiasm. First,
they felt that a redesign of the total platform—including firmware, external case, internal
components, and TFT (thin-film transistor) display—was feasible if management could extend time
to market by six more months; the resulting product would be superior and the project would be
more stimulating technically to the team members. Second, they had several other ideas for new
products that they believed would position TerraCog to capitalize on growth in cycling and fitness
GPS applications, and Project Aerial was forcing them to put aside these more exciting projects.
Finally, with company co-founder Harold Whistler preparing for retirement, Roth was eager to prove
his readiness to be the next VP of design & development. The Aerial project impeded his building of
a product line he could truly call his own.

Aerial Pre-Launch Meeting

As promised, the design team completed Aerial by the end of 2007. Late in January 2008, the
production team received the design specifications it needed to establish production methods,
conduct a pilot run, and estimate costs. As the new executive vice president, Emma Richardson was
tasked with overseeing the product launch. She scheduled a launch meeting in early March with
sales, production, and design & development.

In the past, Fiero and Whistler had been very involved in new products and tended to make quick
decisions. TerraCog’s growth forced Fiero to take a step back from the launch process, while
Whistler had cut back to part-time hours. There were many more employees involved in Aerial than
in past product launches, and Richardson worried that the size of the group might threaten the focus
and thwart decision making. She needed to finalize decisions on costs, pricing, and initial production
volume.

At the start of the pre-launch meeting on March 7, Richardson looked down the table, seeing Ed
Pryor, Allen Roth, production director Tony Barren, Cory Wu representing software and firmware,
and Alice Gorga representing hardware. Richardson opened the meeting by asking Barren to present
his cost estimates. Barren looked around grimly and did not mince words: “This thing’s expensive to
build. It looks the same, but Aerial’s got higher-end components and it’s more complex to
manufacture.” He gave a high-level overview of product-cost breakdown and concluded by saying,
“I’ll be blunt. You’re going to have to sell this product for a lot more than you thought. If anything,
we have been too aggressive in our cost estimate. We can’t lower it beyond what I’ve presented.”

ED PRYOR: I know you think your estimates are sound, but that isn’t going to help us. With these
numbers, we would have to sell Aerial for $550 to maintain our margin. We’d be two years late to
market with a price point $100 over BirdsI.

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial | 2184

HARVARD BUSINESS SCHOOL | BRIEFCASES 5

CORY WU: Tony, those cost estimates are surprisingly high. We tweaked the firmware without
overhauling it, so it’s basically the same components. It doesn’t seem justified that the costs should
come out as you say.

ALICE GORGA: I’m not sure, Cory. Those costs look realistic, given how my team upgraded the
hardware. Sales probably just needs to reconsider how to position this thing. I think —

ED PRYOR: Wait, are we here to talk about positioning or pricing? Fiero and I already figured out
how we’ll position the product, so let’s just get the pricing straightened out.

TONY BARREN: Well, we don’t perform miracles. The cost won’t change, and I’m not going to cut
corners in production. I had my head handed to me the last time we had quality issues.

ED PRYOR: Allen, your designers must be able to tweak something, right? Hate to say it, but maybe
you guys need to go back to the drawing board and figure out how to solve this problem.

ALLEN ROTH: Ed, we’re well into a couple of other projects now. Frankly, I don’t think this is
Sales’ call to make. We already put other projects on hold for Aerial, and we’ve given you what you
requested.

The discussion continued, but it became clear the group was at an impasse. Richardson suggested
ending the meeting: “Why don’t we wrap it up for now and meet again next week? In the meantime
Tony and the Design team should look for opportunities to cut these costs.”

As the meeting adjourned, Pryor announced to the room, “If we can’t lower these costs and fix the
finances on Aerial, I can’t sell it. I won’t try.”

Resumption of Aerial Pre-Launch Meeting One Week Later

Roth and Barren had spent much of the intervening week reviewing the cost estimates. When the
team re-gathered on March 14, the participants seated themselves around the conference table (see
Exhibit 5 for a seating chart). Becky Timmons, the CFO, was in attendance. At the last minute,
Harold Whistler also decided to join the meeting.

As Emma Richardson passed out copies of the new cost estimates, she explained that Roth and
Barren had agreed to make minor changes to the Aerial prototype, and that they now felt it could be
produced for approximately 8% less than the prior week’s estimate. On this basis, the Aerial could be
priced at $475, about $100 more than the current full-featured TerraCog GPS.

A long silence followed, then Cory Wu spoke up. “Eight percent—that’s all? I don’t understand it.
I’d like to know where the differences lie between our costs and PostHaste’s on BirdsI. There’s got to
be room for more cuts.”

Barren snorted derisively. “You can’t wish away the costs,” he said. “We’ve cut what we can.
Last time we got pressured into being too aggressive on cost estimates and then we got burned when
the price of plastic went through the roof. I’m not making that mistake again.”

Allen Roth concurred, pointing to the drivers of the cost increase: “Incorporating the satellite
imagery requires five times as much memory as our standard graphics. That increases cost—but if
you cut it, you undermine Aerial’s value proposition. Then we also did some reconfiguration
internally, and that increases the labor required to put one of these together.” He paused, surveying
the frustrated faces around him. “I don’t like the situation any more than the rest of you do, but

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

2184 | TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial

6 BRIEFCASES | HARVARD BUSINESS SCHOOL

we’ve got to be realistic. Look at the numbers in front of you—there’s nothing we can do to further
reduce costs.”

As everyone scrutinized the new cost estimates, the meeting broke into several conversations. Ed
Pryor and Richardson huddled together, while Allen Roth and Tony Barren carried on a conversation
with Harold Whistler. After several minutes, Richardson realized she needed to get the discussion
back on track. She addressed the entire group, saying, “We have the estimates, so we just have to set
a price that makes sense for the company. What do you think, Ed?”

ED PRYOR: We have to consider the selling price of the Aerial relative to competition. Posthaste is
selling at around $250 to dealers, which means they retail for around $400. Garmin just announced
their satellite-image version, which will hit shelves at somewhere around $395 MSRP. You all are
talking about $475 retail, and that’s too high. We have to be in the ball park with our list price or we’ll
be shut out of the game.

EMMA RICHARDSON: What should list price be?

ED PRYOR: $425 tops—but we should be lower than that if we are going to be aggressive at
recapturing lost share. Let’s not kid ourselves. The way we’re trending, same-store sales will be
down 10% this year. And this is with the GPS handheld market growing.

HAROLD WHISTLER: What if we relax our margin requirements for once?

BECKY TIMMONS: Absolutely not. We’re cutting it close already.

HAROLD WHISTLER: Okay, then how about a redesign? Let’s go to market with what we have,
and I’ll have my team take another look at possible changes that we can incorporate later.

ALLEN ROTH: I’m afraid that’s wishful thinking, Harold. Given the product requirements the sales
team called for, the cost is as low as it will get.

CORY WU: I don’t know about that. The changes we made to meet sales’ requirements were not
enormous. Why would they contribute to such a large increase in projected labor costs?

TONY BARREN: Cory, you and Harold can run the numbers for yourself. Then you’ll see that these
high costs are real.

BECKY TIMMONS: I’d still feel more comfortable if we could price it at $500 retail. With fuel costs
still rising, the cost to get these here from Shenzhen will only increase, and we run the risk of our
margins really getting squeezed.

ED PRYOR: Becky, you don’t understand how competitive this market is! Even at $475, why would
anyone give us shelf space? We are late to market and we’d be pricing at a substantial premium.
And is the product superior enough to justify that premium? I’ve been trying out our prototypes and
I’m concerned about the speed. The update speed is terrible, and switching between functions is
just —

ALLEN ROTH: Come on, don’t start talking speed now. We were clear from the start that we would
trade some speed to get new functionality and features.

ED PRYOR: Well, my sales managers are going to be fuming. Fiero told them Aerial would be
available at $400, and now you are talking about a minimum of $75 more than that. I still want to see
Cory or Harold take a crack at reducing unit cost.

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial | 2184

HARVARD BUSINESS SCHOOL | BRIEFCASES 7

Emma Richardson took a deep breath. The company needed a “go/no-go” decision on the
existing Aerial, and whether to do so at a competitive price in the hope that costs might be cut in the
future, or at a high price. She wondered fleetingly what the consequences might be if the company
abandoned Aerial altogether. As things stood, the arguments and finger-pointing were bound to
continue, and the group would never come to a decision on its own.

Richardson would have to push them to one.

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

2184 | TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial

8 BRIEFCASES | HARVARD BUSINESS SCHOOL

Exhibit 1 TerraCog Management Organization Chart, 2008

President
Richard Fiero

CFO
Becky Timmons

VP, Sales
Ed Pryor

VP, Design &
Development

Harold W histler

Executive VP
Emma Richardson

Director,
Production

Tony Barren

Regional Sales
Managers

Director, Design
& Development

Allen Roth

Manager,
Software &

Firmware Design
Cory W u

Manager,
Hardware Design

Alice Gorga

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial | 2184

HARVARD BUSINESS SCHOOL | BRIEFCASES 9

Exhibit 2 TerraCog Management Bios (in alphabetical order)

Tony Barren, Director of Production—Barren joined TerraCog as Director of Production in 2002
and spends half of his time onsite at the production facility in Shenzhen, China. Prior to TerraCog, he
was the Purchasing Manager for Markham Instruments. Barren graduated with a BS in Electrical
Engineering from Rutgers University.

Richard Fiero, President—President since 1992, Fiero began his career at TerraCog Inc. in sonar
equipment sales in 1985. He has a BS in Business Administration from Notre Dame.

