Category Archives: Startups

Fly the Aircraft First: The Potentially Paralyzing Effects of Fundraising

Startup CEOs can learn an important lesson from pilots.  Specifically, to always fly the aircraft first.  Sounds obvious, like maybe you shouldn’t need to remind pilots to do this, but here’s what they teach them and why:

From the earliest days of flight training, pilots are taught an important set of priorities that should follow them through their entire flying career: Aviate, Navigate, and Communicate.  The top priority — always — is to aviate. That means fly the airplane by using the flight controls and flight instruments to direct the airplane’s attitude, airspeed, and altitude …

A famous example of a failure to aviate is the December 1972 crash of Flight 401, an Eastern Airlines Lockheed L-1011. The entire crew was single-mindedly focused on the malfunction of a landing gear position indicator light. No one was left to keep the plane in the air, as it headed towards a shallow descent into the Florida Everglades. Four professional aviators … were so focused on a non-critical task that they failed to detect and arrest the descent.

In my work with startups, I periodically see CEOs surprisingly stop flying the aircraft first.  When does that happen?  When they are raising money, or think that they might be soon.  I know they’re not flying the aircraft first because they say things like:

  • “I want to replace the CRO, but I can’t because I’ll be out fundraising next quarter.”
  • “We need to reduce the burn rate because cash-out is about 9 months away, but I don’t want to cut expenses now because I’m trying to raise money.”
  • “I’m no longer excited about the new product that we’re building, but I want to keep funding it because we’re out raising money and VCs like it as part of the pitch.”

Not flying the aircraft first means not making operational changes that you normally would because you are fundraising, or believe you soon will be.  It means running your business differently because you are trying to raise money.

This begs two questions:

  • Are you actually fundraising or just talking to venture capitalists? There is a difference.
  • Even if you actually are fundraising, is deferring such changes a good idea?

Are You Actually Out Raising Money?

Sometimes you want to keep the burn rate high while fundraising to stay on a hypergrowth trajectory and enable the big, next round.  Other times, you’re growing at 30%, and not particularly efficiently, and that next round is more fantasy than reality. Happy ears can help you avoid unpleasant-but-necessary decisions for another few weeks or months.

Are you out raising money or are you simply talking to VCs? How can you tell the difference?

  • By receiving a term sheet.  VCs don’t need your permission to make you an offer, though such proactive term sheets are less common than they used to be.  Remember that if all that love is real, there’s an easy way for a VC to show it.
  • By asking.  Remember the first rule of VC (and M&A):  the amount of time you invest, access you offer, and data you provide must always be proportional to the odds you see of actually closing a deal.  Ask if they’re thinking of making an offer, why or why not, and when.  Ask what additional information they need and provide it only if they are clearly doing their homework, signal an acceptable valuation range, and express valid concerns that you can resolve.

If you’re not getting term sheets and are starting to doubt some of the answers you’re hearing, then look for these clues that you’re more talking to VCs than raising money:

  • You have been talking with an analyst or associate for 2+ meetings without talking to a principal or partner.
  • Your meetings get rescheduled and responses to your communications come slowly. That likely means the VC doesn’t see your deal as urgent and probably thinks of your interactions more as a simple chat or check-in.
  • The VC doesn’t appear to be doing their homework. They ask questions that are answered in the material you’ve already shared, they don’t communicate the due diligence agenda ahead of meetings, and they don’t follow-up on the data requests they’ve made. VCs do a lot of work preparing for their internal investment committee, and you can usually tell when they’re doing it.
  • The VC focuses only on financial metrics, which could indicate that they’re just updating their database or, worse yet, are looking at another player in the space and using you as a data point. [1]

But even if an investor is genuinely working on an offer, if that offer is not qualified on valuation or, increasingly in these days, terms, then once again you’re not out raising money, you’re just talking to VCs. On valuation, it’s pretty clear — if an investor says they’re working on a term sheet at a valuation of 4x revenues when your absolute minimum is 6x, then you’re not out raising money; you’re just talking to VCs.

With terms (also known as structure), things are somewhat more subtle. You can receive a term sheet with an attractive headline valuation only to discover it’s dirty because it contains terms such as:

  • A multiple liquidation preference, where the new investor gets not the usual 1x their money back before the common shareholders, but perhaps 1.5x or 2.0x.
  • Participating preferred stock, where the new class of preferred stock gets both its liquidation preference and its pro rata, as opposed to one or the other.
  • Redemption rights, where the company has the obligation to repurchase the shares from the new investor after some time period (e.g., five years). (Which begs the question of where it’s going to get the money to do so?)

