Chris Arsenault

Category: The Musings

Long-form writing on technology, leadership, AI strategy, and execution inside complex organizations.

  • The Hire I Got Right (and Why)

    Looking back: April 2026

    I’ve made a lot of hires in my career. I’ve gotten many of them wrong. I’ve gotten a few spectacularly right. The one I think about most isn’t the most senior hire I’ve ever made — it was an early, mid-level hire that worked in ways that surprised me and taught me what the hiring evaluation should actually look for.

    The Hire

    The role was for someone to own a specific function within a growing organization — a role that required both execution capability and the judgment to navigate ambiguity. The function didn’t have a playbook. The person would have to build one.

    The candidate was not the most senior applicant. Not the most credentialed. Not the most impressive on paper. They were competent, they had the relevant experience, but by traditional hiring metrics they were middle of the pack.

    What they had — which most of the other candidates didn’t — was a specific quality I now look for deliberately: the ability to diagnose the real problem before jumping to a solution.

    What the Interview Revealed

    In the interview, I presented a scenario. The function had three competing priorities, limited resources, and unclear direction. How would they approach it?

    Most candidates went straight to a solution. They’d tell me how they’d structure the team, which priorities they’d rank first, what metrics they’d track. The answers were fine but fungible — any of them could have been written in a how-to-manage book.

    This candidate did something different. They spent most of the interview asking questions. Why were those three priorities in competition? Had we tried to decouple them? What did leadership actually want this function to achieve? What would “working well” look like to the board? Who outside the function was affected by its output?

    By the end of the conversation, they’d mapped the real problem — which was not a resource allocation issue but a clarity issue at the leadership level. The function was in conflict because leadership hadn’t made the hard trade-off decisions, and no amount of internal prioritization would resolve that until the upstream clarity was fixed.

    The insight wasn’t groundbreaking. What was notable was that they’d gotten there through disciplined questioning rather than through delivering a prepared answer. They’d treated the scenario as a real problem to diagnose, not as an interview question to perform against.

    What Happened After They Started

    The function they took over was a mess, as I’d described. Within three months, they had done three things:

    1. Escalated the real problem to leadership clearly and got the upstream clarity that had been missing.
    2. Built a short-term operating cadence that let the function deliver against its most urgent priorities even while the broader issue was being resolved.
    3. Built a longer-term structure that would hold up as the function grew.

    None of this was visible externally for several weeks. They weren’t making big announcements. They were doing the quiet work of diagnosing and structuring. It looked, from the outside, like not much was happening.

    At month three, the function hit a tempo I hadn’t seen from that role before. At month six, the entire surrounding system was working better because the hire had fixed upstream problems that had been affecting multiple other functions.

    By month twelve, they were the quietest high performer on the team. They didn’t seek credit. They didn’t manage upward. They just made the function work, helped the rest of the organization work, and occasionally surfaced new issues that needed to be worked through.

    What I Learned From This Hire

    Several lessons that changed how I’ve hired since:

    1. The quality of the hire’s questions matters more than the quality of their answers.
    Interview questions produce rehearsed answers. Scenario walkthroughs where the candidate asks their questions reveal how they actually think. I now spend significant interview time watching candidates engage with real problems, not evaluating how polished their answers are.

    2. The best hires are often not the most impressive on paper.
    Credentials, prior titles, impressive-brand companies — these are proxies. Sometimes the proxies align with ability. Often they don’t. The candidate I hired had a fine but unremarkable resume. The judgment and questioning quality weren’t on the resume; they only showed up in conversation.

    3. Diagnosis > prescription, especially for senior roles.
    A hire who can correctly diagnose a problem will eventually find or build the right solution. A hire who can prescribe solutions without diagnosis will execute well on problems that happen to match their prescription, and poorly on everything else. The difference between the two compounds over years.

    4. The quiet high performers are disproportionately valuable.
    They don’t manage upward. They don’t generate visibility for their work. They can be underweighted in organizational politics. But they produce disproportionate value because the rest of the system depends on them in ways that aren’t legible. Finding them, valuing them, and protecting them is one of the highest-leverage things an executive does.

    How I Hire Now

    Because of this hire, my hiring process evolved to include:

    • Extended scenario-based conversations where the candidate does most of the talking and most of the questioning
    • A deliberate exercise where the candidate is asked to diagnose a real current problem in the organization — not to solve it, just to diagnose it
    • Reference conversations that specifically ask about how the candidate approached ambiguity
    • A de-emphasis on resume impressiveness in favor of judgment indicators

    The hit rate on hires since this change has been materially higher. Not because the process is magical. Because it’s designed around what actually predicts performance in the kinds of roles I’m hiring for.

    What I Still Look For

    The quality I now most deliberately hire for — the one this hire surfaced for me — is:

    Can this person see the real problem before they act?