Alice Gorga, Manager, Hardware Design—Gorga joined TerraCog after leaving Apple to move
closer to her family in the Midwest. Gorga graduated from the Rhode Island School of Design with a
degree in Industrial Design.

Ed Pryor, VP of Sales—Pryor started at TerraCog as a National Account Manager in 2000 and
became VP of Sales in 2006. Before joining the company, he was a category manager for camping
gear at REI. Pryor has a BA in History from UCLA.

Allen Roth, Director of Design & Development—Roth joined TerraCog in 2003 after working in
product development at Suunto, Motorola, and Research in Motion. He has a degree in Industrial
Design from Northwestern University and an MBA from USC.

Emma Richardson, Executive Vice President—Richardson was promoted to Executive VP since
January 2008. She began her TerraCog career in 1996 and has worked in both Sales and Production.
She was instrumental in shifting production from Taiwan to Shenzhen, China. Richardson has an
MBA from the University of Michigan.

Becky Timmons, CFO—Timmons was promoted to Controller in late 2007. She has been in the
Accounting department at TerraCog since 1993. Timmons is a CPA and graduated from the
University of Illinois with a degree in Accounting.

Cory Wu, Manager, Software & Firmware Design—Wu has been in software development at
TerraCog since 1998, and he was promoted to Manager in early 2007. He graduated from the
University of Wisconsin with a BS in Computer Science.

Harold Whistler, VP of Design & Development—Whistler was a co-founder of TerraCog in 1977.
His background in engineering and design enabled him to develop much of the company’s key sonar
equipment products in the 1980s. Since 1991, Whistler has served in a supervisory and advisory role
for the Design & Development team.

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

2184 | TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial

10 BRIEFCASES | HARVARD BUSINESS SCHOOL

Exhibit 3 Comparison Screen Shots

Posthaste BirdsI

TerraCog GPS

Exhibit 4 Timeline of Events

20
06

Q3 Q4

20
07

Q1 Q2 Q3 Q4

20
08

Q1 Q2 Q4Q3

BirdsI
prototype

introduced at
2006 OR Show

BirdsI holiday
2006 launch

Aerial project
launched

Design and
development

of Aerial

Aerial
completed

and passed to
Production

Planned Aerial
holiday 2008

launch

20
06

Q3 Q4

20
07

Q1 Q2 Q3 Q4

20
08

Q1 Q2 Q4Q3

BirdsI
prototype

introduced at
2006 OR Show

BirdsI holiday
2006 launch

Aerial project
launched

Design and
development

of Aerial

Aerial
completed

and passed to
Production

Planned Aerial
holiday 2008

launch

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

TerraCog Global Positioning Systems: Conflict and Communication on Project Aerial | 2184

HARVARD BUSINESS SCHOOL | BRIEFCASES 11

Exhibit 5 Seating Chart at Resumption of Pre-Launch Meeting, March 14, 2008

Harold Whistler
(Vice President,

Design & Development)

Tony Barren
(Director of
Production)

Allen Roth
(Director of Design &

Development)

Cory Wu
(Manager, Software &

Firmware Design)

Absent: Alice Gorga (Manager, Hardware Design)

Becky Timmons
(CFO)

Emma Richardson
(Executive

Vice President)

Ed Pryor
(Vice President

Sales)

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

Received: 31 March 2016 Revised: 23 May 2017 Accepted: 13 June 2017

DOI: 10.1002/jclp.22516

R E V I E W A R T I C L E

Case conceptualization research in cognitive
behavior therapy: A state of the science review

Michael H. Easden1 Nikolaos Kazantzis2

1Massey University, New Zealand

2Monash University, Australia

Correspondence

MichaelEasden,MasonClinicRegionalForensic

PsychiatryServices,PrivateBag19986,Avon-

dale,1746,Auckland.

Email:Michael.Easden@waitematadhb.govt.nz.

WethankGregoryMumma,WillemKuyken,

andCatherineSmithforfeedbackonaprevious

versionofthismanuscript.

Objective: Prominent models of cognitive behavior therapy (CBT)

assert that case conceptualization is crucial for tailoring interven-

tions to adequately address the needs of the individual client. We

aimed to review the research on case conceptualization in CBT.

Method: We conducted a systematic search of PsychINFO, MED-

LINE, Psychology and Behavioral Science Collection, and CINAHL

databases to February 2016.

Results: A total of 24 studies that met inclusion criteria were identi-

fied. It was notable that studies (a) focused on the assessment func-

tion of case conceptualization, (b) employed diverse methodologies,

and, overall, (c) there remains a paucity of studies examining the in-

session process of using case conceptualization or examining rela-

tions with outcome.

Conclusion: Results from the existing studies suggest that experi-

enced therapists can reliably construct some elements of case con-

ceptualizations, but importance for the efficacy of case conceptual-

ization in CBT has yet to be demonstrated. Research that involves

direct observation of therapist competence in case conceptualiza-

tion as a predictor of CBT outcomes is recommended as a focus for

future hypothesis testing.

K E Y W O R D S

case conceptualization, case formulation, cognitive behavior ther-

apy, cognitive therapy, review

There is widespread agreement that case conceptualization is an important and necessary foundation for the com-

petent practice of cognitive behavior therapy (CBT; J. Beck, 2011; Kuyken, Padesky, & Dudley, 2009; Persons, 1989,

2008). Case conceptualization serves as a framework for drawing together important developmental history with key

cognitions and behaviors as a basis for developing hypotheses about the etiology and maintenance of the client’s psy-

chopathology (Eells, Kendjelic, & Lucas, 1998) and treatment targets (Nezu, Nezu, & Lombardo, 2004; Persons, 1989,

2006). Comprehensive case conceptualization is not intended to capture a client’s entire life, or the “entire person,” but

to identify psychopathology through the identification of cognitive and behavioral strengths and weaknesses (Bieling

& Kuyken, 2003; Kuyken et al., 2009). The case conceptualization is also fluid and draws on client information from

each session and the experiences of the client resulting from the activities between sessions, and it could be based on

356 c© 2017 Wiley Periodicals, Inc. wileyonlinelibrary.com/journal/jclp J. Clin. Psychol. 2018;74:356–384.

EASDEN AND KAZANTZIS 357

any moment that the client interacts with the therapist (e.g., Cronin, Lawrence, Taylor, Norton, & Kazantzis, 2015). In

this way, the case conceptualization is technically never “complete”; it is an evolving hypothesis that both informs and

is informed by therapy (Kuyken et al., 2009).

Prominent models of CBT for different problems and populations emphasize the use of case conceptualization as

the primary means of ensuring that interventions and strategies are relevant and adapted to the client’s specific cir-

cumstances (e.g., Beck, Rush, Shaw, & Emery, 1979; Butler, Fennell, & Hackman, 2008; Clark & Beck, 2010; Cooper,

Fairburn, & Hawker, 2004; Epstein & Baucom, 2002; Rapee, Wignall, Hudson, & Schniering, 2000). Specific practice

guides focusing on case conceptualization in CBT have also been published (i.e., Kuyken et al., 2009, Persons, 1989,

2008). However, the empirical basis for the reliability and validity of case conceptualization in CBT remains unclear.

Previous reviews have critiqued research on the reliability and validity of case conceptualization (Flinn, Braham, & das

Nair, 2015; Persons & Hong, 2016; Rainforth & Laurenson, 2014). However, these have limitations because of a focus

on either reliability or validity and/or because they used a nonsystematic approach or examined other therapeutic

modalities. This is the first systematic review of reliability and validity of case conceptualization in the context of CBT.

1 DEFINING CASE CONCEPTUALIZATION IN CBT

1.1 Forms of conceptualization

The case conceptualization was first outlined as a means for tailoring CBT interventions to meet the specific needs

of the client (A. T. Beck et al., 1979; J. S. Beck, 1995). A variety of forms and formats of CBT conceptualization exist,

which can be focused on understanding problematic situations or iterations of a problem situation in which a problem

isreduced,doesnotrecur,orwhenaclientiseffective.Atthislevel,conceptualizationusuallycentersonidentifyingsit-

uational antecedents, along with cognitive, emotional, physiological, and behavioral experiences as their consequences

(Padesky & Mooney, 1990). This “situation”-level conceptualization has been described elsewhere as a CBT equiva-

lent of “functional analysis” (Haynes & O’Brien, 1990, 2000) and as being similar to the “A-B-C” framework in rational

emotive behavior therapy (Ellis, 1962, 1991, 1994) and “chain analysis” in dialectical behavior therapy (Linehan, 1980,

1993; Waltz & Linehan, 1999).

Another popular approach to conceptualization is “the problem list” (Persons, 1989), which prioritizes centrally

important client problems and measurable goals for intervention. However, the skilled therapist also works to identify

key developmental experiences, core beliefs, intermediate beliefs, assumptions, rules, and cross-situational behavioral

strategies (Young & Beck, 1980). “Overdeveloped” and “underdeveloped” behavioral strategies can also be included in

the conceptualization to assist the therapist in distinguishing between skill deficits and the role of beliefs in client skill

repertoires (e.g., a client may have assertiveness skills, but find them difficult to express in certain relationships).

Some comprehensive conceptualization formats require the therapist to specifically identify key thoughts, emo-

tions, behaviors, and physiology across multiple problematic situations (e.g., “Cognitive Case Conceptualization Dia-

gram” in J. Beck, 2011), whereas others incorporate attachment styles and values within attention to relationship his-

tory together with all the above (e.g., “Cognitive Case Conceptualization with a Relational Focus” in Kazantzis, Dattilio,

& Dobson, 2017). Thus, the CBT case conceptualization can provide a comprehensive understanding of both etiology

and maintenance of the client’s problems within a theoretical framework that explains the client’s psychopathology.