For more information on dirty term sheets, see this excellent post by Janelle Teng. If you’re in receipt of one, I recommend doing these three things:

  • Do the work to understand the terms. Ask your CFO, lawyer, or banker to create liquidation waterfalls to model the outcomes in numerous liquidation scenarios.
  • Ask the investor if they would provide an equivalent offer on clean terms [2].  Understand and compare a possible down-round to a flat- or up-round on less desirable terms.
  • Remember that terms only get worse.  No investor wants to invest on terms inferior to the prior investors.  This suggests that if you’re going to do a round with dirty terms that you make it big enough (and/or your path-to-profitability fast enough) to be pretty sure it’s your last [3].

But let’s zoom back up.  Why are we talking about dirty term sheets again? 

Just as spending time with VCs who will eventually make you an offer at an unacceptable valuation is not actually “out fundraising,” so is spending time with VCs who will eventually make you an offer on unacceptable terms. It’s more subtle, but it’s the same issue. And to defer necessary operational changes because of it is a big mistake.

Is Deferring Change a Good Idea?

Let’s say that you are actually out raising money — not just talking to VCs, but talking to VCs who are likely to make you an offer at an acceptable valuation and on acceptable terms.

In this case, is not flying the aircraft first a good idea? I think this is a hard question because of the risk of derailing a potentially transformational financing round. 

Even here, I think it’s wrong to defer the changes [4].  Why?  Because it’s:

  • Dishonest.  You shouldn’t start your long-term relationship with a new investor by saying, “just kidding when I said the CRO was great, we need to replace them.” [5] You might be working together for the next 5-10 years.
  • Ineffective.  It has spectacular backfire potential. Specifically, the investor is likely to detect the problem with the CRO and your vigorous (but disingenuous) defense of the CRO may cause them to question you and drop out the round as a result. [6]
  • Bad for the company.  Failing to make a desired operational change is definitionally bad for the business.

As one VC said to me, “We don’t invest in perfect companies. We invest in companies where the upside is greater than the downside. I have invested in companies where the CRO, CTO, and CFO were all recently or in the midst of transition. The important thing is that we talk about the changes and understand them. We’re going to be investors for the next 5-10 years.”

What does that mean to me? If you, as CEO, think something needs to be done, then do it. Always fly the aircraft first.

# # #


Thanks to Michael Lavner for his comments and review.

[1] This happened to me. If the VC had just told me they were doing diligence in the space on another potential investment (as opposed to seeming to express genuine interest in my company), then I wouldn’t have felt burned. Plus, a note to VCs: if you meet me, ask me all about the space, and then announce a deal with one of my competitors in about 2 weeks, I’m going to know what happened and feel used in the process.

[2] I love this one because they do the work for you and show you an offer that is mathematically equivalent (to them). Comparing the two sheets (and associated liquidation waterfalls) shows you the cost of maintaining the headline valuation and avoiding the consequences of a down-round.

[3] To be clear, some investors will be scared off by finding a lot of structure in previous rounds. So it’s not simply a question that you will have to raise subsequent rounds on equal or inferior terms. You may not be able to raise at all, or at least from the investors who you want to raise from.

[4] Though, perhaps sadly, it takes me longer to reach that conclusion. While I “get” the theory that I’m preaching, I’ve also raised money on the back of a CRO transition, and it wasn’t easy. Nevertheless, I still preach the theory.

[5] You easily could have said, “they’re nearing the end of their runway” during the process, instead.

[6] Or, more simply, they may just detect the deception.

Metrics That Matter in 2023: My KiwiSaaS Presentation Slides

Just a quick post to share the slides from the presentation I gave today at the KiwiSaaS conference to discuss the SaaS metrics that matter in 2023 and 2024.

The presentation has three sections:

First, an introduction which quickly reviews the ways the startup world has changed in the past 6 months.  Simply put:  Toto, I have a feeling we’re not in 2021 anymore.

Second, a three-step set of recommendations for what to do about that:  (1) extend your runway, (2) make a plan to re-earn your last round valuation, and (3) enable your next round, likely in 18-24 months, by focusing on the metrics that matter in this new world.

  • Phrasing these strategies in terms of songs/albums:  (1) Staying Alive, (2) Get Back [to where you once were valued], and (3) Born to Run [convince VCs that, “tramps like us, baby, we were born to run” — i.e., that we have a lean machine where ARR is a predictable output of VC investment.]
  • Note that I also did a Balderton webinar (Balancing Growth and Burn in 2023) on this topic with David Thevenon.

