Three Marketing Lessons from The Realm of Politics

Silicon Valley marketing communications are, simply put, not the major league.  By comparison to Washington, DC and political communications, we are AAA baseball [1].  In fact, to be less kind, if DC communications are the major league, you could argue that consumer marketing is AAA, and we in Silicon Valley are only AA.  We play for the love of the game [2].

Without overstretching the metaphor, let’s try and agree to two things:

  • We aren’t the top league.
  • Therefore, we can learn from studying the leagues above us.

I study the higher leagues from time to time in this blog, e.g., by looking at consumer marketing cases such as the goosebump-inspiring Olay example in Playing to Win. But I generally refrain from studying political examples [3] for many reasons, mostly for fear that I’ll end up in political arguments when my actual point is to study marketing and communications techniques, and not whether I agree or disagree with what someone stands for and/or is saying.

For example, while I don’t necessarily agree with Frank Luntz’s politics, I have great respect for his work.  Words That Work [4] is a great book.  He’s amazing at linguistic reframing (e.g., climate change vs. global warming).  He relies on a heavily research- and data-driven approach to communications, including numerous focus groups that he often personally runs.  Like him or not, agree with his views or not, the man is not afraid to roll up his sleeves and he is good at what he does.

In that spirit, in today’s post, I’m going to discuss lessons marketers can learn from today’s political right.  I pick the right because I think they execute against three principles particularly well [5]:

  • Demonstrate an understanding of the problem.
  • Framing is everything.
  • The power of consistency.

Demonstrate an Understanding of the Problem

Politicians know that the problem is safer ground than the solution.  Think: “I’m in favor of food for the hungry.”  It’s hard to disagree with that.  The devil, of course, is in the detail of how you want to do it. To illustrate this, let’s break down a marketing message into three parts:

  • Problem — describing the problem and its consequences, empathizing.
  • Solution — presenting the solution to the problem, naming and explaining.
  • Proof — providing evidence that the solution will work, typically via technical explanations and/or customer stories and references.

For example, let’s use this recent Trump CPAC excerpt (and we’re going to ignore the diction), break it down, and determine the percent of the lines used in each of those three areas:

Before Biden came into office, we had illegal immigration at a record low, refugees were at the lowest level in history. Human trafficking, women and children was at the lowest in 30 years. And drug dealers were finding the US border a very inhospitable place to be. It was very inhospitable. In my last year, less drugs came through the southern border than had been seen in many, many decades. We weren’t playing games. Now we have complete chaos. Fentanyl is pouring in. Families are being wiped out, destroyed, and there’s death everywhere, all caused by incompetence. Millions of illegal aliens are stampeding across our border. Interior enforcement has been shut down. Everyone is overstaying their visas. Nobody even thinks about reporting it anymore. My wonderful travel ban is gone. I had a travel ban, it was so wonderful.  Refugee numbers are through the roof. And spies and terrorists are infiltrating our country totally unchecked like never before.

When I’m back in the White House, the very first reconciliation bill I will sign will be for a massive increase in Border Patrol and a colossal increase in the number of ice deportation officers […] Under my leadership, we will use all necessary state, local, federal, and military resources to carry out the largest domestic deportation operation in American history. Other countries are emptying out their prisons, insane asylums and mental institutions and sending all of their problems right into their dumping ground, the USA. Think of it, they’re emptying out their prisons, and you’ve heard me say that, but they’re also emptying out their mental institutions and to use a strong couple of words, insane asylum.

Now, let’s analyze it:

  • Problem:  illegal immigrants, drugs, spies, refugees, other countries emptying prisons, insane asylums, and mental institutions into the US.  (58%)
  • Solution:  increase border patrol budget, largest deportation effort in history.  (21%)
  • Proof:  claims that problems were at record lows during prior tenure. (21%) [6]

The underlying logic:  if you don’t take the time to demonstrate an understanding of the problem and its impacts, then why would I care about your solution and how it might work?  Convince me you understand the problem, care about the problem, make me feel seen and heard, and then I might give you a chance to try and solve it.

Consider the rise of the viral country hit, Rich Men North of Richmond.  While we can demand solutions and proof from politicians, it’s not a reasonable ask for singers, so I won’t decompose the lyrics here.  I will, however, share two reactions from the target audience on understanding the problem:

“And just like that you became the voice of 40 or 50 million working men,” read one comment that received 11,000 likes.

“You’ve captured the anger, the angst, and the disbelief of every hard-working, law-abiding, patriotic American who can’t believe what our country has become.”

People outside the target audience have plenty to say, too but I won’t jump into that fray.  I will say that Oliver Anthony demonstrated an understanding of the problem to his audience.  That, plus a pretty husky singing voice, is how you rise to #1 on Apple, Spotify, and iTunes in just a few days.

