Category Archives: Venture Capital

How to Present an Operating Plan to your Board

I’ve been CEO of two startups and on the board of about ten.  That means I’ve presented a lot of operating plans to boards.  It also means I’ve had a lot of operating plans presented to me.  Frankly, most of the time, I don’t love how they’re presented.  Common problems include:

  • Lack of strategic context: management shows up with a budget more than a plan, and without explaining the strategic thinking (one wonders, if any) behind it.  For a primer, see here.
  • Lack of organizational design: management fails to show the proposed high-level organizational structure and how it supports the strategy.  They fail to show the alternative designs considered and why they settled on the one they’re proposing.
  • A laundry list of goals. OKRs are great.  But you should have a fairly small set – no more than 5 to 7 – and, again, management needs to show how they’re linked to the strategy.

Finance types on the board might view these as simple canapes served before the meal.  I view them as critical strategic context.  But, either way, the one thing on which everyone can agree is that the numbers are always the main course. Thus, in this post, I’m going to focus on how to best present the numbers in an annual operating plan.

Context is King
Strategic context isn’t the only context that’s typically missing.  A good operating plan should present financial context as well.  Your typical VC board member might sit on 8-10 boards, a typical independent on 2 (if they’re still in an operating role), and a professional independent might sit on 3-5.  While these people are generally pretty quantitative, that’s nevertheless a lot of numbers to memorize.  So, present context.  Specifically:

  • One year of history. This year that’s 2021.
  • One year of forecast. This year that’s your 2022 forecast, which is your first through third quarter actuals combined with your fourth-quarter forecast.
  • The proposed operating plan (2023).
  • The trajectory on which the proposed operating plan puts you for the next two years after that (i.e., 2024 and 2025).

The last point is critical for several reasons:

  • The oldest trick in the book is to hit 2023 financial goals (e.g., burn) by failing to invest in the second half of 2023 for growth in 2024.
  • The best way to prevent that is to show the 2024 model teed up by the proposed 2023 plan. That model doesn’t need to be made at the same granularity (e.g., months vs. quarters) or detail (e.g., mapping to GL accounts) as the proposed plan – but it can’t be pure fiction either.  Building this basically requires dovetailing a driver-based model to your proposed operating plan.
  • Showing the model for the out years helps generate board consensus on trajectory. While technically the board is only approving the proposed 2023 operating plan, that plan has a 2024 and 2025 model attached to it.  Thus, it’s pretty hard for the board to say they’re shocked when you begin the 2024 planning discussion using the 2024 model (that’s been shown for two years) as the starting point.

Presenting the Plan in Two Slides
To steal a line from Name That Tune, I think I can present an operating plan in two slides.  Well, as they say on the show:  “Dave, then present that plan!”

  • The first slide is focused on the ARR leaky bucket, metrics derived from ARR, and ARR-related product.ivity measures
  • The second slide is focused on the P&L and related measures.

There are subjective distinctions in play here.  For example, CAC ratio (the S&M cost of a dollar of new ARR) is certainly ARR-related, but it’s also P&L-driven because the S&M cost comes from the P&L.  I did my best to split things in a way that I think is logical and, more importantly, between the two slides I include all of the major things I want to see in an operating plan presentation and, even more importantly, none of the things that I don’t.

Slide 1: The Leaky Bucket of ARR and Related Metrics

Let’s review the lines, starting with the first block, the leaky bucket itself:

  • Starting ARR is the ARR level at the start of a period. The starting water level of the bucket.
  • New ARR is the sum of new logo (aka, new customer) ARR and expansion ARR (i.e., new ARR from existing customers). That amount of “water” the company poured into the bucket.
  • Churn ARR is the sum of ARR lost due to shrinking customers (aka, downsell) and lost customers. The amount of water that leaked out of the bucket.
  • Ending ARR is starting ARR + new ARR – churn ARR. (It’s + churn ARR if you assign a negative sign to churn, which I usually do.)  The ending water level of the bucket.
  • YoY growth % is the year-over-year growth of ending ARR. How fast the water level is changing in the bucket.  If I had to value a SaaS company with only two numbers, they would be ARR and YoY ARR growth rate.  Monthly SaaS companies often have a strong focus on sequential (QoQ) growth, so you can add a row for that too, if desired.