Third, a one-slide-per-metric review of the set of metrics that matter in 2023:  ARR growth, free cashflow margin, Rule of 40 score, subscription gross margin, burn multiple, ARR/FTE, CAC ratio, CAC payback period, NRR, and GRR.

  • This includes an explanation of why I excluded (what I view as old school) churn, lifetime value (LTV) and LTV/CAC analysis from those metrics.  That explanation is also available in considerably longer form in my SaaStr talk:  Churn is Dead, Long Live NDR.

The presentation is chock full of links to interesting articles (e.g., an amazing 75-page interview transcript with Sequoia founder Don Valentine as part of an oral history of Silicon Valley, a great breakdown on stock-based compensation by Janelle Teng) and it includes a slide on people to follow and sites to visit if you are interested in this material.  An image of it is pasted below, the presentation itself has live links.

The slides of the presentation are embedded below.

For those who don’t use Slideshare, the presentation is also available on Google Drive.

Thanks to KiwiSaaS for inviting me, the audience for putting up with a remote live presentation, and to the sources I included as data in the slides — particularly RevOps Squared, on whose 2022 SaaS Benchmarks survey I relied fairly heavily.

My Thoughts on the SVB Meltdown

(Revised 8:56 am 3/19)

Looks like I picked the wrong week to be off-grid in Argentina.

When I came back on-grid last night, I quickly discovered that the world, or more precisely, my Silicon Valley business world, had basically exploded while I was flyfishing in Patagonia.

A few weeks ago there had been talk of a mass extinction event for startups in 2023.  It was about funding, not banking, and the prediction was for the second half of 2023.  But perhaps it had come early and for a different reason.

Instead of writing yet-another explainer article, I’ll do two things:

  • Provide links to the best explainer articles I’ve found thus far
  • Share some of my own views on the situation, reminding readers that I am go-to-market person and former CEO (and not a finance person or former CFO)

The Best Explainer Posts I’ve Found

My Personal Views on the Situation

I’ll quickly share my personal views on the situation here:

  • Almost every company I work with uses SVB.  They are the default startup bank in Silicon Valley.  Many keep all their cash there because it’s a fairly standard term of an associated venture debt loan.  If depositors lose their funds I believe large numbers of startups could fail, eliminating the thousands of jobs that they provide.  The Alderaan scenario.  I think it’s unlikely, but absolutely must be avoided.
  • Startup death is a natural part of the Silicon Valley ecosystem, the Darwinian process that produces the innovation that drives a large part of our economy.  Startup death is a natural part of the process — but it should result from a bad idea or a unworkable product.  Not from your bank failing.
  • There is a blame game with three primary parties involved:  VCs for provoking the bank run, the Fed for raising rates (which devalued SVB’s long bonds), and SVB for putting themselves in an weak position.  Who you blame seems to say more about you than the situation.  People who like SVB blame the Fed.  People who dislike VCs blame them.
  • Answering the question “what happens to us if rates go up?” seems absolutely core to the operation of a bank.  (Think:  it’s what we do here.)  SVB put themselves into a situation where the liquidity rumors couldn’t be easily dismissed.  Yes, VCs likely provoked the bank run, but SVB put themselves in a place where they couldn’t stop it and bungled communications on top of that.
  • You cannot overstate the interconnectedness around SVB.  I know startups with all their money there.  I know VCs who are unable to provide bridge loans to startups because all their working capital is also at SVB.  I’ve heard of founder/CEOs who have all their personal money there as well, so they are unable to even use their own funds to bail out their companies.  The single worst story I’ve heard is a startup who had all their money in SVB successfully arranged a loan to cover payroll and wired that money to their payroll provider … who then put it in SVB.  Additionally, startups often sell to other startups, so the web is intereconnected not just across investors, but companies and customers.
  • SVB’s depositors must be protected.  I’m not talking about bailing out SVB investors or management.  I’m talking about protecting depositors, thousands of startups, the jobs they provide today, and their potential to become world-leading tech companies  — the next Oracle, Cisco, or Salesforce might be killed off if we don’t.

Personally, while I’m not an expert in banking, I am uncharacteristically optimistic because SVB owns plenty of high-quality assets and, as mentioned above, those assets exceed deposits in value (though that is a function of valuation method as discussed in the Rubinstein article).