(The term “conceptualization” will be used primarily for present purposes, but it is understood that “formulation” is

used interchangeably in the literature, e.g., Persons, 2008).

1.2 Conceptualization as a guide for interventions

In organizing the client’s “data,” the case conceptualization also serves as a guide for the effective targeting and priori-

tization of treatment strategies (Eells, 2007; Needleman, 1999; Persons, 2006). For example, a client who had already

learned perspective taking and how to evaluate situational thoughts during a previous course of CBT for depression

358 EASDEN AND KAZANTZIS

might find it useful to focus on identifying and changing “processes in thinking,” such as through experiments that

involve attentional refocusing on feedback that does not support their negative core belief about themselves (e.g., “I

am incompetent”). A different client with depression may benefit from a historical evaluation of the same core belief.

Even though the intervention in these two examples is clearly different, both clients may experience symptom relief

and benefit in terms of the prevention of depressive relapse. Deciding between each treatment pathway will be guided

by the client’s presentation (e.g., extent of concurrent negative core beliefs about self, activated schema, competing

compensatory strategies) and attributes (e.g., cognitive capacity, including executive functions). Thus, the case concep-

tualization provides the decision-making framework for selecting interventions and then individually tailoring those

interventions to meet the needs of the client.

1.3 Relational processes in conceptualization

Wehavedefined(a)differenttypesofcaseconceptualization,(b)conveyedthatcaseconceptualizationcanbedesigned

to service initial assessment and therapy, and (c) provided a beginning account of how the effective use of con-

ceptualization depends to a large extent on therapist skill. A final consideration in this topic is the way in which

(a) and (b) are achieved within the therapeutic relationship. A comprehensive definition is important at the out-

set of this review because it presents the context for empirical study into the role of case conceptualization in

CBT.

Collaborative empiricism has long been considered central to the therapeutic relationship and therapeutic process

in CBT (A. T. Beck et al., 1979), often aligned with the process of guided discovery, the naive or curious enquiry, and

Socratic questioning. These elements of the therapeutic relationship intersect with the therapists’ case conceptualiza-

tion. That is, collaborative empiricism is the process by which the therapist and client work together toward develop-

ing and achieving shared goals drawing on each other’s unique knowledge (Dattilio & Hanna, 2012; Kazantzis, Beck,

Dattilio, Dobosn, & Rapee, 2013). Kuyken et al. (2009) provide the metaphor of the “case conceptualization crucible”

whereby CBT theory, research, and data from the client’s experience are integrated to promote change. Consequen-

tially, the different components of the CBT case conceptualization (e.g., thoughts, emotion, behavior, relevant back-

ground information) provide hypotheses for evaluation through data collection.

Thus, the case conceptualization forms the basis for a process of empiricism in which the client is a co-

investigator or self-detective, further testing and refining adaptive behavioral and cognitive skills in collaboration

with the therapist. This might be achieved, for example, either in vivo or as homework tasks; through behav-

ioral experiments, reality testing, or simply seeking client feedback; and checking the fit of the data with the

client’s experience and the evolving case conceptualization (Tee & Kazantzis, 2011). Collaborative empiricism

requires a focus beyond psychological disorder and conceptualization of problems to encompass skills, strengths,

and resilience essential to a comprehensive case conceptualization (Kazantzis, Tee, Dattilio, & Dobson, 2013). As a

result, researchers have used diverse methods to empirically capture and evaluate the reliability and validity of case

conceptualization.

In the next section, we present a systematic review of the existing evidence for case conceptualization in CBT. The

aim was to include empirical studies that provided data on the reliability and/or validity of CBT case conceptualization.

Giventheabovedefinition, thepresentreviewallowedforthepossibilitythatboththecontentandtheprocessofusing

case conceptualization could form the focus of empirical study. The specific aim was to clarify the current state of the

evidence, as well as highlight its strengths and limitations. We examined whether research has (a) correctly used tests

of reliability to determine whether therapists can, for instance, reliably construct and use case conceptualization, and

whether reliability requirements have been sufficiently satisfied to (b) determine the validity of case conceptualiza-

tion, particularly as this related to therapeutic outcomes. Given the interdependence of reliability and validity, it was

considered most useful to consider both in this review. We will then present a synthesis of the above definition with

the findings from the review and offer a research agenda.

EASDEN AND KAZANTZIS 359

Records identified through database
searching
(n = 6612)

Sc
re
en
in
g

In
cl
ud
ed

E
lig
ib
ili
ty

Id
en
ti
fi
ca
ti
on Additional records identified

through other sources
(n = 2)

Records after duplicates removed
(n = 6312)

Records screened
(n = 6312)

Records excluded
(n = 6075)

Articles assessed for
eligibility
(n = 237)

Articles excluded, with
reasons:

• Not data-based / non-
empirical / non-
experimental
(n = 131)

• Not oriented to
cognitive-behavioral
model
(n = 53)

• Single N case study
(n = 26)

• No reliability or
validity data (n = 3)

Studies included in
systematic review

(n = 24)

FIGURE 1 Flow diagram of systematic search

2 METHOD

2.1 Search strategy

Studies examining case conceptualization were identified by (a) searching the PsychINFO, MEDLINE, Psychology

and Behavioral Science Collection, and CINAHL databases to February 2016 and using the text phrases “case con-

ceptualization,” “case conceptualisation,” or “case formulation,” and (b) manually searching the reference sections

of articles identified and included (see Figure 1). The present research adopted a definition of “Cognitive Behav-

ior Therapy” or “CBT” as reflecting either behaviorally focused or cognitively focused therapies (see discussions

in Kazantzis, Freeman, & Reinecke, 2010; Mennin, Ellard, Fresco, & Gross, 2013). Studies examining “formulation-

driven” case conceptualization were included, while the present study did not seek to provide a review on manual-

ized versus individualized treatment protocols and included studies with a focus on case conceptualization. Studies

were included in the review if they were (a) published in English, (b) had more than one participant (i.e., case stud-

ies were excluded unless multiple raters were used to assess the individual case), and (c) specifically targeted some

aspect of CBT conceptualization and therapeutic intervention (i.e., studies coming from non-CBT orientations were

excluded).

360 EASDEN AND KAZANTZIS

2.2 Data extraction and synthesis

The diversity of methodologies and reporting parameters identified in selected studies meant that application of sta-

tistical pooling was limited. The results are therefore presented in line with the Preferred Reporting Items for System-

atic Reviews and Meta-Analyses (PRISMA) and incorporate elements of narrative synthesis. Specifically, the findings

are charted and summarized where possible by those outlined by the Cochrane Handbook for Systematic Reviews of

Interventions (Higgins & Green, 2011). Of the studies, 50% were double coded by an independent, doctoral-level clini-

calpsychologistandyieldedhighreliabilityinselectionandextraction(i.e.,98.1%agreement).Anycodingdiscrepancies

were resolved by discussion. We offer an overview of data able to be synthesized in a systematic manner, followed by

limited synthesis of data and findings in each subsection, and accompanied by a narrative description of select studies

sufficient to facilitate an understanding of the frequent idiosyncrasies of each study even within study groupings.

3 RESULTS

3.1 Overview of data

A total of 24 studies were identified for the review. Table 1 provides a summary of key reliability and validity data

extracted from each study. Table 2 provides additional data on study aims, model integrity, and outcome assessment.

3.1.1 Reliability

Over half of all studies (n = 14, 58.3%) aimed to evaluate interclinician reliability in constructing case conceptualiza-
tions with similar case material. An expert benchmark or criterion conceptualization was frequently adopted (n = 15,
62.5%), primarily to investigate reliability; but in two studies, the expert conceptualization was used to gauge validity.

One study (4.2%) reported on test-retest reliability of interpersonal scenarios (i.e., “mini” formulations).

3.1.2 Validity

Studies also sought (directly or indirectly) to evaluate case conceptualization in relation to symptom change (n = 16,
66.7%). Seven of these studies (29.2%) provided some measure of effect size (i.e., r, R2 or Cohen’s d) concerning the

relationship between aspects of case conceptualizations and patient symptomology, but three of these were small-N

repeated measures (i.e., close analysis and serial observations of a small number of subjects) and each had disparate

methods and/or level of focus. Four studies (17%) sought client or therapist feedback on the perceived clinical utility of

case conceptualizations. Where CBT was delivered, only two studies (8.3%) reported data relevant to integrity checks

to ensure fidelity to the CBT model and/or the treatment condition.

3.1.3 Training

A total of 11 studies (45.8%) carried out their aims in the context of conceptualization training or in the provision of

prescriptive training as part of the research design. Five studies (20.8%) reported some measure of effect size for sta-

tistical associations between training / experience and hypothesized quality of written case conceptualization and/or

ability to identify themes or content based on vignettes.

3.2 Participant characteristics

Table 1 includes a breakdown of raters, therapist and clients/vignettes. A total of 501 clients/vignettes formed the

bases of the total pooled sample. A total pool of 542 therapists and raters were used across studies identified for

review. However, the therapists often served as raters and vise-versa, as well as different numbers of therapists and

raters being used for different analyses. As a result, it is less meaningful to provide the number of therapist versus

raters (i.e., 363 raters and 179 therapists took part across studies).