Now, let’s zoom back to Silicon Valley and think about how a sales team might allocate their energy across problem/solution/proof.  Example (dramatized):

Problem:  yes, we’re aware of the problem with totaling some types of measures at the end of a period.  That’s called semi-additive measures, it’s common in OLAP systems, and for what it’s worth Excel doesn’t handle it well, either.

Solution:  the schmumbleator engine understands semi-additive measures and let me tell you how that works …

Proof:  we invented the schmumbleator after our founder graduated MIT and he decided to make an OLAP engine that used metadata to overload functions like TOTAL.  So, when the schmumbleator TOTALs an additive measure, the value for the year will be the sum of the four quarters, whereas when it TOTALs a semi-additive measure, like headcount, the value will not be the sum of the four quarters, but instead the period value for the fourth quarter.  By the way, this is kind of recursive because just like headcount is semi-additive across quarters of the year, it’s also semi-additive across months of the quarter, right?  Q1 headcount isn’t the sum of January through March, it’s just March.  The schmumbleator can do a lot of other interesting things as well.  I love telling people about the schmumbleator, …

What are we doing wrong here? Lots.

  • Not talking enough about the problem.  Is the customer convinced we understand the problem and that we understand its impacts?  Do we understand how the problem affects them personally?  Would the customer say, “they get me” at the end of this interaction?
  • Not talking about enough about the solution, either. 
  • Spending all our time talking about the proof.  Presumably, that’s what interests us — “wait, this is really cool” — if not the customer.
  • Offering only technical explanations as proof, not offering any stories about companies like theirs who also faced the problem, solved it with our product, and received benefits X, Y, and Z.
  • Speaking in technical language and jargon.  Something else a smart politician would never do, but all too common in Silicon Valley.

Thus, our first lesson:  spend more time demonstrating an understanding of the problem, its impacts, and empathizing with the customer.  Spend less time on proof — and offer the right kind of proof for the situation.  Sometimes proof means a deep technical explanation (so write a white paper), sometimes it’s a reference story, and sometimes it’s just a smiling, “that’s what we do here.”

Framing is Everything

Allow me to introduce Kellogg’s Two Rules of Communications:

  1. Framing is everything
  2. See rule 1

I somewhat arbitrarily break framing into three levels:

  • Issue:  what are we actually talking about?
  • Narrative:  what’s the bigger story in play?
  • Linguistic:  what words do we use to describe it?

Issue-level framing answers the question, what are we actually talking about?

  • Air-traffic controller working conditions or an oath?  (Reagan in the 1981 PATCO strike.)
  • A candidate’s age or experience?  (Reagan in the 1984 debate.  Yes, he was great at framing.)
  • Protecting life or the right to make one’s own medical decisions?  (You’re familiar with this one.)
  • The freedom to practice one’s religion or the right to discriminate against others?  (Ditto.)

In general, if you win the framing, you win the argument.  Issue-level framing works because — if you can get away with it — you split the issue into A vs. B where there is a fairly obvious choice between the two.  Examples:

  • Should people stick to their oaths?  Well, yes.  I think they should.
  • Should people be able to practice their chosen religion?  Well, yes.  I think the country was founded on that.

Narrative-level framing kicks it up a level.  It answers the question:  what’s the bigger story here?  Examples: 

  • All these indictments and investigations, they’re just part of an ongoing witch hunt designed to interfere with the 2024 election and prevent Trump from being president. 
  • This is really the story of a con man, someone who’s stiffed contractors, evaded taxes, and bankrupted casinos — someone who’d lie to you about the time of day just for the practice. [7]

Narrative-level framing works by ingesting every new story into a bigger narrative.  It moves the attention from the  individual story (e.g., Trump was indicted on 13 counts including racketeering) to the bigger, more favorably framed narrative.  And it can work:

He suggested Trump’s opponents are using the charges to impede his electability.  “They’re trying their very best they can to keep him from running,” Nannet said. “Because they know they can’t beat him.”

The idea is to string together a series of events into a bigger narrative so that each new story just feeds the narrative.  That allows you to respond consistently (see next rule) to the series of events, instead of making a specific response each time.

Linguistic-level framing answers the question, what words do we use to describe this? 

  • Gaming vs. gambling
  • Energy exploration vs. drilling
  • Death tax vs. estate tax
  • Obamacare vs. the ACA
  • Entitlements vs. social security [8]

To contrast, issue-level framing is about concepts:  is refusing to bake a cake an act of religious freedom or an act of discrimination?  Linguistic-level framing is simply about words.  People react differently to the same concept expressed with different words.  You can guess that a death tax is less popular than an estate tax, even if it’s the same thing.  You can guess that the reaction to Obamacare vs. the Affordable Care Act will be a function of Obama’s popularity ratings.  If you’re trying to rehab your industry’s reputation, gaming sounds a heck of a lot better than gambling.