They are not sitting atop a pile of incredibly complex, thinly-traded derivatives (e.g., CDOs, CDO swaps).  They are sitting atop a pile of long government bonds.   This is not 2008.  SVB is not Lehman Brothers.  Because of this, I think there is a good chance that someone acquires them this weekend (or soon thereafter), finding opportunity in SVB’s wreckage and ending this industry-wide liquidity crunch.

Let’s hope so, at least.

Is a Dream a Lie if It Don’t Come True? Founders, Aspirations, and Company Potential

“Is a dream a lie if don’t come true?” — Bruce Springsteen

The River was one of my favorite songs in college and whenever I listen to the above line near the end, I start thinking about Silicon Valley.

Consider three entrepreneurs.

Founder 1:  Elizabeth Holmes.  Was she simple con artist who chose fraud over business failure or a broken visionary trying to walk in the footsteps of Steve Jobs?

Founder 2:  Adam Neumann who dressed up the equivalent of Regus as a tech company and successfully raised money at valuations up to $47B before flaming out on the approach to an IPO.  (Now seemingly running a similar play with Flow.)

Founder 3:  Joe, our friend at BigCo who quit his VP-level job to found a company, spent 10 years sweating it out, pivoted, recapped, and finally threw in the towel for a carve-out in a $30M sale that didn’t clear the preference stack.

Which are they?  Were they dreamers or liars?

To try and sort that out, I’d consider three questions:

  • Did they truly believe in the dream?  It’s hard to know what anyone truly believes [1] — and we need to separate visionaries from lunatics [2] — but in many situations you can develop a sense for whether someone is a true believer or a poser.  This one’s hard to assess, but important.
  • Were they lying about progress?  While there is a small gray zone of exaggeration, misunderstanding, and embellishment [3], for the most part this one is black and white.  Were the numbers real?  Was the demo faked?  Were the milestones hit?
  • Were they making big money before realizing the dream?  This didn’t used to be possible in Silicon Valley, but a side effect of the recent financing environment [4] was the rise of secondary sales that made it possible for founders to reap 10s to 100s of millions before a liquidity event that shared success more broadly across investors and employees.  Situations where a founder can make “done” (or “lifestyle changing”) money before realizing the dream can present the potential for conflicts of interest.

We ask a lot from founders.  And what we ask is often in diametric oppposition.   We ask founders to be:

  • Unreasonable, but reasonable.  Founding a startup against long odds is an inherently unreasonable thing to do.  But, aside from that, we want them to be reasonable people.
  • Optimistic, but realistic.  We want them to believe they can accomplish the nearly impossible, but be realisitic in setting goals and operating plan targets.
  • Big-picture, but detail-oriented.  We want them to create a disruptive, big-picture vision of the market, but be able recite SaaS metrics from memory.

This alone is a good reason to have both co-founders and a strong executive team.  While F Scott Fitzgerald said, “the test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time,” it’s hard to do so in every situation, all the time.

While plenty has been written about Holmes and Neumann inspired a TV series, nobody talks much about Joe.   And there are a lot of Joe’s out there.

Joe’s employees and investors are disappointed in him and might be mad at him.  After all, Joe probably said:

  • The company faced an amazing opportunity
  • The space had the potential to produce a public company
  • He believed they could beat BadCo and WorseCo to win the market

But what did we want Joe to say?   Yes, Joe needs to be careful.  He needs to speak precisely and make precise claims.  He needs to hedge his language and not make promises.  But Joe is not only allowed to be optimistic, it’s his job.

As I’ve often said,

“As CEO, even if you’re standing neck-deep in shit, you need to be looking at the stars.”

Put differently, while the CEO needs to be aware of the company’s situation and have credible plans to address it (i.e., the neck-deep part), they must always be focused on and believe in the potential of the company [5].  The day they can’t do that is when they should turn in their badge.

So, going back to Springsteen, “is a dream a lie if it don’t come true — or is it something worse?”

# # #


[1]  I’d argue it’s sometimes easier to know when they don’t — e.g., if they’re telling all their friends they’re running a scam.

[2] For example, the difference between trying to emulate Steve Jobs and thinking that you have been sent by God as the reincarnation of Steve Jobs.

[3] Larry Ellison reportedly once said, “sometimes I just get my verb tenses mixed up” when speaking about product capabilities vs. roadmap.

[4] E.g., higher valuations, longer time to liquidity, higher bar on IPOs.

[5] My take on what’s known as the Stockdale paradox.