    If the answer is yes, almost everything else is teachable. If the answer is no, everything else is brittle — because actions taken on incorrectly diagnosed problems usually make things worse.

    Most of the hires I’ve made were adequate. A few were wrong in predictable ways. The ones that were right almost always shared some version of the quality the hire I got right surfaced for me. I’ve been looking for it deliberately ever since.

    Hiring is one of the hardest things operators do. The hire that works well will teach you more about how to hire than any book or framework. Pay attention to the lessons. Apply them deliberately. Your next hire is a better investment when the prior hire taught you something specific.

  • The Board Meeting That Changed How I Forecast

    Looking back: April 2026

    There was a specific board meeting, years ago, that changed how I thought about forecasting. I hadn’t realized until that meeting how deeply I’d absorbed a culture where forecast commitment was treated as a display of confidence rather than as a probabilistic estimate. The board conversation that surfaced this was uncomfortable, instructive, and one of the more useful feedback moments of my career.

    The Setup

    I was walking the board through the current quarter’s forecast. Coverage looked reasonable. Pipeline quality was mixed. I’d committed to a number that was toward the optimistic end of the range because — as I told myself — that’s what you did. Boards wanted confidence. Leaders delivered confidence. The forecast was aspirational because aspiration was the expected register.

    I walked through the slides. Numbers matched expectations. I was ready to move to the next topic.

    An experienced board member interrupted. “I want to push on this. You’ve told us the commit number. I want to understand what you think the actual number is going to be.”

    The Question That Reframed Everything

    The question was simple but cracked something open. The commit number and the expected number had quietly become different things in my head — and I hadn’t been articulating the difference. The commit was aspirational. The expected was what I actually thought would happen. Those numbers differed by about 15%. I’d been showing the first and privately carrying the second.

    I said something like: “The commit is X. My personal expectation is closer to Y.”

    The board member nodded. “Thank you. That’s the number I want. I need to make decisions based on what you actually think, not on what you think you’re supposed to say.”

    The meeting shifted. We spent the next 40 minutes talking about the honest forecast — the one I actually believed — and the uncertainty around it. Where were the biggest risks? What were the leading indicators? What would I see in the next 30 days that would update my estimate?

    The conversation was more useful than any previous board conversation I’d had about revenue. Not because the number was better. Because the number was honest.

    What I Realized Afterward

    The lesson that took a few weeks to fully land: forecast commitment and forecast honesty are different things, and treating them as the same thing produces worse decisions.

    I’d been trained — through years of watching leaders perform confidence, through cultural messaging about “missing your number” being unforgivable, through my own discomfort with presenting uncertainty — to converge on a single commit number that was optimistic by design. The optimism served a signaling function (look how confident the leader is) but undermined the real function of forecasting (help decision-makers plan).

    The board member’s question forced a separation I hadn’t been making consciously: the commit is a promise to the team and the market; the expected is an analytical estimate. Different purposes, different numbers, different conversations.

    How I Forecast Now

    After that meeting, I shifted my forecasting practice in specific ways:

    1. I present ranges, not point estimates.
    The honest forecast has a low, mid, and high. The commit is usually somewhere between low and mid. The expected is around mid. The upside — what happens if things go unusually well — is the high. Three numbers, not one, force a real conversation about uncertainty.

    2. I explicitly state what would move the forecast.
    “If these three deals close on current timelines, we hit the upside. If any two slip, we hit the midpoint. If all three slip, we’re at the low end.” This gives the board (or any audience) the ability to track the forecast themselves, which builds trust faster than presenting a clean number ever did.

    3. I separate commit from expectation deliberately.
    The commit is what I’m willing to publicly stand behind and be measured on. The expectation is what I analytically believe. I present both, name them differently, and explain the gap. Some audiences find this jarring at first. They end up appreciating it.

    4. I review my forecast accuracy quarterly, honestly.
    The single most important forecasting discipline is looking back. Was my forecast within 5%? 10%? 20%? The pattern over four quarters tells me whether I’m a reliable forecaster, a systematically optimistic one, or a sandbagger. Every forecaster has a personal bias. Knowing yours is a precondition for adjusting.

    The Broader Lesson

    The board meeting taught me something beyond forecasting specifically. It taught me that performing confidence — in any domain — often costs more than it provides. Audiences (boards, teams, customers) are usually more sophisticated than the performance assumes. They can tell when the presentation is calibrated for signaling rather than for honesty. The leaders who consistently present honestly — with appropriate uncertainty, with explicit reasoning, with acknowledged risks — build trust that the performative leaders can’t match.

    This is hard in cultures that reward confidence. It’s the right bet anyway, because the alternative is a long career of building trust on unstable ground. One honest forecast compounds into permission to be honest about everything else. One performative forecast forces the next one to also be performative.