How can we apply these framing lessons to Silicon Valley sales and marketing?  Let’s provide several examples:

  • “This is not about picking the best product today, it’s about picking the best vendor with whom to partner over the long-term.”  Reframes vendor as partner and reframes technological advantage as fleeting.  Used frequently by market leaders to dismiss startups.
  • “This is not about which vendor has feature X, it’s about which system delivers the best overall performance.”  Moves attention from a feature that you lack that is supposed to improve overall performance, and back onto overall performance.  The inverse also works when you have a differentiating feature.
  • “This is not just about compliance, it’s about security.”  Reframing that properly separates compliance from security.  You can comply with lots of standards and still have weak security.  People want both.
  • “While I know you were initially shopping for a financial planning system, don’t you want to integrate your sales plan with your financial plan, and thus shouldn’t you be looking for a system that does both?”  Reframing that moves the goal post.  If your competitor only offers financial planning and you win this argument, you win the deal.
  • “If you want data governance to be effective, you should not tell the user to go the data governance system, you should bring data governance to the point of user access.”  Reframes separate data governance systems as undesirable and frames integrated access and discovery as more desirable and more effective.
  • “The question isn’t the price of the yearly subscription, but the total cost of ownership (TCO) and the total return on investment (ROI)?”  Reframes from looking at subscription price to TCO and adds a focus on ROI.
  • “It’s not about features XYZ.  If you’re looking to solve problem A, then you need to be looking at features PDQ and let me tell you why.”  Reframes the product selection criteria around a specific problem and the feature requirements for solving it.

These examples are all issue-level framing.  Narrative-level reframing is usually used when it comes to corporate messaging around innovation (e.g., “GoodCo is once again setting the bar as part of our ongoing technology leadership”), ongoing disputes (e.g., “another example of BadCo making poor imitations of our products”), or rivalries (e.g., “this market continues to be a two-horse race”).

Linguistic framing examples are harder to find in Silicon Valley marketing.  I’ll mull on this more and share some if I find them [9].

The Power of Consistency

Let’s wrap up by discussing consistency [10].  Consistency matters across three dimensions:

  • Language.  We need to consistently use the same words to describe things (e.g., consistently say “witch hunt” and not use synonyms like “fishing expedition”).
  • Spokesperson.  Each spokesperson needs to communicate the same messages (e.g., if we send five spokespeople to cover the Sunday morning talk shows, they all need to communicate the same talking points).
  • Time.  You must stick with the same message over long periods of time.  You can’t get bored with your message before the entire audience has heard it — and heard it several times.  David Ogilvy reminds us: “you aren’t advertising to a standing army, you are advertising to a moving parade.” I think politicians are quicker to understand this than marketers [11], hence the notion of stump speech

In Silicon Valley, we are not very good at consistency:

  • Companies tend to change their messaging every 18 months.  This is a by-product of changing CMOs at the same rate.  There are two ways to fix this:  reduce turnover in the CMO position or challenge the need to change messaging with the arrival of each new CMO.  To me, it’s a huge red flag when a new CMO wants to rebrand simply to put their mark on the company.
  • Companies get bored with their messages before the customers do.  Example:  Tableau has been talking about building data culture for over a decade.  Chief data officers (CDO) list building data culture as a top-three priority.  Instead of seeing this as a long-unfulfilled need, some marketers will see data culture as “tired” and want to talk about something else.  That’s a mistake.  You should not get bored with your message before your customers do.  Ivory soap has been “99 and 44/100ths percent pure” since 1882[12]
  • Companies confuse strategic and tactical messaging.  The company’s message shouldn’t be the latest product launch or marketing campaign.  Those can and often should dominate the hero on your homepage.  But your company’s message should be on the about-us page, start with your origin story, and change little over time.  The easiest way to ensure consistency is to stratify your message with some parts changing fairly frequently and others not changing at all.
  • Companies are terrible with synonyms and naming.  Most startups have about 3-5 names for roughly the same thing. These pseudo-synonyms are often loosely defined and used interchangeably when they shouldn’t be.  For example, ask your EPM seller the difference between planning, budgeting, and modeling.  Or ask your DI vendor about the names and types of metadata.  Product marketers need to get control over this by draining the language swamp, defining terms, and training the company to repeat the standard terms in the standard way.  If this seems like too small a battle, go inspire yourself by listening to some recordings of sales calls. You’ll be starting a glossary by lunch.

In this post, I’ve discussed three important communications principles that I think politicians execute well, shown how you can see them at work every day if you’re looking, and demonstrated how to apply them to the world of Silicon Valley marketing and communications.

Those principles are:

  • Demonstrate an understanding of the problem.   Don’t skip over this critical step in your rush to offer technical proof.
  • Framing is everything.  You can win deals by changing the customer’s view on what they should be buying.
  • The power of consistency.  Repetition works.  Pick a standard set of messages and words, train your team on them, and enforce standard usage.  Use this maxim to help: it’s better to be consistent than better.

Peace out.