The next block has two rows focused on change in the ARR bucket:

  • Net new ARR = new ARR – churn ARR. The change in water level of the bucket.  Note that some people use “net new” to mean “net new customer” (i.e., new logo) which I find confusing.
  • Burn ratio = cashflow from operations / net new ARR. How much cash you consume to increase the water level of the bucket by $1.  Not to be confused with cash conversion score which is defined as an inception-to-date metric, not a period metric.  This ratio is similar to the CAC ratio, but done on a net-new ARR basis and for all cash consumption, not just S&M expense.

The next block looks at new vs. churn ARR growth as well as the mix within new ARR:

  • YoY growth in new ARR. The rate of growth in water added to the bucket.
  • YoY growth in churn ARR. The rate of growth in water leaking from the bucket.  I like putting them next to each other to see if one is growing faster than the other.
  • Expansion ARR as % of new ARR. Percent of new ARR that comes from existing customers.  The simplest metric to determine if you’re putting correct focus on the existing customer base.  Too low (e.g., 10%) and you’re likely ignoring them.  Too high (e.g., 40%) and people start to wonder why you’re not acquiring more new customers. (In a small-initial-land and big-expand model, this may run much higher than 30-40%, but that also depends on the definition of land – i.e., is the “land” just the first order or the total value of subscriptions acquired in the first 6 or 12 months.)

The next block focuses on retention rates:

  • Net dollar retention = current ARR from year-ago cohort / year-ago ARR from year-ago cohort. As I predicted a few years back, NRR has largely replaced LTV/CAC, because of the flaws with lifetime value (LTV) discussed in my SaaStr 2020 talk, Churn is Dead, Long Live Net Dollar Retention.
  • Gross dollar retention = current ARR from year-ago cohort excluding expansion / year-ago ARR from year-ago cohort. Excluding the offsetting effects of expansion, how much do customer cohorts shrink over a year?
  • Churn rate (ATR-based) = churn ARR/available-to-renew ARR. Percent of ARR that churns measured against only that eligible for renewal and not the entire ARR base.  An important metric for companies that do multi-year deals as putting effectively auto-renewing customers in the denominator damps out

The next block focuses on headcount:

  • Total employees, at end of period.
  • Quota-carrying reps (QCRs) = number of quota-carrying sellers at end of period. Includes those ramping, though I’ve argued that enterprise SaaS could also use a same-store sales metric.  In deeper presentations, you should also look at QCR density.
  • Customer success managers (CSMs) = the number of account managers in customer success. These organizations can explode so I’m always watching ARR/CSM and looking out for stealth CSM-like resources (e.g., customer success architects, technical account managers) that should arguably be included here or tracked in an additional row in deeper reports.
  • Code-committing developers (CCDs) = the number of developers in the company who, as Elon Musk might say, “actually write software.” Like sales, you should watch developer density to ensure organizations don’t get an imbalanced helper/doer ratio.

The final block looks at ARR-based productivity measures:

  • New ARR/ramped rep = new ARR from ramped reps / number of ramped reps. This is roughly “same-store sales [link].”  Almost no one tracks this, but it is one of several sales productivity metrics that I like which circle terminal productivity.  The rep ramp chart’s 4Q+ productivity is another way of getting at it.
  • ARR/CSM = starting ARR/number of CSMs, which measures how much ARR each CSM is managing.  Potentially include stealth CSMs in the form of support roles like technical account manager (TAM) or customer success architects (CSAs).
  • ARR/employee = ending ARR/ending employees, a gross overall measure of employee productivity.

Slide 2: The P&L and Related Metrics

This is a pretty standard, abbreviated SaaS P&L.

The first block is revenue, optionally split by subscription vs. services.

The second block is cost of goods sold.

The third block is gross margin.  It’s important to see both subscription and overall (aka, blended) gross margin for benchmarking purposes.  Subscription gross is margin, by the way, is probably the most overlooked-yet-important SaaS metric.  Bad subscription margins can kill an investment deal faster than a high churn rate.