The Key to Making Market Research Actionable, Part II

In the first part of this two-part series we discussed the importance of timing in ensuring that market research is actionable.  The moral was to time the arrival of research (e.g., win/loss reporting, NPS surveys, awareness and marketing funnel studies) with the cadence of your company’s quarterly and annual strategic goal setting process.

Research that arrives asynchronously gets read (if you’re lucky) and then forgotten.  Research that arrives synchronously becomes a homework assignment for the meeting and a session on the agenda.  That way, its findings are top-of-mind when you sit down to decide priorities and hammer out OKRs.

In part II, we’ll take a more strategic look at the question.  Ultimately, to make market research actionable, you need to ensure five things:

  • Good timing.  It must show up at a time when you’re ready to absorb and action it.  The subject of the first post in this series.
  • High relevance.  It needs to help answer your most important strategic questions.
  • Action-oriented framing.  Work to ask questions in a way that provides action-oriented answers.  You can ask, “do you have plans to move to the cloud in the next five years?” or you can ask, “do you plan to move to the cloud in the next year, and if so, what are your top three evaluation criteria?”
  • Time for consensus building.  You can’t just spit out the answer from a black box.  At each stage of the process, you want to have discussion and get buy in, so that when the end is reached people feel the process was valid and buy into the conclusions.
  • A qualitative component.  Quantitative answers what, but not why.  Qualitative can lead to understanding why.  Pair surveys with interviews for this reason.

Put differently, as my friends at Topline Strategy say, market research that gets turned into action is market research that was designed from the outset to be actionable.

Let’s drill into relevance and action-orientation a bit more.  To ensure you’re asking relevant questions, you should do two things.

First, create what I call the hypothesis file.  This is a file where you write down, over the course of the year, every time you hear an assertion or a hypothesis that you’d love to validate.  Examples:

  • The problem is nobody’s ever heard of us.  We’re just not seeing enough deals.
  • The issue is we’re not making the short list and that because we’re not seen as a leader.
  • We can’t sell use-case A because we’re seen as weak on features 1, 2, and 3.
  • The leakage point in our funnel is demo.  We lose too many deals there and that’s because of our UI.
  • We’re not speaking to the business buyer’s priorities.
  • Everyone’s tired of talking about (e.g.,) data culture, we need a new message.
  • If we just focused on BigCo replacements, we could do the numbers on that alone.

These are rarely offered as hypotheses.  They’re usually statements, often presented as self-evident facts.  You need to tune your ears to hear them and write them down.  Don’t fight every one in real time.  But be keenly aware that these are the foundations of your internal corporate mythology — and it’d sure be nice to know if they’re true or not.

When it’s time to do a market study, review the questions in the file and decide which ones you want answered.  Picking the hottest questions will guarantee that people will be champing at the bit to see the results.

Second, to ensure high relevance, try to identify the core strategic questions you’re facing, whether they appear in the hypothesis file or not.  Such as:

  • In which segment are we really most successful, not just at winning deals but renewing and expanding them?
  • If our market is transitioning to a platform, what are the key elements that must be included?  Where do customers want us to partner so they can buy best of breed?
  • How can we easily regain some product differentiation that matters to customers with our limited R&D capacity?
  • Is our Microsoft partnership a key asset on which to double down or a liability to unwind?
  • Will our category be entirely absorbed into a broader suite or can we sustain a moat of product differentiation to protect us?

By debating these questions, and which ones to include in the study, you again guarantee a high degree of interest in learning the answers once they are available.

To ensure you’re asking action-oriented questions, as you build the study, question-by-question you need to ask yourself, “what would I do differently based on the answer?”

  • For multiple choice, what would I do differently if the majority answer were A vs. C?
  • For rankings, what would I do knowing the top three ranked choices were 123?  Or the bottom, 789?
  • For progression questions, what would I do if I notice everyone was dropping out at stage 3 of our funnel?

Sometimes, the answer is ask more questions.  So build follow-up and drill-down questions into the survey.  Sometimes, the answer is you’re circling the wrong question.  Think:  we asked a lot of questions that will help us determine the TAM.  Say we conclude it’s $20B, then what?  I think we’re asking the wrong question, I don’t want to know what the TAM is, I won’t to know the velocity with which it’s coming to market.  Ultimately, how many deals will be happening in the space next year and how many of those do we need to participate in to make our numbers?

Simply put, you can research how big the TAM is, or you can research how much of it is coming to market next year.  The latter is a lot more actionable than the former.

So let’s wrap up.  If we want to ensure that our market research is actionable, then we need to:

  • Time its arrival
  • Study the right questions
  • Ask those questions in a way that provides action-oriented answers