# # #

Dedication

While researching this post, I was saddened to learn of the passing of Alan Kelly, the only PR titan I know who, after crushing it in Silicon Valley (e.g., by putting Oracle on the map in the 1990s), decided to challenge himself, move to DC, and bring both his firm and his communications system to the major leagues. Ave atque vale.

Notes

[1] For my European friends, this is the soccer equivalent of premier league vs. championship in England or Ligue 1 vs. Ligue 2 in France

[2] Which I mistakenly took as the AA baseball motto, but in fact it’s the motto of the American Association of Professional Baseball, an independent professional baseball league.

[3] Exceptions:  Communications Lessons from Mayor Pete which I wrote after watching him do a town hall or The Introvert’s Guide to Glad-Handing, inspired by watching Jackie Speier work a room.

[4] Right down to its subtitle:  It’s Not What You Say, It’s What People Hear.

[5] Which brings to mind the old Will Rogers quip: “I am not a member of any organized political party.  I am a democrat.”

[6] I’m also not going to drill into the extent to which these claims are supported.

[7] The fact that I knew exactly what to write in the first narrative and had to struggle coming up with second is a testament to my beliefs about execution, consistency, and the quip in note [5].  See this Tweet for a reference on the quote.

[8] I know entitlements is broader definitionally so they’re not really equivalent. However, note that social security and medicare/caid constitute the vast majority of entitlement spending.

[9] Every example I’ve thought of ends up actually being issue-level reframing.  I think it’s relatively rare when our arguments depend solely on the words chosen for expressing the exact same concept.  This can happen with feature naming and branding, but that’s not really the same thing.

[10] I call consistency one of the “three Cs” of communications:  clear, credible, and consistent.

[11] Put differently, do you ever wonder if Trump gets tired of saying “hoax” or “witch hunt”?  Using synonyms would be more refreshing for him.  But to drive the message home, it’s better to consistently repeat the same words. 

[12] Even inspiring the country song Pure Love with the chorus, “ninety-nine and forty-four one-hundredths percent pure love”.

PLG Resources and Wrap-Up

I put blog ideas into a to-write folder which contains more than 300 files full of brain dumps, outlines, and rants, all in various forms of disrepair. It’s my process. In that folder, there are 21 posts with PLG (product-led growth) in the draft copy and 5 with PLG in the title. It’s a topic I’ve always been interested in, but two things have prevented me from attempting a seminal post on PLG.

  • There is so much great work out there already. Every time I start to write, I hear stand on the shoulders of giants ring through my head followed by, “Dave, stop. Stop, will you? Stop Dave.”
  • I still consider myself more student than master. I grew up in enterprise. I have experience with market-seeding strategies and open source. I work with some velocity SaaS businesses. I’ve debated founders and boards on how enterprise companies should think about a PLG motion. But I’ve never run a PLG company and I’ve never worked in-depth on a PLG process.

Thus the drafts remain unfinished. But after having learned a fair bit, met many interesting people, located some great resources, and developed some strong opinions, I didn’t want to just drop everything. So I decided to write this summary report to provide links to PLG resources and share a few PLG thoughts before moving on.

PLG Resources

Thoughts and Opinions

Here are some opinions forged by my PLG research and conversations.

  • Don’t do PLG just because a board member wants you to. Try introducing a PLG product or motion if and only if you (as CEO) want to. While we are past peak-hype on PLG, there can still be a lot of pressure to try it because everyone else is doing it. Resist.
  • Try PLG if you think it can help your business. There is a lot to learn from PLG companies so I don’t recommend resistance for its own sake. But, if you weren’t born PLG, the odds your enterprise product will be ready overnight for a PLG motion are low, so keep expectations in check. Here’s a great post to help you get started on the journey. Alternatively, you could run PLG motion on a teaser product to pull people into your overall offering (e.g., Clearbit’s de-anonymization can lead to a later purchase of enrichment, Moz’s online presence can lead to a later purchase of its SEO).
  • I like growth teams. In general, I like silo-busting where you take groups of experts in different functions and aim them at a business goal. ABM does this, uniting sales, marketing, and SDRs in the quest to crack target accounts. Pods do this, uniting different sales and marketing functions, typically within a geography or vertical. Growth teams do this, uniting product managers, designers, marketers, sellers, and analytics team members on maximizing PLG conversion rates. They’re a good idea.
  • The product doesn’t sell itself. Knowing how expensive sales and marketing people can be — with S&M expenses typically running at twice R&D in a typical software company — I think for some people the PLG dream was to elminate S&M with a product that sold itself. That didn’t happen. In PLG, most of the time, the product helps sell itself. That’s certainly a lot better than not helping (I’ve seen that too), but we’ll still need those pesky sellers and marketers after all. See the McKinsey paper.
  • PLG ain’t cheaper. While a few PLG companies end up ahead on the S&M vs. R&D expense trade-off, most end up spending more on both. See the Tunguz post or the McKinsey paper. Don’t do PLG to save money. Do it because you think you’ll have a better product, grow faster, take market share, and build leadership. PLG is not a cost-cutting strategy.
  • PLG still needs marketing. How do you think we get people to do all those trials anyway? Yes, you may have a viral product or a strong community (word of mouth), but you’ll still also rely on marketing to drive people to your trial via SEO, SEM, and making TRY-IT the primary call-to-action on your website. (This always makes me think of Boston with the anaphoric “listen to the record” on the back cover of their debut album.)
  • If you’re starting a new company, try to be born PLG. For example, were I to start a new EPM company, I’d try hard to build a PLG motion in from day one. That’s not particularly easy in a space with separate buyers and end-users (where end-users need to be connected to the corporate system), but I’d sure try. You’ll likely end up with a better product as a result. Or a teaser product linked to a core one.
  • PLG is another pipeline source. Traditionally we’ve had four pipeline sources (marketing/inbound, partners, sales/outbound, SDR/outbound). When do you PLG, you’re not only adding any direct revenue, you’re also adding a pipeline source (with its own vernacular, e.g., PQLs) much as ABM adds a pipeline source (with its own vernacular, e.g., MQAs).