The fourth block is operating expense (opex) by major category, which is useful for benchmarking.  It’s also useful for what I call glideslope planning, which you can use to agree with the board on a longer-term financial model and the path to get there.

The penultimate block shows a few more SaaS metrics.

  • CAC ratio = S&M cost of a $1 in new ARR
  • CAC payback period  = months of subscription gross profit to repay customer acquisition cost
  • Rule of 40 score = revenue growth rate + free cashflow margin

The last block is just one row:  ending cash.  The oxygen level for any business.  You should let this go negative (in your financial models only!) to indicate the need for future fundraising.

Scenario Comparisons
Finally, part of the planning process is discussing multiple options, often called scenarios.

While scenarios in the strategy sense are usually driven by strategic planning assumptions (e.g., “cheap oil”), in software they are often just different version of a plan optimized for different things:

  • Baseline: the default proposal that management usually thinks best meets all of the various goals and constraints.
  • Growth: an option that optimizes growth typically at the expense or hitting cash, CAC, or S&M expense goals.
  • Profit: an option that optimizes for cash runway, often at the expense of growth, innovation, or customer satisfaction.

Whatever scenarios you pick, and your reasons for picking them, are up to you.  But I want to help you present them in a way that is easy to grasp and compare.

Here’s one way to do that:

I like this hybrid format because it’s pulling only a handful of the most important rows, but laying them out with some historical context and, for each of the three proposed scenarios, showing not only the proposed 2023 plan also the 2024 model associated with it.  This is the kind of slide I want to look at while having a discussion about the relative merits of each scenario.

What’s Missing Here?
You can’t put everything on two slides.  The most important things I’m worried about missing in this format are:

  • Segment analysis: sometimes your business is a blended average of multiple different businesses (e.g., self-serve motion, enterprise motion) and thus it’s less meaningful to analyze the average than to look at its underlying components.  You’ll need to add probably one section per segment in order to address this.
  • Strategic challenges. For example, suppose that you’ve always struggled with enterprise customer CAC.  You may need to add one section focused solely on that.  “Yes, that’s the overall plan, but it’s contingent on getting cost/oppty to $X and the win rate to Y% and here’s the plan to do that.”
  • Zero-based budgeting. In tough times, this is a valuable approach to help CEOs and CFOs squeeze cost out of the business.  It takes more time, but it properly puts focus on overall spend and not simply on year-over-year increments.  In a perfect world, the board wouldn’t need to see any artifacts from the process, but only know that the expense models are tight because every expense was scrutinized using a zero-based budgeting process.

Conclusion
Hopefully this post has given you some ideas on how to better present your next operating plan to your board.  If you have questions or feedback let me know.  And I wish everyone a happy and successful completion of planning season.

You can download the spreadsheet used in this post, here.

The Balderton Founder’s Guide to B2B Sales

Working in my capacity at as an EIR at Balderton Capital, I have recently written a new publication, The Balderton Founder’s Guide to B2B Sales, with the able support of Balderton Principal Michael Lavner and the entire Balderton Capital team.  This guide is effectively a new edition, and a new take, on the prior, excellent B2B Sales Playbook.

The guide, which is now published as a microsite, will soon be available in PDF format for downloading.

I’ll put the opening quote here that the editors omitted because it’s nearly unparseable:

“I have learned everything I need to know about sales.  Sales is saying ‘yes’ in response to every question.  So, now, when a customer asks if the product has a capability that it currently lacks, I say, ‘yes, the product can’t do that.'”

— Anonymous CS PhD founder who didn’t quite learn everything they needed to know about sales.

In short, this guide’s written for you, i.e., the product-oriented founder who thought they founded a technology business only to discover that SaaS companies, on average, spend twice as much on S&M as they do on R&D, and ergo are actually running a distribution business.