    What I’d Tell a First-Time CEO

    If you’re a first-time CEO or a first-time revenue leader, resist the temptation to converge to a single commit number that performs confidence. Present the range. Explain the uncertainty. Separate commit from expectation. Your board will be more helpful, your team will trust you more, and your own decision-making will be cleaner.

    The board meeting that taught me this was uncomfortable for me at the time. It’s also the meeting I most often refer back to when I think about what kind of operator I want to be.

    Honesty about uncertainty is a professional muscle. Most leaders aren’t taught it. Many cultures actively punish it. The leaders who develop it anyway build something more durable than the confident-performer alternative.

    It’s a quieter way to lead. Over time, it’s also more effective. I’m grateful for the board member who asked the question that forced me to see the difference.

  • A Partnership That Taught Me to Write Kill Criteria

    Looking back: April 2026

    There’s a specific partnership in my past that I think about more than most. It’s the partnership that made me a believer in kill criteria — the written, specific, measurable conditions under which a partnership gets formally reassessed and potentially wound down.

    Before this partnership, I would have told you kill criteria were a good idea. After it, I made them a non-negotiable condition of entering any new partnership. The shift in practice was the difference.

    The Partnership

    The details don’t matter in specifics — the partnership was with a complementary company in an adjacent market, announced with some fanfare, and intended to produce joint revenue through co-selling and integrated offerings.

    The opening months were what every partnership launch is. Press release. Joint customer kickoff. Executive enthusiasm on both sides. A shared vision for the quarter that would demonstrate the partnership’s value.

    The problems started quietly. Initial joint deals took longer than expected. The integration work between our products required more engineering than either side had scoped. The go-to-market motion required more enablement than the partner’s sales team absorbed. Each individual friction was small. In aggregate, they added up to a partnership that was consuming real resources and producing only modest returns.

    The Slow Drift

    Over the following 18 months, the partnership went through the classic lifecycle of partnership decay:

    • The initial ops cadence was weekly. It dropped to biweekly within four months. Monthly within eight. Quarterly within twelve. By month 15, we hadn’t had a formal review in over two months.

    • The executive sponsor on their side rotated into a new role in month six. The replacement didn’t have the same investment. They were polite but clearly had other priorities.

    • The dedicated partnership operator on our side moved to a different role in month ten. We didn’t backfill specifically — the responsibility got absorbed by a senior BD person who had four other priorities.

    • Joint pipeline stopped growing. Joint revenue flatlined. Neither of us formally raised the concern because neither of us wanted to be the one to call the partnership into question.

    By month 18, the partnership existed on paper and in a shared Slack channel that had gone mostly silent. We’d both spent significant resources. The revenue return was negative when we factored in the real costs.

    Why It Persisted

    Here’s the lesson I took from watching this happen — to myself, to the partner, to multiple partnerships I observed from the outside in subsequent years:

    Partnerships persist past their usefulness because nobody has the authority or the political space to end them without specific triggers.

    Our executives liked their executives. Nobody wanted to be the one who called the partnership dead. The formal termination would require a conversation that would be awkward, and with nothing forcing it, the path of least resistance was to let the partnership quietly continue doing nothing.

    Kill criteria would have forced the conversation. With specific written triggers — “if joint revenue is below $X by month Y, we review” — the partnership would have been formally assessed at the right time. Whatever the outcome of that assessment, it would have been a clearer, faster, less politically loaded process than the slow drift we actually experienced.

    What I Wrote Into the Next Partnership

    When I set up my next significant partnership — same general category, different partner, different structure — I insisted on explicit kill criteria written into the partnership agreement. Specifically:

    • At 90 days: named operators in place, weekly ops cadence running, minimum number of joint prospects in pipeline.
    • At 180 days: specific revenue threshold for joint-won deals. If below, formal review by both executive sponsors.
    • At 365 days: minimum ROI threshold for continued investment. Below that, formal reassessment — either restructure, scope reduction, or wind-down.

    The partner pushed back initially. “It feels like we’re planning for failure.” I explained my reasoning — not planning for failure, planning for clarity. Both sides of a partnership deserve to know what success looks like and when the relationship should be reassessed. Without those markers, partnerships drift past their usefulness.

    They agreed. We wrote the criteria into the agreement. And when we hit the 180-day mark, we sat down with the data and had an honest conversation. The partnership was performing — but below target. We adjusted scope, revised the cadence, and committed to a 90-day rework period. At 270 days we reassessed.

    We restructured. The second phase of the partnership was meaningfully more productive than the first — not because the underlying potential was different, but because the kill criteria had forced the conversations that made us redesign the motion.

    The Principle

    Kill criteria are not about ending partnerships. They’re about creating forcing functions that convert partnerships from drift mode to decision mode. Every partnership I’ve seen run well has had them. Every one I’ve seen drift into zombie status has lacked them.