Appearance on Data Radicals: Frameworks and the Art of Simplification

This is a quick post to highlight my recent appearance on the Data Radicals podcast (Apple, Spotify), hosted by Alation founder and CEO, Satyen Sangani. I’ve worked with Alation for a long time in varied capacities — e.g., as an angel investor, advisor, director, interim executive, skit writer, and probably a few other ways I can’t remember. This is a company I know well. They’re in a space I’m passionate about — and one that I might argue is a logical second generation of the semantic-layer-based BI market where I spent nearly ten years as CMO of Business Objects.

Satyen is a founder for whom I have a ton of respect, not only because of what he’s created, but because of the emphasis on culture and values reflected in how did it. Satyen also appreciates a good intellectual sparring match when making big decisions — something many founders pretend to enjoy, few actually do, and fewer still seek out.

This is an episode like no other I’ve done because of that history and because of the selection of topics that Satyen chose to cover as a result. This is not your standard Kellblog “do CAC on a cash basis,” “use pipeline expected value as a triangulation forecast,” or “align marketing with sales” podcast episode. Make no mistake, I love those too — but this is just noteably different content from most of my other appearances.

Here, we talk about:

  • The history and evolution of the database and tools market
  • The modern data stack
  • Intelligent operational applications vs. analytic applications
  • Why I feel that data can often end up an abstraction contest (and what to do about that)
  • Why I think in confusing makets that the best mapmaker wins
  • Who benefits from confusion in markets — and who doesn’t
  • Frameworks, simplification, and reductionism
  • Strategy and distilling the essence of a problem
  • Layering marketing messaging using ternary trees
  • The people who most influenced my thinking and career
  • The evolution of the data intelligence category and its roots in data governance and data catalogs
  • How tech markets are like boxing matches — you win a round and your prize is to earn the chance to fight in the next one
  • Data culture as an ultimate benefit and data intelligence as a software category

I hope you can listen to the episode, also available on Apple podcasts and Spotify. Thanks to Satyen for having me and I wish Alation continuing fair winds and following seas.

The CEO Job Description: In Reductionist Form

Long-time readers will know I like making reductionist mission statements for roles in the organization. For example:

Some people take issue with these:

  • “That defines marketing into a purely tactical role!” No, you did that — not me. You can make sales easier in very strategic ways, like picking great target markets and working with product to uniquely meet customer requirements.
  • “Shouldn’t HR be about employee experience?” Yes, but indirectly. We don’t need HR to be gossips, gadflies, or self-appointed employee advocates. If HR focuses on making managers more effective almost everything else falls into place. No more blindside-hit terminations. No avoided conversations on performance. Clear goals. Legal compliance. Recruiting support. It’s not the “HR police” fighting against evil managers and protecting employees from them. It’s HR supporting managers to do their jobs better.
  • “Shouldn’t customer success be about relationships?” Well, if you want to get someone’s renewal it sure helps if you have a relationship with them, understand the value they’re getting value from the software, and can be relied upon to ensure their issues are resolved. So yes, it’s about relationships — but as a means to an end. And that end is a renewal. Or the identification of an upsell opportunity.

So, all that said, what’s the reductionist mission statement for the CEO?

The CEO’s job is to get what matters right.

Let me explain this with a story. Before joining Business Objects, I spent 10 years across two startups: Ingres, which got destroyed by Oracle (while nevertheless growing from $30M to $240M during my tenure) and exited for less than 1x revenue and Versant, which we turned around despite existing within a doomed category (object database) and went on to a lackluster IPO.