The guide has seven parts:

  • Selling: what founders need to know about sales
  • Building: how to build a sales organization
  • Managing: how to manage a sales organization
  • Renewing/expanding: teaming sales and customer success
  • Marketing: using marketing to build sales pipeline
  • Partnering: how to use partners to improve reach and win rate
  • Planning: planning and the role of key metrics and benchmarks

While there are numerous good SaaS benchmarking resources out there, the guide includes some benchmark figures from the Balderton universe (i.e., European, top-tier startups) and — hint, hint — we expect to release those benchmarks more fully and in a more interactive tool in the not-too-distant future.

The guide is also chock full of links which I will attempt to maintain as sources change over time.  But I’ve written it with both in-line links (often to Kellblog) and end-of-section links that generally point to third-party resources.

I’ve packed 30 years of enterprise software experience into this.  I come at sales from an analytical viewpoint which I think should be relatable for most product-oriented founders who, like me, get turned off by claims that sales has to be artisanal magic instead of industrial process.

I hope you enjoy the guide.  Feel free to leave comments here, DM me on Twitter, or reach me at the contact information in my FAQ.

Key Takeaways from the 2022 KeyBanc SaaS Metrics Survey

KeyBanc Capital Markets (KBCM) recently published their 13th annual private SaaS company survey.  This post has three purposes:  to let you know it’s out, to provide you with a link so you can get it, and to offer some quick takeaways on skimming through the results.

The first thing to remember about this survey is that it’s private SaaS companies.  Unlike Meritech Public Comps, where you can see metrics for the best [1], public SaaS companies, this private company data is somewhat harder to come by (the only other source that springs to mind is RevOps Squared) and, for most of us, it provides much more realistic comparables than Meritech [2].

The second thing to remember is that there are a lot of smaller companies in the sample:  about 20% of respondents are less than $5M in ARR and about 40% are less than $10M.   (The overall median is $13MM.)  Depending on who you want to compare to, this may be a good or a bad thing.  In addition, for most of the metrics they exclude companies <$5M in ARR from the calculations, which brings up the overall median for that set to $17.6M.

Net:  this is not VC-backed SaaS companies (62% are), this is not IPO-track SaaS companies (presumably some small subset of that 62%).  This is all private SaaS companies, including 22% PE-backed and 13% boostrapped.

One of my new benchmarking themes is that people need to pay more attention to matching their benchmarks with their aspirations. If your aspirations are to raise money from top VCs at a good valuation, my guess is you should be thinking 75th precentile of this data set; if they’re to IPO, you should be thinking 90th.

That said, let’s meet the Joneses, who have median:

  • ARR growth of 31%, lower than I’d hope.
  • Forecast 2022 ARR growth of 36%, so they’re planning to accelerate.  Everyone’s an optimist.
  • Expansion ARR of 46%, higher than I’d hope.
  • Net dollar retention (NDR) of 109%.
  • Customer acquisition cost (CAC) ratio of 1.2 blended, 1.8 new, and 0.6 expansion, in line with my expectations.
  • Gross churn of 14%, in line, perhaps a tad high, relative to my guess.
  • Available to renew (ATR) gross churn of 10%, but it’s hard to understand how ATR rate can be lower than gross churn rate [3].
  • Margin profile of 77% subscription, 73% blended.  In line.
  • Sales and marketing (S&M) expense of 40% of revenues.  They’re frugal, but they’re not growing that fast, either.
  • Free cashflow (FCF) margin of -5%.
  • New ARR per seller of $673K, which I if I understand, is what I’d call sales productivity.
  • Contract length and billing frequency of one year.
  • ARR/FTE of $143K, lower than I’d guess (for public companies it’s nearly double that).
  • Valuation of 6.1x ARR at their most recent round (in 2021 or later).

Since I don’t want to lift too many of their slides, I’ll extract just two.  The first shows S&M spend as a function of growth rate.

If there’s one area where you really need to look at metrics as a function of growth rate, it’s customer acquistion cost and, by extension S&M spend, on the theory that in enterprise SaaS you need to invest up front to grow.  Therefore a high-growth company is theoretically carrying the cost of as-yet-unproductive capacity where as a steady-state one is not.  You can see this pretty clearly here where the sub-20% growth companies spend 27% on S&M, which surprisingly drops to 17% at the 30-40% bucket, but then begins a steady upward march to 59% for those growing faster than 80%.