EASDEN AND KAZANTZIS 361

T
A
B
L
E
1

S
u
m
m
a
ry

o
f
ke
y
fi
n
d
in
g
s
fo
r
re
li
a
b
il
it
y
a
n
d
v
a
li
d
it
y
in
se
le
ct
e
d
st
u
d
ie
s
(N

=
2
4
)

S
a
m
p
le
si
ze

(n
)

K
e
y
fi
n
d
in
g
s

R
a
te
rs

T
h
e
ra
p
is
ts

C
li
e
n
ts
/

v
ig
n
e
tt
e
s

R
e
li
a
b
il
it
y

V
a
li
d
it
y

1
.
C
h
a
d
w
ic
k
e
t
a
l.

(2
0
0
3
)U

K
1

2
1
3

•
N
o
n
si
g
n
ifi
ca
n
t
re
su
lt
s
fo
r
im

p
a
ct
o
f
ca
se

fo
rm

u
la
ti
o
n
o
n

a
n
x
ie
ty

o
r
d
e
p
re
ss
io
n
u
si
n
g
F
ri
e
d
m
a
n
’s
A
N
O
V
A

(D
e
p
re
ss
io
n
:C

h
iS
q
u
a
re
d
=
6
.7
8
,d
f
=
3
,n
s:
A
n
x
ie
ty
:C

h
i

sq
u
a
re
d
=
3
.3
2
,d
f
=
3
,n
s)

•
S
ig
n
ifi
ca
n
t
in
cr
e
a
se

in
cl
ie
n
t
se
lf
-r
e
p
o
rt
ra
ti
n
g
s
o
f

th
e
ra
p
e
u
ti
c
a
ll
ia
n
ce

(H
A
q
)m

a
d
e
a
ft
e
r
fo
cu
se
d
u
se

o
f

ca
se

fo
rm

u
la
ti
o
n
in
-v
iv
o
b
e
tw

e
e
n
T
1
a
n
d
T
3
(T

=
−
2
.1
2
,

p
<
0
.0
5
)a
n
d
T
1
a
n
d
T
4
(T

=
−
2
.2
5
,p

<
0
.0
5
)b
u
t
n
s

fi
n
d
in
g
s
fo
r
o
th
e
r
ti
m
e
s
(i
.e
.,
T
1
/T
2
a
n
d
T
3
/T
4
w
e
re

n
s)

•
Q
u
a
li
ta
ti
v
e
in
te
rv
ie
w
s
w
it
h
cl
ie
n
ts
re
p
o
rt
in
g
sh
a
ri
n
g
o
f

ca
se

fo
rm

u
la
ti
o
n
e
n
h
a
n
ce
d
u
n
d
e
rs
ta
n
d
in
g
o
f
o
w
n

p
ro
b
le
m
s
(n

=
9
),
p
o
si
ti
v
e
e
m
o
ti
o
n
s
(n

=
6
),

u
n
d
e
rs
ta
n
d
in
g
o
f
th
e
ir
th
e
ra
p
is
t
(n

=
3
),
n
e
g
a
ti
v
e

e
m
o
ti
o
n
s
(n

=
6
),
n
o
e
m
o
ti
o
n
a
li
m
p
a
ct
(n

=
3
)o
r
fi
n
d
in
g

ca
se

fo
rm

u
la
ti
o
n
s
co
m
p
li
ca
te
d
(n

=
2
)

2
.
D
u
d
le
y
e
t
a
l.

(2
0
1
0
)U

K
8
2
/
3

n
/a

1
•

P
e
rc
e
n
ta
g
e
a
g
re
e
m
e
n
t
o
n
ex
p
e
rt
b
e
n
ch
m
a
rk

ra
n
g
e
d

fr
o
m
3
2
.2
%
to

9
1
.6
%

•
H
ig
h
e
r
a
g
re
e
m
e
n
t
fo
r
d
e
sc
ri
p
ti
v
e
/s
u
rf
a
ce

le
v
e
la
sp
e
ct
s

a
n
d
lo
w
e
r
a
g
re
e
m
e
n
t
fo
r
m
o
re

in
fe
re
n
ti
a
l/
d
e
e
p
e
r
le
v
e
l

co
n
te
n
t
o
f
ca
se

fo
rm

u
la
ti
o
n

•
T
h
e
“n
u
m
b
e
r
o
f
C
B
T
ca
se
s
se
e
n
”
w
a
s
p
o
si
ti
v
e
ly

a
ss
o
ci
a
te
d
(𝛽

=
0
.2
8
,t
=
2
.1
7
,p

<
.0
5
)w

it
h
th
e
m
e
a
n

n
u
m
b
e
r
o
f
in
fe
re
n
ti
a
lc
o
m
p
o
n
e
n
ts
id
e
n
ti
fi
e
d
.S
im

il
a
rl
y

“o
v
e
ra
ll
y
e
a
rs
o
f
ex
p
e
ri
e
n
ce
”
(𝛽

=
0
.4
2
,t
=
2
.8
5
,p

<
.0
1
)

h
a
d
a
p
o
si
ti
v
e
a
ss
o
ci
a
ti
o
n
w
it
h
th
e
to
ta
lf
o
rm

u
la
ti
o
n

m
e
a
n
sc
o
re
,a
lt
h
o
u
g
h
th
e
“n
u
m
b
e
r
o
f
p
sy
ch
o
si
s
ca
se
s

se
e
n
”
h
a
d
a
n
in
v
e
rs
e
re
la
ti
o
n
sh
ip
(𝛽

=
−
0
.3
7
,t
=
−
2
.3
8
,p

<
.0
5
)

3
.
E
e
ll
s
e
t
a
l.
(2
0
0
5
)

U
S
A

6
6
5

6
•

K
a
p
p
a
co
e
ffi
ci
e
n
ts
b
a
se
d
o
n
‘d
e
sc
ri
p
ti
v
e
’(
𝜅
=
0
.6
1
),

‘d
ia
g
n
o
st
ic
’(
𝜅
=
0
.8
1
),
‘in
fe
re
n
ti
a
l’
(𝜅

=
0
.6
2
)a
n
d

‘t
re
a
tm

e
n
t’
(𝜅

=
0
.6
9
)

•
R
e
li
a
b
il
it
y
(u
si
n
g
tw

o
-w

ay
ra
n
d
o
m
e
ff
e
ct
s
IC
C
)f
o
r

co
m
p
o
n
e
n
ts
/
ca
te
g
o
ri
e
s
o
f
co
n
te
n
t
ra
n
g
e
d
fr
o
m
.6
9
to

.8
9

•
E
x
p
e
rt
co
n
ce
p
tu
a
li
za
ti
o
n
s
w
e
re

ra
te
d
a
s
m
o
re

co
m
p
re
h
e
n
si
v
e
,e
la
b
o
ra
te
d
,c
o
m
p
le
x
,a
n
d
sy
st
e
m
a
ti
c

th
a
n
le
ss
ex
p
e
ri
e
n
ce
d
th
e
ra
p
is
ts

•
G
re
a
te
r
d
e
ta
il
a
n
d
re
le
v
a
n
ce

w
a
s
o
b
se
rv
e
d
in
th
e

ex
p
e
ri
e
n
ce
d
th
e
ra
p
is
ts
’c
o
n
ce
p
tu
a
li
za
ti
o
n
s

(C
on
ti
nu
es
)

362 EASDEN AND KAZANTZIS

T
A
B
L
E
1

(C
o
n
ti
n
u
e
d
)

S
a
m
p
le
si
ze

(n
)

K
e
y
fi
n
d
in
g
s

R
a
te
rs

T
h
e
ra
p
is
ts

C
li
e
n
ts
/

v
ig
n
e
tt
e
s

R
e
li
a
b
il
it
y

V
a
li
d
it
y

4
.
E
e
ll
s
e
t
a
l.
(2
0
1
1
)

U
S
A

6
6
5

6
•

T
h
e
th
e
ra
p
is
ts
,v
ig
n
e
tt
e
s,
tr
a
n
sc
ri
p
ti
o
n
a
n
d
co
n
te
n
t

co
d
in
g
p
ro
ce
d
u
re
s
a
re

th
e
sa
m
e
a
s
d
e
sc
ri
b
e
d
in
E
el
ls

et
al
.(
2
0
0
5
)a
lt
h
o
u
g
h
w
it
h
d
if
fe
re
n
t
re
se
a
rc
h
q
u
e
st
io
n
s

a
n
d
a
n
a
ly
si
s.
A
s
su
ch

re
li
a
b
il
it
y
d
a
ta

is
d
u
p
li
ca
te
d
.

•
R
e
su
lt
s
y
ie
ld
e
d
sm

a
ll
to

m
o
d
e
ra
te

e
ff
e
ct
si
ze
s
fo
r

ex
p
e
rt
s
g
e
n
e
ra
ti
n
g
m
o
re

d
e
sc
ri
p
ti
v
e
(d

=
.3
3
),
d
ia
g
n
o
st
ic

(d
=
.2
4
),
in
fe
re
n
ti
a
l(
d
=
.3
5
),
a
n
d
tr
e
a
tm

e
n
t
p
la
n
n
in
g

in
fo
rm

a
ti
o
n
(d

=
.3
2
)i
n
th
e
ir
ca
se

fo
rm

u
la
ti
o
n
s
th
a
n
th
e

n
o
n
ex
p
e
rt
s

•
E
x
p
e
rt
s
h
a
d
a
g
re
a
te
r
fo
cu
s
o
n
tr
e
a
tm

e
n
t
(d

=
.3
4
)a
n
d

p
sy
ch
o
lo
g
ic
a
lm

e
ch
a
n
is
m
s
o
f
ch
a
n
g
e
(d

=
.2
5
)t
h
a
n

n
o
n
ex
p
e
rt
s.