Let’s just say I saw plenty of problems along the way. At Ingres, I think I had 9 bosses in 7 years. We were so desperate for marketing talent that at one point we hired the former brand manager for Chuck Wagon dog food. We got acquired in an SGI/MIPS-like OEM-buys-technology-provider deal by ASK, the leader in what was then called MRP (the predecessor of ERP). The company that could have been Oracle got acquired by the company that should have been SAP. And then we pulled them down with us in a kind of corporate double-drowning, a Vietnam-esque escalation of commitment on ASK’s part. I was on the seven-member merger integration team, hand-picked by Sandy Kurtzig. I saw all kinds of shit go down.

At Versant, my boss, a high-flying ex-Lotus executive, told me one day, “if this can’t be a $1B company, I don’t want to be a part of it.” Well, I guess he figured it out because one day he — and the rest of the executive team — all vanished and I got a battlefield promotion to marketing VP. While we did turn the company around with a chasm-crossing strategy under the capable Dave Banks, we skimmed the treetops, a few times putting shares in employee compensation envelopes because we didn’t have the cash. (I’m sure this would be illegal today — who knows, maybe it was back then.)

I subsequently joined Business Objects, not long after its IPO. So I had access to a fresh S-1 for diligence. This thing was perfect. Twenty-something quarters of profitable growth. A total of $4M in VC burned from inception to IPO. A growth curve to die for: 1, 5, 15, 30 on revenues. The entire S-1 was as pristine as the Arctic National Wildlife Refuge.

After 10 years of struggles, I was finally going to work at the perfect company. I was so excited. And so naive.

I expected Business Objects to get everything right. It didn’t. I quickly learned lots of things were wrong. But several things weren’t. We had an aggressive, sales-driven culture. We understood how to manage Wall Street. We anticipated trends in the market (e.g., BI consolidation) and got ahead of them. We had strong operational discipline. And dare I say, good marketing.

But wow, did we get things wrong, too. The version 4 product launch darn near killed us. We were slow to the web (but had a great strategy once we went there). Germany was a repeated mess. The US operation continuously struggled. Our inability to build enterprise reporting (via a project ironically code-named after a tragic opera) nearly cost us the market. “This place is as screwed up as Ingres or Versant,” I’d often think.

But it wasn’t. That’s when I first realized that success in business is not about getting everything right. It’s about getting what matters right. And Business Objects did.

This, of course, begs the critically important question: what matters?

For Business Objects, in the end, what mattered was three things:

  • The consolidation of the BI category from separate Q&R, OLAP, and enterprise reporting into a BI suite.
  • The transition to the web (the product started out as 16-bit PC software).
  • Market leadership, mostly in the form of market share but also in market vision.

You’ll notice that I just took nine years of my life in a complex, evolving market and boiled success down to three things. That’s easy to do in hindsight. It’s nearly impossible when you’re in the thick of it.

I’ll share a trick. Before starting any strategy meeting, ask a few people during the coffee: “hey, you were at BigCo, and you folks really succeeded, why was that?” or “hey, you were at LittleCo and that didn’t work out so well, why was that?” Anyone with a strategic bone in their body can provide a pretty good, short answer. They’ve effectively just told you, “what mattered.”

Now do it in the present. “So what matters here at OurCo, today?” You’ll almost always get a laundry list of both strategic and tactical concerns. For example, only slightly dramatized:

  • Artificial intelligence strategy
  • Performance appraisal process
  • Company culture
  • Sales productivity
  • Pipeline diversification
  • Channel expansion
  • Add-on products
  • Customer satisfaction
  • Category creation
  • International operations
  • Long-term financial model
  • QBR format and cadence
  • Organizational structure
  • Glassdoor reviews
  • Account-based marketing
  • Marketing attribution
  • Lawsuits
  • Competitor features
  • G2 reviews
  • Product-led growth
  • Indirect competitors
  • Vision statement
  • Board meeting effectiveness
  • Customer advisory board
  • Annual user conference
  • Usage-based pricing
  • Public cloud strategy
  • Data privacy and compliance

I don’t entirely know why, but people are very good at distiling the essence of the past and just plain rotten at distillation of the present. Thus, that’s the CEO’s job. The very difficult task of figuring out what matters. And then getting that right. And doing their best to ignore or delegate almost everything else.

Remember my favorite passage on strategy, from Good Strategy, Bad Strategy by Richard Rumelt.

After my colleague John Mamer stepped down as dean of the UCLA Anderson School of Management, he wanted to take a stab at teaching strategy. To acquaint himself with the subject, he sat in on ten of my class sessions. Somewhere around class number seven we were chatting about pedagogy and I noted that many of the lessons learned in a strategy course come in the form of the questions asked as study assignments and asked in class. These questions distill decades of experience about useful things to think about in exploring complex situations. John gave me a sidelong look and said, “It looks to me as if there is really only one question you are asking in each case. That question is what’s going on here?”  John’s comment was something I had never heard said explicitly, but it was instantly and obviously correct. A great deal of strategy work is trying to figure out what is going on. Not just deciding what to do, but the more fundamental problem of comprehending the situation. (Bolding mine.)