The second discusses a concept I’ve called The Rule of 56789

Here, KeyBanc is saying roughly what I say, which is [4]:

  • 5 years to $10M (5.6 years, per KCBM)
  • 6 years to $20M (7.1 years, but to $25M)
  • 7 years to $50M (7.6 years)
  • 8 years to $75M (they have no threshold here)
  • 9 years to $100M (9.3 years)

I’m glad they’re now tracking this, along with net burn rate (aka, cash conversion score) though I’d say their implied cash conversion scores are more efficient than I’d guess based on my experience and Bessemer’s data.

Overall, this is a seminal report for SaaS companies.  Every private SaaS company should read it.  Grab yours here.

Notes

[1]  In the sense that even a “bad” public SaaS company (dare I suggest Domo or C3 as two of my favorites to scrutinize) was still good enough to get public in the first place and ergo creme de la creme when viewed more broadly.

[2]  As I said in a recent speech, it’s the difference between benchmark off all SAT test takers and Ivy League applicants.  See slide 13 of this presentation.

[3]  KBCM calls this non-renewal rate, but I think it’s 1 – ATR churn.  The reason it’s hard to believe it’s lower is that it should be the same numerator over a smaller denominator.

[4]  I was looking at European 75th percentiles and they are looking at worldwide (but US-weighted) medians

Slides from my SaaStock Dublin Presentation on GTM Efficiency

Just a quick post to share the slides I presented at SaaStock Dublin today on driving go-to-market (GTM) efficiences over the coming 24 months.  I chose this topic because extending runway is on everyone’s mind and — because it’s usually the single largest contributor to overall operating expense — sales & marketing (S&M) is where companies turn to do so.

After a brief review of the problem, I look at two popular approaches that don’t work:

  • The Excel-induced hallucination, where you make seemingly small but unsupported tweaks to your GTM funnel model that result in massive (and totally unrealistic) productivity gains.
  • Everyone for themselves!  A Lord of the Flies approach, which sales usually wins, resulting in too many mouths to feed with too few supporting resources.

Newly hired sales reps waiting for pipeline

What does is work is to adopt a three-musketeers attitude across sales, marketing, customer success, and professional services.  (Yes, there actually were four muskeeters; they picked up d’Artagnan along the way.)

All for one and one for all to maximize ARR

I then run through a punch list of ideas, some obvious and some less so, structured in four groups, about how you can drive GTM efficiency:

  • Work better together
  • Shoot at richer targets
  • Forward-deploy more resources
  • Improve operating efficiency

The slides are embedded below.  Note that the Slideshare previewer sometimes doesn’t mix well with the Balderton fonts, so I uploaded only a PDF to Slideshare.  If you want it in PowerPoint, go to Google drive here.

 

 

Playing Bigger vs. Playing To Win: How Shall We Play the Marketing Strategy Game?

“I’m an CMO and it’s 2018.  Of course I’ve read Play Bigger.  Duh.  Do you think I live under a rock?” — Anonymous repeat CMO

Play Bigger hit the Sand Hill Road scene in a big way after its publication in 2016.  Like Geoffrey Moore’s Crossing the Chasm some 25 years earlier, VCs fell in love with the book, and then pushed it down to the CEOs and CMOs of their portfolio companies.  “Sell high” is the old sales rule, and the business of Silicon Valley marketing strategy books is no exception.

Why did VCs like the book?  Because it’s ultimately about value creation which is, after all, exactly what VCs do.  In extreme distillation, Play Bigger argues:

  • Category kings (companies who typically define and then own categories in the minds of buyers) are worth a whole lot more than runner-ups.
  • Therefore you should be a category king.
  • You do that not by simply creating a category (which is kind of yesterday’s obsession), but by designing a great product, a great company, and a great category all the same time.
  • So, off you go.  Do that.  See you at the next board meeting.

I find the book a tad simplistic and pop marketing-y (in the Ries & Trout sense) and more than a tad revisionist in telling stories I know first-hand which feel rather twisted to map to the narrative.  Nevertheless, much as I’ve read a bunch of Ries & Trout books, I have read Play Bigger, twice, both because it’s a good marketing book, and because it’s de rigeur in Silicon Valley.  If you’ve not read it, you should.  You’ll be more interesting at cocktail parties.