5
.
E
sb
jø
rn

e
t
a
l.

(2
0
1
5
)D

e
n
m
a
rk

0
8

5
4

•
M
cN

e
m
a
r’
s
te
st
re
v
e
a
le
d
n
o
n
si
g
n
ifi
ca
n
t
re
su
lt
s
fo
r

d
if
fe
re
n
ce
s
o
n
se
lf
-r
a
te
d
a
n
x
ie
ty

le
v
e
ls
a
n
d
d
ia
g
n
o
st
ic

st
a
tu
s
(d
e
te
rm

in
e
d
u
si
n
g
th
e
A
D
IS
-I
V
-C
/P
)b
e
tw

e
e
n
th
e

fo
rm

u
la
ti
o
n
-d
ri
v
e
n
a
n
d
co
n
tr
o
lg
ro
u
p
d
e
sp
it
e
a
tr
e
n
d

to
w
a
rd
s
h
ig
h
e
r
p
e
rc
e
n
ta
g
e
sy
m
p
to
m
re
m
is
si
o
n
th
e
in

fo
rm

u
la
ti
o
n
-d
ri
v
e
n
g
ro
u
p

6
.
F
li
tc
ro
ft
e
t
a
l.

(2
0
0
7
)U

K
0

2
3

1
•

Q
-s
o
rt
m
e
th
o
d
o
lo
g
y
to

d
e
te
rm

in
e
im

p
o
rt
a
n
t
fe
a
tu
re
s
o
f

ca
se

fo
rm

u
la
ti
o
n
th
e
n
a
p
p
li
e
d
to

fa
ct
o
r
a
n
a
ly
si
s

•
T
h
re
e
fa
ct
o
r
so
lu
ti
o
n
w
it
h
e
ig
e
n
v
a
lu
e
s
a
b
o
v
e
1
.0
0
,a
n
d

a
cc
o
u
n
ti
n
g
fo
r
4
9
%
o
f
th
e
v
a
ri
a
n
ce

in
to
ta
l(
F
a
ct
o
r
A

“C
B
T
st
a
te
”
=
2
4
%
;F
a
ct
o
r
B
“h
e
re

a
n
d
n
o
w
”
=
1
8
%
;

F
a
ct
o
r
C
“p
ro
ce
ss
a
n
d
fu
n
ct
io
n
”
=
7
%
)

7
.
G
h
a
d
e
ri
(2
0
0
6
)

S
w
e
d
e
n

2
1

5
0

•
O
f
re
le
v
a
n
ce
,i
n
th
e
co
u
rs
e
o
f
a
d
h
e
re
n
ce

ch
e
ck
s,

d
e
ta
il
e
d
su
m
m
a
ri
e
s
o
f
2
0
ra
n
d
o
m
ly
ch
o
se
n
ca
se
s
(h
a
lf
in

th
e
in
d
iv
id
u
a
li
ze
d
co
n
d
it
io
n
,h
a
lf
fr
o
m
st
a
n
d
a
rd

C
B
T
)

w
e
re

re
v
ie
w
e
d
b
y
th
re
e
in
d
e
p
e
n
d
e
n
t
th
e
ra
p
is
ts
w
h
o

ju
d
g
e
d
w
h
a
t
co
n
d
it
io
n
th
e
p
a
ti
e
n
t
b
e
lo
n
g
e
d
to
,a
n
d
to

a
ss
e
ss
th
e
q
u
a
li
ty

o
f
th
e
tr
e
a
tm

e
n
t.
In
to
ta
l,
th
e
ra
ti
n
g
s

o
f
th
e
co
n
d
it
io
n
s
w
e
re

co
rr
e
ct
fo
r
9
0
%
(t
w
o
p
a
ti
e
n
ts

w
e
re

m
is
id
e
n
ti
fi
e
d
to

b
e
in
st
a
n
d
a
rd
iz
e
d
co
n
d
it
io
n
).

•
M
o
d
e
ra
te

to
la
rg
e
e
ff
e
ct
si
ze
s
(r
e
p
o
rt
e
d
a
s
p
a
rt
ia
lE
ta

sq
u
a
re
d
)f
o
r
g
ro
u
p
s
o
n
so
m
e
E
D
E
su
b
sc
a
le
s
a
n
d

m
e
a
su
re
s
fo
r
‘n
u
m
b
e
r
o
f
w
e
e
k
s
o
f
a
b
st
in
e
n
ce

fr
o
m

o
b
je
ct
iv
e
b
u
li
m
ic
e
p
is
o
d
e
s
(O

B
E
s)
’(
.1
5
),
‘e
a
ti
n
g

co
n
ce
rn
s’
(.
1
1
),
a
n
d
‘b
o
d
y
sh
a
p
e
d
is
sa
ti
sf
a
ct
io
n
(u
si
n
g

B
S
Q
)’
(.
0
8
)i
n
fa
v
o
u
r
o
f
in
d
iv
id
u
a
li
ze
d
ca
se

fo
rm

u
la
ti
o
n
-d
ri
v
e
n
co
n
d
it
io
n
.A

si
g
n
ifi
ca
n
t
g
ro
u
p

d
if
fe
re
n
ce

w
a
s
o
b
se
rv
e
d
fo
r
‘e
xc
e
ss
iv
e
co
m
p
e
n
sa
to
ry

ex
e
rc
is
in
g
’(
.1
5
)a
n
d
‘c
o
m
p
e
n
sa
to
ry

b
e
h
av
io
u
r’
(.
1
0
),
b
u
t

n
o
in
te
ra
ct
io
n

•
T
h
e
re

w
e
re

n
s
g
ro
u
p
o
r
in
te
ra
ct
io
n
e
ff
e
ct
s
fo
r
o
th
e
r

p
re
-p
o
st
m
e
a
su
re
s
in
cl
u
d
in
g
th
e
B
D
I-
II
,R
o
se
n
b
e
rg

S
e
lf
-e
st
e
e
m
sc
a
le
,a
n
d
E
D
I

(C
on
ti
nu
es
)

EASDEN AND KAZANTZIS 363

T
A
B
L
E
1

(C
o
n
ti
n
u
e
d
)

S
a
m
p
le
si
ze

(n
)

K
e
y
fi
n
d
in
g
s

R
a
te
rs

T
h
e
ra
p
is
ts

C
li
e
n
ts
/

v
ig
n
e
tt
e
s

R
e
li
a
b
il
it
y

V
a
li
d
it
y

8
.
H
a
a
rh
o
ff
e
t
a
l.

(2
0
1
1
)N

e
w

Z
e
a
la
n
d

2
6

n
/a

4
•

P
e
rc
e
n
ta
g
e
a
g
re
e
m
e
n
t
d
a
ta

b
e
tw

e
e
n
ra
te
rs
a
n
d

b
e
n
ch
m
a
rk

w
a
s
n
o
t
p
ro
v
id
e
d
fo
r
a
ll
in
d
ic
e
s.
A
u
th
o
rs

n
o
te

th
e
h
ig
h
e
st
a
g
re
e
m
e
n
t
fo
r
‘r
e
le
v
a
n
t
ch
il
d
h
o
o
d
d
a
ta
’,

‘A
x
is
I
d
ia
g
n
o
si
s’
,‘
co
m
p
e
n
sa
to
ry

st
ra
te
g
ie
s’
,a
n
d
‘c
o
re

b
e
li
e
fs
a
b
o
u
t
se
lf
’,
a
n
d
b
ro
a
d
o
m
is
si
o
n
s
le
a
d
in
g
to

v
e
ry

p
o
o
r
a
g
re
e
m
e
n
t
(i
.e
.,
“a
lm

o
st
n
o
a
g
re
e
m
e
n
t”
)o
n
‘t
h
e
ra
p
y

in
te
rf
e
ri
n
g
b
e
h
av
io
r’
a
n
d
‘A
x
is
II
p
e
rs
o
n
a
li
ty

d
is
o
rd
e
r

tr
a
it
s’
.K

a
p
p
a
a
g
re
e
m
e
n
t
b
e
tw

e
e
n
cl
in
ic
ia
n
a
n
d

b
e
n
ch
m
a
rk

fo
r
‘p
ro
b
le
m
li
st
’w

a
s
lo
w
fo
r
a
ll
v
ig
n
e
tt
e
s

(𝜅
s<

3
0
%
)

9
.
K
e
n
d
je
li
c
&
E
e
ll
s

(2
0
0
7
)U

S
A

3
4
3

9
9

•
To
ta
lm

e
a
n
k
a
p
p
a
fo
r
co
d
e
d
fo
rm

u
la
ti
o
n
s
a
cr
o
ss
a
ll

v
a
ri
a
b
le
s
a
n
d
co
d
e
rs
w
a
s
.8
6
(r
a
n
g
e
.6
0
to

1
.0
)

•
B
a
se
d
o
n
a
la
rg
e
m
e
a
n
e
ff
e
ct
si
ze

(d
=
1
.1
2
)t
h
e

re
se
a
rc
h
e
rs
co
n
cl
u
d
e
d
th
a
t
th
e
av
e
ra
g
e
th
e
ra
p
is
t
in
th
e

tr
a
in
in
g
g
ro
u
p
h
a
d
a
b
e
tt
e
r
q
u
a
li
ty

co
n
ce
p
tu
a
li
za
ti
o
n

th
a
n
8
6
%
o
f
th
e
ra
p
is
ts
in
th
e
co
n
tr
o
lg
ro
u
p

1
0
.
K
u
y
ke
n
e
t
a
l.