For advice on how to figure out what matters at your company, read Rumelt’s sequel, The Crux, which will help you identify and make plans for overcoming your biggest strategic challenge.

I was a CEO of two startups for about 12 years in total. I’ve worked as a CEO direct report for 12 years. I’ve sat on 10 boards. The CEO’s desire to run a tight ship is real, particularly for proud perfectionists. The forces pulling CEOs towards everything on the above list are also real. But I view that list as the occupational hazard of the CEO job, not the job description.

The job is to get what matters right. The difficult first step is to figure out what matters. The second step is to focus most of your energy on that. The third step, perhaps the most difficult of all, is to ignore or delegate almost everything else.

What Do “Pipeline Coverage” and “Forecast” Mean When Your Sales Cycle is 30 Days?

I grew up in enterprise.  I have already written a post on the tricky problem of mapping one’s mindset from enterprise to velocity SaaS, meaning smaller deals, shorter contract durations (e.g., month-to-month), and/or monthly-varying pricing [1].  That post was focused on what, if anything, “annual recurring revenue” (ARR) means such an environment, and how that impacts metrics that rely on ARR as part of their definition (e.g., CAC ratio).

In this post, I’ll continue in the velocity SaaS direction by exploring short average sales cycles (ASC), as opposed to short contracts.  Specifically, what does it mean in short ASC companies when you discuss common concepts like pipeline coverage and the sales forecast?

Let’s demonstrate the problem.

In enterprise, quarterly pipeline (defined as the sum of the values of opportunities with a close date in the quarter) is somewhat intertwined the notion of long sales cycles.  Meaning that in a company with 9–12-month sales cycles, virtually every deal that has a chance of closing within the quarter is already in the pipeline at the start of the quarter.  Thus, you can meaningfully calculate “coverage” for the quarter by dividing the quarterly starting pipeline by the quarterly sales target.  Most sales VPs like a 3x ratio [2].

Thus, the concept of pipeline coverage implicitly assumes a sales cycle (significantly) longer than the coverage period.  That’s why most companies don’t look at out-quarter pipeline coverage much (though they should) and if they do, they expect a much lower coverage ratio.

Now, let’s imagine an average sales cycle of 30 days and — rather than futzing with cohorts, statistics, and distributions [3] — let’s assume that all oppties are won or lost in exactly 30 days [4].

In this scenario, at the start of the quarter, what is the pipeline coverage ratio? It’s 1.0x.  Why?  We have zero pipeline for months 2 and 3 of the quarter.  If we assume that we have 3.0x coverage for month one and that the quarterly goal is evenly distributed across months, then we’d have 3.0x, 0.0x, and 0.0x for the three months of the quarter, or 1.0x overall [5].

In this example, quarterly pipeline coverage is basically meaningless because two-thirds of the pipeline you need to close during the quarter hasn’t been created yet.  Assuming a 30-day MQL-to-opportunity lag, one-third is working its way through the high funnel and the other third is still a wink in marketing’s eye.

If quarterly pipeline coverage is basically meaningless in short ASC companies, then what is meaningful?

  • Examining monthly pipeline coverage. Instead of week-3 quarterly pipeline coverage [6], we should look at day-3 monthly pipeline coverage — dividing the starting monthly pipeline by the monthly sales target. (After that, you can use to-go pipeline coverage to get continuous insight.)
  • Treating months 2 and 3 the way you’d treat next-quarter and the quarter thereafter in enterprise. Using a pipeline progression chart to see how the out-month pipeline is shaping up.
  • Getting marketing to forecast starting pipeline for month 2 and month 3, based on what they have already generated in the high funnel and their current pipeline generation plans for month 2.

Inherent in my point of view is that the definition of “coverage” is based on opportunities that already exist in the pipeline. Call me untrusting, but somehow I can’t feel covered by something that hasn’t been created yet.  Some might define quarterly coverage in this environment using month 1 pipeline plus month 2 pipeline forecast and month 3 pipeline plan.  But to me, that’s not coverage.  And it’s objectively not the same thing as pipeline coverage when we use the term in enterprise.

Now, let’s zip back to reality for a minute.  In the velocity companies that I work with, ASC is closer to 60 days and with a pretty broad distribution where maybe 90% of the deals close within 30 and 120 days.  Happily, this means you will have month 2 and month 3 opportunities in the starting quarter pipeline, but it nevertheless also means you will be increasingly reliant on to-be-generated opportunities across the months of the quarter.

In this case, I would make a three-layer forecast:

  • Sales (from existing opportunities). Forecast month 1, 2, and 3 sales using the normal sales forecasting process.
  • Marketing, from the high funnel. Use existing MQLs and your standard conversion rates, ideally time-based time-based (not just the total rate, but the rate split by time period)
  • Marketing, from planned demandgen. Forecast responses, then use standard conversion rates and ideally time-based. (Ideally you can start with your inverted funnel model.)