As with any marketing book, there is no shortage of metaphors.  Geoffrey Moore  had D-Day, bowling alleys, and tornados.  These guys run the whole “something old, something new, something borrowed, and something blue” gamut with lightning strikes (old, fka blitzkreigs), pirates (new to me if not Steve Jobs), flywheels (borrowed from Jim Collins), and gravity (blue in sense of a relentless negative force as described in several cautionary tales).

While I consider Play Bigger a good book on category creation, even a modernized version of Inside the Tornado if I’m feeling generous, I must admit there’s one would-be major distinction that I just don’t get:  category creation vs. category design, the latter somehow being not just about creating and dominating a category, but “designing” it — and not just a category, but a product, category, and company simultaneously.  It strikes me as much ado about little (you need to build a company and a product to create and lead a category) and, skeptically, a seeming pretense for introducing the fashionable word, “design.”

After 30 years playing a part in creating, I mean designing, new categories — both ones that succeeded (e.g., relational database, business intelligence, cloud EPM, customer success management, data intelligence) and ones that didn’t (e.g., XML database, object database) — I firmly believe two things:

  • The best way to create a category is to go sell some software.  Early-stage startups excessively focused on category creation are trying to win the game by staring at the scoreboard.
  • The best way to be a category king is to be the most aggressive company during the growth phase of the market.  Do that by executing what I call the market leader play, the rough equivalent of Geoffrey Moore’s “just ship” during the tornado.  Second prize really is a set of steak knives.

I have some secondary beliefs on category creation as well:

  • Market forces create categories, not vendors.  Vendors are simply in the right place (or pivot to it) at the right time which gives them the opportunity to become the category king.  It’s more about exploiting opportunities than creating markets.  Much as I love GainSight, for example, I believe their key accomplishment was not creating the customer success category, but outexecuting everyone else in exploiting the opportunity created by the emergence of the VP of Customer Success role.  GainSight didn’t create the VP of Customer Success; they built the app to serve them and then aggressively dominated that market.
  • Analysts name categories, not vendors.  A lot of startups spend way too much time navel gazing about the name for their new category.  Instead of trying to sell software to solve customer problems, they sit in conference rooms wordsmithing.  Don’t do this.  Get a good-enough name to answer the question “what is it?” and then go sell some.  In the end, as a wise, old man once told me, analysts name categories, not vendors.
  • Category names don’t matter that much.  Lots of great companies were built on pretty terrible category names (e.g., ERP, HCM, EPM, BTO, NoSQL).  I have trouble even telling you what category red-hot tech companies like Hashicorp and Confluent even compete in.  Don’t obsess over the name.  Yes, a bad name can hurt you (e.g., multi-dimensional database which set off IT threat radar vs. OLAP server, which didn’t).  But it’s not really about the name.  It’s about what you sell to whom to solve which problem.  Again, think “good enough,” and then let a Gartner or IDC analyst decide the official category name later.

To hear an interesting conversation on category creation,  listen to Thomas Otter, Stephanie McReynolds, and me discuss the topic for 60 minutes.  Stephanie ran marketing at Alation, which successfully created (or should I say seized on the market-created opportunity to define and dominate) the data catalog category.  (It’s all the more interesting because that category itself is now morphing into data intelligence.)

Since we’re talking about the marketing strategy game, I want to introduce another book, less popular in Silicon Valley but one that nevertheless deserves your attention: Playing to Win.  This book was written not by Silicon Valley denizens turned consultants, but by the CEO of Proctor & Gamble and his presumably favorite strategy advisor.  It’s a very different book that comes from a very different place, but it’s right up there with Blue Ocean Strategy, Inside the Tornado, and Good Strategy, Bad Strategy on my list of top strategy books.

Why?