(2
0
0
5
)U

K
1
1
5
/
2

n
/a

1
•

P
e
rc
e
n
t
a
g
re
e
m
e
n
t
b
e
tw

e
e
n
1
1
5
in
d
iv
id
u
a
lc
li
n
ic
ia
n
s

a
n
d
ex
p
e
rt
b
e
n
ch
m
a
rk

co
n
ce
p
tu
a
li
za
ti
o
n
ra
n
g
e
d
fr
o
m

7
%
to

7
3
%

•
T
w
o
in
d
e
p
e
n
d
e
n
t
ju
d
g
e
s
ra
te
d
a
g
re
e
m
e
n
t
o
n
ca
te
g
o
ri
e
s

o
f
in
fo
rm

a
ti
o
n
u
n
it
s
(r
a
n
g
e
𝜅
=
0
.6
3
to

0
.9
1
)

1
1
.
M
o
rb
e
rg
-P
a
in

e
t
a
l.
(2
0
0
8
)U

K
2

5
1
3

•
K
a
p
p
a
a
g
re
e
m
e
n
t
fo
r
id
e
n
ti
fi
e
d
th
e
m
e
s
w
a
s
h
ig
h
(𝜅

=
.8
9
)

•
P
a
ti
e
n
ts
e
n
d
o
rs
e
d
a
co
m
b
in
a
ti
o
n
o
f
p
o
si
ti
v
e
(n

=
9
,6
9
%
),

n
e
g
a
ti
v
e
(n

=
9
,6
9
%
)a
n
d
n
e
u
tr
a
l(
n
=
6
,4
6
%
)e
m
o
ti
o
n
a
l

re
a
ct
io
n
s
to

u
se

o
f
ca
se

fo
rm

u
la
ti
o
n
s

(C
on
ti
nu
es
)

364 EASDEN AND KAZANTZIS

T
A
B
L
E
1

(C
o
n
ti
n
u
e
d
)

S
a
m
p
le
si
ze

(n
)

K
e
y
fi
n
d
in
g
s

R
a
te
rs

T
h
e
ra
p
is
ts

C
li
e
n
ts
/

v
ig
n
e
tt
e
s

R
e
li
a
b
il
it
y

V
a
li
d
it
y

1
2
.
M
u
m
m
a
(2
0
0
4
)

U
S
A

1
1

1
•

Id
io
g
ra
p
h
ic
C
o
g
n
it
iv
e
S
ch
e
m
a
(I
C
S
)[
g
e
n
e
ra
te
d
u
si
n
g
th
e

C
B
IS
S
A
I]
fa
ct
o
rs
ex
p
la
in
e
d
a
n
av
e
ra
g
e
o
f
2
9
%
o
f
th
e

d
a
il
y
v
a
ri
a
b
il
it
y
in
sy
m
p
to
m
/d
is
tr
e
ss
ra
ti
n
g
s
(R

2
=
.1
9
to

.4
3
)a
s
m
e
a
su
re
d
o
n
a
n
ID
Q
.T
h
is
w
a
s
co
m
p
a
ra
b
le

b
e
tw

e
e
n
av
e
ra
g
e
sh
ru
n
ke
n
R
2
=
.2
4
fo
r
th
e
IC
S

p
re
d
ic
to
rs
a
n
d
R
2
=
.2
5
fo
r
th
e
n
o
m
o
th
e
ti
c
co
g
n
it
io
n

p
re
d
ic
to
rs
.T
e
m
p
o
ra
l/
in
cr
e
m
e
n
ta
lv
a
li
d
it
y
:I
C
S
fa
ct
o
r

sc
o
re
s
p
re
d
ic
te
d
a
n
av
e
ra
g
e
o
f
6
%
o
f
th
e
d
a
il
y
v
a
ri
a
b
il
it
y

in
sy
m
p
to
m
/d
is
tr
e
ss
o
v
e
r
a
n
d
a
b
o
v
e
th
a
t

________________________________________________________________________________________________________________

HBS Professor Michael Beer and writer Rachel Shelton prepared this case solely as a basis for class discussion and not as an endorsement, a
source of primary data, or an illustration of effective or ineffective management. This case, though based on real events, is fictionalized, and any
resemblance to actual persons or entities is coincidental. There are occasional references to actual companies in the narration.

Copyright © 2012 Harvard Business School Publishing. To order copies or request permission to reproduce materials, call 1-800-545-7685, write
Harvard Business Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored
in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means——electronic, mechanical, photocopying, recording, or
otherwise——without the permission of Harvard Business Publishing.

Harvard Business Publishing is an affiliate of Harvard Business School.

M I C H A E L B E E R

R A C H E L S H E L T O N

BoldFlash: Cross-Functional Challenges in the
Mobile Division

On January 16, 2012, Dr. Roger Cahill walked into his office in BoldFlash’s Waltham,
Massachusetts headquarters at 7 a.m., less than a year into his new role as Vice President of the
company’s Mobile Division. His predecessor’s personal photos and mementos had been packed up
months ago, but they still sat in a corner, patiently waiting for someone to collect them. Cahill didn’t
have the heart to move them because Jim Harrison had died unexpectedly after more than a decade
in the position. Jack Young, the company’s CEO, immediately replaced Harrison with Cahill, a 24-
year employee and highly respected research scientist who previously had led Research in the
Consumer Division.

Cahill had faced a big challenge since assuming his new role. BoldFlash, a producer of flash
memory components for electronic devices, had fallen behind its competitors on several fronts. It was
experiencing pressure on both prices and the ability to get new products to market quickly. Shortly
before Harrison’s death, the Mobile team had developed redundant customized chips for mobile
phone market customers while completely missing a critical market in storage devices used for
tablets, thus handing its competitors an advantage in a critical growth market. A brilliant
entrepreneur whose vision drove the growth of the Mobile Division, Harrison was perceived as an
autocrat who fostered a compliant culture in which people protected themselves and their territory
from his strong top-down directives. Cahill wondered if his predecessor’s leadership style
contributed to the conflicts he was observing. More generally, he had been unsure what to expect
from the groups in the division——Product Development, Marketing and Sales, and Manufacturing——
and he did not want to judge based on anything other than personal experience and observation.

After a few months on the job, Cahill felt that he had sized up the situation. There was a lot of
unproductive conflict between the functional departments, particularly around the product
development process. Manufacturing was the dominant function and seemed to be pulling its
weight, but overall he saw a division that was underperforming and struggled with communication
and teamwork. The product development process seemed dysfunctional. The division’s quarterly
product development meeting was also dysfunctional. ‘‘The product development meeting should be
the primary opportunity to plan the creation of new products. So I was surprised when I found out
that Jim Harrison usually didn’t attend and instead left leadership of the meeting up to Marketing,’’
he observed. Cahill had broken with tradition and attended the most recent meeting two months ago.

4 4 3 8
M A Y 3 1 , 2 0 1 2

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

4438 | BoldFlash: Cross-Functional Challenges in the Mobile Division

2 BRIEFCASES | HARVARD BUSINESS PUBLISHING

‘‘I did not see the strong leadership that I would expect from a team with this much talent,’’ he
observed. ‘‘The meeting lasted two days, yet I had a hard time summarizing what we had
accomplished.’’

With the next two-day product development meeting coming up in less than a month, Cahill
wondered what changes he would have to make to the meeting, the organization, and the product
development process to develop truly innovative, market-leading products that anticipated customer
needs.

Company History and the Flash Memory Market

Two computer science professors founded BoldDisk in Waltham, Massachusetts in 1982. Initially,
it manufactured computer storage media, primarily floppy disks, at a plant in Waltham. The
company continued to focus on storage devices as technology evolved (with the exception of an ill-
fated foray into MP3 players in the early 1990s), although its product mix evolved with the market.
By 2012, the company——now re-named BoldFlash——focused on flash memory and had a significant
customer base in both OEM (original equipment manufacturer) and direct-to-consumer markets.

Flash memory is a solid-state drive (SSD) that maintains data storage without moving parts, even
when disconnected from power. It is commonly used in consumer devices such as digital cameras,
mobile phones, USB drives, MP3 players, DVD players, GPS systems, tablet computers, and notebook
computers. Although some flash memory is embedded into the product, consumers can also
purchase flash memory devices to increase the device’s storage capacity. Industry production is
spread among a number of manufacturers that run the gamut from small specialty manufacturers to
global technology powerhouses. Purchasing power is concentrated, with one customer alone
consuming one-third of the market for the newer, high-density NAND-type flash memory. In 2011,
smartphones accounted for 40% of the overall market.

Industry revenues, $20 billion in 2010, were expected to increase to $44 billion by 2014. The sector
was highly competitive. Although demand was consistently strong and expected to grow 22% in
2012, margins were constantly squeezed due to pricing pressure (prices were expected to decline 30%
in 2012) and manufacturers sought to reduce costs wherever possible. This environment led
BoldFlash to move most of its manufacturing outside the U.S. By 2012 the Mobile Division’s largest
plant was in Shanghai; its remaining U.S. plant was in Austin, Texas, and it also had a smaller plant
in Hamilton, Ontario, Canada. Unlike some of its competitors, BoldFlash operated its own plants; it
did not subcontract or have any joint venture agreements with competitors. The industry was
sometimes prone to supply shortages, so fast, flexible manufacturing practices were critical in
meeting customer demand.