This approach is preferable to looking only at pipeline generation (pipegen) because a pipegen approach:

  • Tends to ignore the oppties that are already there
  • Almost always ignores that time-based nature of close rates
  • Uses an average sales price (ASP) as the proxy value for an opportunity [7].

In the example above you can clearly see how much of the forecast comes from existing opportunities (51%), how much from the existing high funnel (36%), and how much from planned demandgen activities (13%).

Finally, I have the same problem with the word “forecast” as I do with “coverage” in the short ASC world. They’re not quite the same thing as they are in enteprise. First, let me define “forecast,” along with its cousins, “plan” and “model.”

  • The plan is about accountability. It’s what we signed up for and accountable to. Budget is a synonym [8].
  • The model is a driver-based model of the business. It’s a calculated output (e.g., opportunities generated) given assumptions for a number of inputs and the way they interact (e.g., demandgen spend, MQLs generated, conversion rates).
  • The forecast is about prediction. It’s someone’s latest prediction for an output (e.g., bookings) given all available information at the time it’s made.

The plan is what we were willing to sign up for last December (when we received board approval). The forecast is what we think is going to happen now.  We used models to help build the original plan and we can certainly re-run those models today using actuals as inputs to see what they produce.

In enterprise, the sales forecast is all about the deals in play.  What if Mike closes deals A, B, and either C or D.  The buyer at deal E promised me they’d give us the order.  Given everything we know about Sally’s deal F, what value do we think it will close at?  Sales VPs spend hours in Excel (or a modern forecasting tool like Clari) running scenarios to arrive a number.  It’s usually more about different combinations of deals than it is about probabilities and expected values.

In the velocity world, as discussed above, the forecast cannot be only about existing deals. If you want to forecast a quarter, you’ll need to include results from the high-funnel and planned demangen. I’d still call it a forecast, but I’d know that it’s not quite the same thing as a forecast in enterprise. And by presenting in the three layers above, you can remind everyone of that.

# # #

Notes

[1] Monthly-varying SaaS is a different concept, which I used in that post, featuring short contracts (e.g., month-to-month) where the spend can vary every month, usually as the result of a flexible user-based pricing model, a consumption-based pricing model, or a hybrid pricing model (e.g., base + overage).  In such environments, simple SaaS concepts like ARR can quickly lose meaning, as do the metrics that rely on them (e.g., CAC ratio).

[2] Which I think had its ancient origins in the idea that you win 33%, lose 33%, and 33% slip. (Thus assuming a 50% competitive win rate.) Regardless of its roots, 3x (starting) coverage is a widely accepted norm, so much so that I fear it’s often a self-fufilling prophecy.

[3] We’re ignoring the distribution of average sales cycle length for closed/won deals, its standard deviation, and the fact the three different outcomes (i.e., win, loss, slip) will likely have three different average opportunity cycle lengths (e.g., you usually lose faster than you win), each with its own distribution.

[4] And, most unrealistically, that deals never slip to a subsequent period. We’re also assuming that all opportunities are generated on the first day of month, an exactly 30-day lag from MQL to opportunity, and that all MQLs are generated on the first day of month, and convert in exactly 30 days. (And, for the detail-oriented, that every month is 30 days.) Overall, with these simplifying assumptions, you start every month with only the opportunities generated from MQLs generated the prior month and only those opportunities. There is no leftover pipeline sloshing around to confuse things.

[5] The reality is likely somewhat less than 1.0x because we’d normally expected to some backloading (“linearity”) of the quarterly target across the months of the quarter.  In enterprise, that backloading is severe (e.g., most enterprise cash models assume a 10/20/70 distribution). In velocity SaaS, I’ve seen from 30/30/40 (i.e., pretty flat) to 10/20/70 (i.e., as backloaded as enterprise), typically reflecting a quarterly (as opposed to a monthly) sales cadence which is usually a mistake in a velocity model.

[6] To intelligently compare pipeline across quarters we need to fix a point in time to snapshot it. In enterprise, I prefer day one of week three because it’s early enough to take actions (e.g., reducing expenses), but late enough so sales can no longer credibly claim they need more time for pipeline cleanup (aka, scrubbing).

[7] In enterprise, this is a major sin because deal sizes vary significantly and values should be inserted only after discovery and price-point socialization (e.g., “you do know that this costs $150K?”)  In velocity, it’s a lesser sin because the deal sizes tend to be more similar.  Either way, if all we’re doing is counting opportunities and multiplying by a constant, then why not just admit it and count opportunities directly? The more sophisticated the proxy, the more I like it (e.g., using $10K for SMB, $25K for MM, and $75K for ENT).

[8] Technically, I’d say budget is a synonym for the financial part of the plan. That is, a budget is only one part of a plan. A plan would also include strategic goals, objectives for attaining them, and organization structure.