  • Consumer packaged goods (CPG) is the major league of marketing.   If they can differentiate rice, yogurt, or face cream, then we should be able to differentiate our significantly more complex and inherently differentiated products.  We have lots to learn from them.
  • I love the emphasis on winning.  In reality, we’re not trying to create a category.  We’re trying to win one, whether we happened to create it or not.  Strategy should inherently be about winning.  Strategy, as Roger Burgelman says, is the plan to win.  Let’s not dance around that.
  • I love the Olay story, which opens the book and alone is worth the price of the book.  Take an aging asset with the wrong product at the wrong price point in the wrong channel and, instead of just throwing it away, build something amazing from it.  I love it.  Goosebumps.
  • It’s practical and applied.  Instead of burying you in metaphors, it asks you to answer five simple questions.  No pirates, no oceans, no tornados, no thunderstorms, no gorillas, no kings, no beaches.

Those five questions:

  • What is your winning aspiration? The purpose of your enterprise, its motivating aspiration.
  • Where will you play? A playing field where you can achieve that aspiration.
  • How will you win? The way you will win on the chosen playing field.
  • What capabilities must be in place? The set and configuration of capabilities required to win in the chosen way.
  • What management systems are required? The systems and measures that enable the capabilities and support

Much as I love metaphors, I’d bury them all in the backyard in exchange for good answers to those five questions.  Strategy is not complex, but it is hard.  You need to make clear choices, which business people generally resist.  It’s far easier to fence sit, see both sides of the issue, and keep options open (which my old friend Larry used to call the MBA credo).  That’s why most strategy isn’t.

Strategy is about answering those questions in a way that is self-consistent, consistent with the goals of the parent organization (if you’re a brand or general manager in a multi-product company), and with the core capabilities of the overall organization.

In our view, Olay succeeded because it had an integrated set of five strategic choices that fit beautifully with the choices of the corporate parent. Because the choices were well integrated and reinforced category-, sector-, and company-level choices, succeeding at the Olay brand level actually helped deliver on the strategies above it.

I won’t summarize the entire book, but just cherrypick several points from it:

  • As with Burgelman, playing to win requires you to define winning for your organization in your context.  How can we make the plan to win if we don’t agree on what winning is?  (How many startups desperately need to have the “what is winning” conversation?)
  • Playing to win vs. playing to play.  Which are you doing?  A lot of people are doing the latter.
  • Do think about competition.  Silicon Valley today is overloaded with revisionist history:  “all we ever focused on was our customers” or “we always focused only on our vision, our north star.”  Ignoring competition is the luxury of retired executives on Montana ranches.  Winning definitionally means beating the competition.  You shouldn’t be obsessed with the competition, but you can’t ignore them either.
  • While they don’t quite say it, deciding where you play is arguably even more important than deciding what you sell.  Most startups spend most of their energy on what (i.e., product), not where (i.e., segment).  “Choosing where to play is also about choosing where not to play,” which for many is a far more difficult decision.
  • The story of Impress, a great technology, a product that consumers loved, but where P&G found no way to win in the market (and ultimately created a successful joint venture with Clorox instead), should be required reading for all tech marketers.  A great product isn’t enough.  You need to find a way to win the market, too.
  • The P&G baby diapers saga sounds similar to what would have happened had Oracle backed XQuery or when IBM originally backed SQL — self-imposed disruptions that allowed competitors entry to the market.  IBM accidentally created Oracle in the process.  Oracle was too smart to repeat the mistake.  Tech strategic choices often have their consumer analogs and they’re sometimes easier to analyze in that more distant light.
  • The stories of consumer research reveal a depth of desired customer understanding that we generally lack in tech.  We need to spend more time in customers’ houses, watching them shave, before we build them a razor.  Asking them about shaving is not enough.
  • I want to hug the person who described the P&G strategy process as, “corporate theater at its best.”  Too much strategy is exactly that.

Overall, it’s a well-written, well-structured book.  Almost all of it applies directly to tech, with the exception of the brand/parent-company intersection discussions which only start to become applicable when you launch your second product, usually in the $100M to $300M ARR range.  If you don’t have time for the whole book, the do’s and don’ts at the end of each chapter work as great summaries.

To wrap this up, I’d recommend both books.  When thinking about category creation, I’d try to Play Bigger.  But I’d always, always be Playing to Win.