In 2011, overall revenues at BoldFlash were $3.9B, with approximately half of sales to OEMs and
half directly to consumers through warehouses, big box retailers, and electronics stores. Mobile
Division revenues were $1.5B (see Exhibits 1A and 1B for detailed financial statements); smartphone
manufacturers were its largest customer group. The company focused solely on organic growth and
had never made an acquisition.

Competitors and customers respected BoldFlash’s reputation for quality. Consistently headed by a
PhD-level scientist, the company maintained a strong commitment to ongoing research. It placed a
premium on intellectual property, and employees receiving patents were richly rewarded and
recognized. The company held more than 600 patents in the U.S. alone; however, it had not always
successfully commercialized the technology it developed, and its developments did not always align
with market needs. Analysts sometimes criticized the firm as doing research for its own sake rather

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

BoldFlash: Cross-Functional Challenges in the Mobile Division | 4438

HARVARD BUSINESS PUBLISHING | BRIEFCASES 3

than to improve products or meet new customer needs and faulted a corporate culture that was not
always focused on results.

Because of rapid changes in customer preferences, the Mobile Division had to identify customer
needs and respond more swiftly than in the past with new products that adapted its technology to
the needs of Marketing and Sales. The number of new products developed by Mobile in the past few
years had slowed considerably, however, accounting for its lack of growth. Because of cost
pressures, new products had to be designed with cost of manufacturing in mind, not just technical
elegance.

Leadership of BoldFlash Mobile

Roger Cahill had spent his career working in R&D at Waltham headquarters. He joined in 1984 as
a co-op student while he completed his degree in Electrical Engineering, and although he left for four
years in the early 1990s to finish his doctorate in the same field, he rejoined the week after
graduation. He himself held fifteen patents, the most of anyone in the company at his level.

Cahill reported to CEO Jack Young (see Exhibits 2A and 2B for organization charts), and everyone
on the Mobile Division leadership team was a long-term BoldFlash employee. Except for the Vice
President and plant-based manufacturing employees in China and Canada, most of the rest of
Mobile’s 650 employees had been based in Austin since 1995, when corporate leadership spun the
Mobile Division out of the OEM group and moved the plant from Waltham. The central Texas
location offered lower cost of manufacturing and benefited from proximity to the other high-tech
firms in the area.

Because of the tablet miss, Cahill was under pressure from Young to get Mobile back on track
quickly. Cahill said, ‘‘I’m not a naturally quick decision-maker. I’m a scientist. I like to analyze all the
options and make sure I’m doing the right thing. But in this case, Jack told me in no uncertain terms
that ‘time’ was something I didn’t have. So I moved quickly,’’ said Cahill. ‘‘I’m still not sure if what I
did was right, but I had to do something.’’ Within his first six months in Mobile, he had made several
big changes.

Two months into his tenure, Cahill had separated Marketing and Sales, which had previously
operated together under Director Chip Bryant. ‘‘The two are essentially different functions and were
separate units in the Consumer Division. Marketing is a strategic function while Sales’ role is to
knock on doors and sell,’’ Cahill said. As part of this re-organization, Cahill appointed Kavita Patel, a
colleague from the Consumer division, as Director of Marketing and retained Bryant as head of Sales.

Another high-profile move was consolidating the division’s non-manufacturing employees by
relocating all Austin-based corporate staff and the Shanghai-based Director of Manufacturing to
Waltham headquarters. ‘‘I needed all my leaders in the same time zone,’’ said Cahill. ‘‘I didn’t like
isolating manufacturing leadership halfway around the world. I’d even move the plants to Waltham
if that wouldn’t kill my margins.’’ He also decided to hold Manufacturing, not Sales, accountable for
meeting delivery dates.

These changes were not without controversy. He lost some key staff members, including the
incumbent Manufacturing Director, who refused to relocate from his native China; Cahill named
Kevin Cheng, previously the Shanghai plant manager, to this position.

Cahill’s reports perceived clear differences between their new boss’s leadership style and that of
predecessor Jim Harrison. ‘‘Jim could be kind of overpowering, and people became more protective
and political because they felt intimidated and unable to disagree. But you always knew where he

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

4438 | BoldFlash: Cross-Functional Challenges in the Mobile Division

4 BRIEFCASES | HARVARD BUSINESS PUBLISHING

stood.’’ said Karl Melzer, Director of Product Development. ’’Roger, on the other hand, is a great
guy. He is much more open and generates trust and involvement, but I think he’s a scientist at heart,
not yet a general manager. I’m not sure why he’s making all these changes.’’ Cahill’s vast technical
knowledge sometimes overwhelmed his colleagues. Kavita Pavel said, ‘‘When Roger gets rolling,
watch out. I had a meeting with him to review our 2012 forecasts and he spent the entire hour talking
about a new design technique that would increase NAND capacity. His expertise is incredible, but it
didn’t help me with my forecasts.’’

Mobile Division Units

Manufacturing

Each of the division’s three manufacturing plants operated as its own profit center and was
evaluated on gross margin and manufacturing-specific metrics such as quality, on-time delivery, and
inventory management. New Manufacturing Director Kevin Cheng, promoted from Shanghai plant
manager, was an American who joined BoldFlash in 1995 as a quality control engineer after receiving
a dual graduate degree in business and engineering. ‘‘Like my old boss, I didn’t want to move from
China and thought about leaving the company,’’ he said, ‘‘but I like working in an environment that
challenges me intellectually. I don’t always agree with my colleagues, but for the most part I respect
their intelligence——especially Roger’s.’’

The managers of the plants in Austin and Ontario were Americans and BoldFlash lifers; the new
Shanghai plant manager was a native Chinese who had attended a top engineering school in the U.S.
The three managers were a competitive group and often compared their results during their monthly
conference calls. Managers at the plant with the lowest margins were subject to sometimes good-
natured ridicule from their peers, but results were serious business. ‘‘I want to encourage a high-
performing culture,’’ said Cheng. ‘‘Knowing we’ll be held accountable makes everyone work harder.’’

Because BoldFlash no longer manufactured anything at corporate headquarters, employees at the
plants could feel isolated. ‘‘I can’t tell you the last time we saw anyone from corporate up here,’’ said
the Ontario plant manager. ‘‘The top people in the States don’t ask us for much input. They mainly
tell us what to make and how much by when.’’ Likewise, Product Development also caused
headaches for the plant by making frequent special requests to run trials of new products on the
manufacturing lines. This lowered efficiency and margins and jeopardized delivery dates. ‘‘Product
Development doesn’t consider the line set-up when they design the new chips, and that leads to costs
that are wrecking our margins,’’ said the manager of the Austin plant. ‘‘It’s very frustrating. All they
have to do is ask and we’d tell them exactly what we think and what they should do. But they rarely
ask. Their actions send us down the wrong track with products that we could have told them cannot
be manufactured at the cost required to make a profit.’’

Manufacturing also frequently butted heads with Sales, which they saw as selling products at any
price they could in order to meet their revenue goals regardless of the impact on margins. Cheng
recalled when they were trying to rush a chip for a new phone model to the market: ‘‘Sales promised
one of our best customers a high level of product at a low price in a short timeframe. To avoid
disappointing the customer, my Ontario plant had to deliver as promised. We lost money on that
deal. Sales got credit for a fifty-million dollar sale. We were mad and let Sales know about it in no
uncertain terms.’’

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

BoldFlash: Cross-Functional Challenges in the Mobile Division | 4438

HARVARD BUSINESS PUBLISHING | BRIEFCASES 5

Marketing

The Marketing group’s responsibilities included sales projections, long-range market planning,
and exploiting new markets. The group’s most important activity was market development——
identifying and managing the development of new product opportunities. Marketing Specialists
were currently were working on a dozen new products they were responsible for introducing. They
found they had to coordinate the activities of the other functions in order to move the product
development process along. Marketing Specialists were recent graduates of business schools, and
although they were smart and analytical, they lacked experience in the activities they were trying to
coordinate. Moreover, the role and authority of Marketing Specialists, who were in effect operating
as new product development project leaders, had never been clearly defined. Consequently they
found it difficult to get people in Product Development, Sales, and particularly the manufacturing
plants to adhere to commitments that the heads of their functional units had made in the quarterly
product development meetings. There were other problems as well. Patel said ‘‘We’re supposed to
suggest new products, but PD does that too. We’re supposed to make projections, but so is Sales.
We’re a new group, but I wish our role had been better defined before we were spun out.’’

Because new mobile devices were constantly coming to market, each requiring more storage at a
better price, the flash memory market was highly volatile, which made it difficult to predict long-
range revenues and profits of new products under development. Revenue and profit projections for
new products did not always materialize, which reduced the credibility of Marketing Specialists and
the Marketing Department as a whole. The group sometimes felt that their colleagues failed to
appreciate the difficulty of their mission and had unreasonable expectations. ‘‘We take a very
rigorous approach to our projections, but this is not an exact science. There is no ‘right answer’ to the
equation, and it can be difficult to explain that to an engineer or operations manager,’’ said Patel.

Other division leaders also struggled to understand Marketing’s mission. Manufacturing’s Kevin
Cheng said, ‘‘They should be listening to customers instead of writing so many reports——I don’t even
read them half the time.’’ Karl Melzer, head of Product Development, agreed: ‘‘Last year we
developed at least 15 product variations and Marketing couldn’t find buyers for the majority of them.
They just don’t understand the technology and how it can benefit customers.’’

Sales

Chip Bryant had spent three years at FlashForce, a larger competitor, before joining BoldFlash in
1999. Like Marketing, Sales had a substantial number of new employees. Because entry-level sales
positions in this group were seen as stepping stones to other internal roles and to other technology
companies, new college graduates actively sought them out. Competition for entry-level roles was
fierce, and expectations of rapid upward mobility led to high turnover in the group. ‘‘It’s hard to keep
a cohesive team when your top producers keep leaving because they didn’t get promoted fast
enough,’’ said Bryant.

The group developed sales projections for individual customers largely by applying ‘‘educated
instinct’’ to information it gleaned from existing customers. This approach faced frequent criticism
from colleagues in other departments, such as this observation by Marketing’s Patel: ‘‘Sales basically
has tunnel vision. I saw it when I went to our industry’s big trade show last year. Our salespeople
stayed in our booth, when they could have done some good ‘walking around’ research by visiting
our competitors. They missed a great opportunity.’’

Each of BoldFlash’s three manufacturing plants had at least one on-site sales representative to
facilitate on-time delivery and customer service. However, although Chip Bryant and Kevin Cheng
were close friends outside of work, that goodwill did not always extend to lower-level staffers. The

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

4438 | BoldFlash: Cross-Functional Challenges in the Mobile Division

6 BRIEFCASES | HARVARD BUSINESS PUBLISHING

manager of the Austin plant said, ‘‘Our sales team rarely consults with us before sending out
proposals, even though we have someone on-site. Even worse, they don’t listen when we try to tell
them they’re wrong. Last year, I tried to tell Sales there was no way we could make as many chips as
they had promised in just two weeks. Jim Harrison ordered us to ‘do it anyway,’ so we did. Well, the
system overload crashed the software that runs the line and we had to shut down for almost a day
while we waited for someone in engineering to fix it.’’

This Sales group’s performance metric was very straightforward——revenues. Incentives, not
offered to all groups, were based on performance against revenue targets.

Product Development

This science-oriented department developed advances in technology leading to new products as
well as extensions to existing product lines. Inside the Mobile Division, they were known as the
‘‘geeks.’’ Many were lifelong BoldFlash employees. All had scientific backgrounds, and over two-
thirds of managerial-level staffers held PhDs. All of the division’s patents came from the Product
Development team. Led by Dr. Karl Melzer, a native Austrian who joined BoldFlash in 1985 after
finishing his doctorate, the group tended to keep to itself. ‘‘I’ve been trying to meet with Karl to talk
about a new product launch, but he’s always off presenting at some conference,’’ said Sales Director
Bryant. ‘‘I know he’s a legend in the field, but sometimes he stays up in the clouds.’’

Product Development likewise sometimes experienced trouble communicating with both Sales
and Marketing about product development initiatives. ‘‘Anytime I ask Chip for anything, he says
‘Ask Kavita.’ When I ask Kavita, she tells me to see Chip,’’ said Melzer. ‘‘At this point I have stopped
asking. I just talk to Kevin Cheng or one of the plant managers and somehow we make it work. Kevin
can really get things done when he needs to.’’

Product Development was evaluated on a number of mostly academic criteria, including the
number of new patents approved, presentations made at industry conferences, and academic papers
published in peer-reviewed journals. These criteria hailed back to the Jim Harrison days. ‘‘When Jim
was here, our big successes came from research driven by technological curiosity, not a business
mandate. He didn’t want us worrying about day-to-day business concerns and always kept them
away from us so we could focus on our research,’’ said Melzer.

The Current Product Development Process

Mobile’s product development process mirrored practices in other BoldFlash divisions whose
markets did not change quite as rapidly as that of the Mobile Division. On paper, the process
appeared fairly structured. Four times a year, about 20 division leaders met off-site for two days to
review financial results, set sales projections, and make plans for new product releases. The head of
Market Development presided. Attendance had grown to accommodate the next level of managers
because the heads of functional departments lacked the information they needed on each new
product development to make decisions.

With the mandate from his CEO entrenched in his thinking, Cahill felt his presence would be
critical to improving his division’s results. When he attended his first meeting in November 2011, he
said little, observing and assessing instead. He later followed up with his four functional leaders and
was surprised at what he learned. Cahill said, ‘‘It turned out that the meetings were not very
effective. When a functional department missed its objective with regard to a given product
development the underlying reasons were never confronted. Participants often met afterwards in
small groups to talk about problems that had not been resolved. Meeting progress was slowed

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

BoldFlash: Cross-Functional Challenges in the Mobile Division | 4438

HARVARD BUSINESS PUBLISHING | BRIEFCASES 7

because people who were up top speed on a given new product development project had to be called
by phone for the latest information about progress in developing the new product — this despite the
fact that the meeting had grown to include the next level managers. Karl told me that in 2009 the
team had a critical discussion centered around the next generation of NAND chips, yet the Product
Development scientist in charge of that product wasn’t even invited and was actually on vacation
during the meeting.’’

The lack of structure around the product development process bothered others on the leadership
team. ‘‘This process lacks discipline and focus,’’ said Karl Melzer. ‘‘I get very frustrated every time I
have to sit in one of those meetings when I could be working on my research. I hope that Roger will
hold us to our goals instead of moving the finish line if we don’t make them.’’ The head of Market
Development, who chaired the product development meetings, confided to Cahill that he often had
sleepless nights before meetings because of the difficulties in resolving issues.

To address some of these concerns, Cahill had decided to chair the next meeting himself.
However, he was unsure how to balance the need for discipline with maintaining positive
relationships with his leadership team. ‘‘I guess I should just go in and take control, but I can’t take
the chance of alienating people,’’ he said. ‘‘I’m still not sure if I’m doing the right thing——I know that
I’m a competent scientist, but many aspects of leadership are new to me. I’m going on instinct here.’’

Conclusion

Cahill looked back at Harrison’s boxes in the corner of his office and wondered what his
predecessor would have done. ‘‘I put together a great team, so this shouldn’t be happening,’’ he
thought. ‘‘Why can’t these smart people all work together?’’ The miss in the tablet market had really
devastated the group. Morale was low, margins were squeezed, and tempers were starting to fray. If
Cahill’s group didn’t introduce successful new products quickly, the division’s future would be on
the line. It might not survive its next miss. He wondered whether the changes he had already made
and his decision to chair the next quarterly product development meeting would solve the
coordination and collaboration problems undermining the product development process. If not, what
additional organizational changes would be needed?

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

4438 | BoldFlash: Cross-Functional Challenges in the Mobile Division

8 BRIEFCASES | HARVARD BUSINESS PUBLISHING

Exhibit 1A BoldFlash Corporate Operating Results ($ thousands)

2011 2010 2009 2008 2007

Income
Total Sales 3,928,710 3,591,432 3,300,964 2,699,623 2,871,092

Expenses
Cost of Goods Sold 2,317,939 1,964,513 1,950,870 1,630,572 1,722,655

Gross Profit 1,610,771 1,626,919 1,350,094 1,069,051 1,148,437

SG&A Expense 245,544 222,669 198,058 156,578 175,137
Research & Development 628,594 628,501 524,853 377,947 367,500
Amortization/Depreciation 243,580 193,937 234,368 143,080 235,430
Total Operating Expense 3,435,657 3,009,620 2,908,149 2,308,178 2,500,721

Operating Income 493,053 581,812 392,815 391,445 370,371

Interest Income 280 200 170 102 145
Interest Expense 94 80 65 54 37

Net Income Before Taxes 493,239 581,932 392,920 391,493 370,479
Provision for Income Taxes 88,783 87,290 74,655 54,809 62,981
Net Income 404,456 494,642 318,265 336,684 307,497

Exhibit 1B BoldFlash Mobile Operating Results ($ thousands)

2011 2010 2009 2008 2007

Income
Total Sales 1,579,000 1,564,000 1,582,000 1,518,000 1,584,000

Expenses
Cost of Goods Sold 931,610 855,508 934,962 916,872 950,400
SG&A 98,688 96,968 94,920 88,044 96,624
Research & Development 314,297 339,390 304,415 192,753 191,100
Total Expenses 1,344,594 1,291,866 1,334,297 1,197,669 1,238,124

Net Incomea 234,406 272,134 247,703 320,331 345,876

a Amortization, depreciation, interest, and taxes are reported at the corporate level.

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

BoldFlash: Cross-Functional Challenges in the Mobile Division | 4438

HARVARD BUSINESS PUBLISHING | BRIEFCASES 9

Exhibit 2A BoldFlash Organization Chart

Vice President
Consumer Division

Vice President
Mobile Division

Vice President
OEM Division

Vice President
R&D

Vice President and
Chief Financial Officer

Vice President and
Chief Operating Officer

Vice President and
Chief Administrative Officer

Vice President
International

Vice President
Marketing and Sales

President
and CEO

For the exclusive use of R. Valldas, 2022.

This document is authorized for use only by Rishika Valldas in TCM 710-898 Project Leadership-Spring 2022 taught by RICHARD CALLAHAN, Missouri State University from Feb 2022 to Jun
2022.

44
38

–

10
–

E
xh

ib
it

2
B

B
ol

d
F

la
sh

M
ob

il
e

D
iv

is
io

n
O

rg
an

iz
at

io
n

C
h

ar
t

M
an

ag
er

A
us

tin
, T

X
P

la
nt

M
an

ag
er

T
or

on
to

, O
N

P
la

nt

M
an

ag
er

Sh
an

gh
ai

, C
hi

na
P

la
nt

K
ev

in
C

he
ng

D
ir

ec
to

r,
M

an
uf

ac
tu

ri
ng

M
an

ag
er

Pr
od

uc
t E

xt
en

si
on

s

M
an

ag
er

N
ew

P
ro

du
ct

E
ng

in
ee

ri
ng

M
an

ag
er

In
te