    The same principle applies to other domains — strategic initiatives, product investments, hiring experiments, organizational structures. Anything that could drift past its useful life benefits from written decision triggers.

    What I Tell Anyone Starting a Partnership Now

    Write the kill criteria before the press release. If the other side won’t agree to them, think carefully about whether you actually have a partner or a vanity alliance. Real partners are willing to commit to the conditions under which the partnership gets formally reassessed, because real partners understand that a clear, structured reassessment is in both parties’ interest.

    The partnerships that work are the ones that both sides take seriously enough to hold each other accountable. Kill criteria are the mechanism for accountability. Without them, accountability drifts.

    The partnership that taught me this was an expensive education in what drift costs. The restructured version of the next partnership — the one where we’d learned from the prior experience — was a meaningfully better relationship.

    I’d rather have had the lesson earlier. Now that I have it, I pass it on every time I’m involved in structuring a partnership.

    The cost of writing kill criteria is an awkward conversation at the start. The cost of not writing them is an awkward 18 months later.

    Choose the earlier conversation. Every time.

  • The Day I Realized the Channel Was Broken

    Looking back: April 2026

    I’ve been through multiple channel partnerships over the course of my career — both as a vendor managing a channel and as an observer of companies that did. The moment I most clearly remember — the one that rewired how I think about channel health — was a specific Tuesday afternoon, in a specific meeting, when I realized a partnership I’d been running for almost two years had been dead for at least eight months without my noticing.

    The Setup

    The channel in question was a distribution partnership — our product going to market through a partner’s sales motion, revenue shared according to a written agreement, with joint pipeline reviews scheduled quarterly.

    On paper, everything was fine. We had joint customers. We had a contract. We had quarterly reviews. Revenue had plateaued but not dropped, which I’d attributed to natural maturation of the relationship.

    What I hadn’t noticed was that every one of those “normal” signals had been deteriorating for months. The joint customer base wasn’t growing. The quarterly reviews had become increasingly perfunctory. The specific contacts I’d been working with at the partner had rotated out, and the new ones didn’t know me or care about the partnership in any meaningful way.

    Revenue was flat, which in a growth environment is actually decline. But flat revenue doesn’t trigger alarms in most organizations the way declining revenue does, so I hadn’t been alerted.

    The Meeting

    The Tuesday that clarified things was a routine quarterly review. I walked in expecting the usual: pipeline update, a review of joint customers, a light discussion of what was coming next.

    The partner’s new regional head — who I’d met once, briefly — opened by saying: “I want to be direct with you. This partnership doesn’t have an operator on our side anymore. The person who championed it is gone. I’m inheriting it without context, and frankly, I’m not sure what it’s for. Can you walk me through what success would look like from your perspective?”

    I couldn’t.

    Not because I didn’t know what success looked like in theory. I knew that. I couldn’t walk him through it because, on reflection, the partnership hadn’t been set up with kill criteria, hadn’t been rebuilt when the original champion left, and hadn’t had a real operating motion for many months. It had been running on momentum from a past era, and I’d been managing the appearance rather than the substance.

    The meeting was cordial. We agreed to reassess. Within 90 days, we wound it down formally.

    What I Should Have Seen Earlier

    Several signals had been present that I’d either missed or rationalized:

    The original champion left, and we didn’t replace the relationship.
    Six months before the Tuesday meeting, my primary contact at the partner had moved to another role. The handoff had been brief. The new contact didn’t have the same investment. I noticed at the time but assumed the relationship would rebuild naturally. It didn’t.

    The meetings got thinner but we kept holding them.
    The quarterly reviews had been degrading in substance. Earlier ones had real discussion of pipeline, roadmap, customer issues. Later ones were status reports with no decisions attached. I’d read the drift as normal evolution rather than as a warning sign.

    Revenue was flat in a growth market.
    In hindsight, this was the clearest signal. Both companies had been growing during the relevant period. The partnership’s contribution had not. That delta — partnership growth vs. organic growth — was negative, even though the absolute number was flat. I hadn’t looked at the comparison that way until much later.

    The partner had stopped volunteering opportunities.
    Earlier in the partnership, the partner had brought deals to us regularly. By the time of the Tuesday meeting, all the “joint deals” were deals we’d brought to them. The flow had reversed without my noticing.

    The Lesson That Generalized

    After that experience, I developed a more rigorous framework for evaluating any channel or partnership I was running:

    1. Operator continuity on both sides.
    If the original operator left and hasn’t been replaced with someone who has real investment, the partnership is at risk. This is the most reliable leading indicator I’ve found.

    2. Growth contribution delta.
    Not absolute revenue — revenue growth relative to organic growth in the same period. Partnerships that aren’t accelerating above baseline are decelerating, even if the absolute number looks stable.

    3. Bilateral deal flow.
    Real partnerships have deals flowing in both directions. When one side stops volunteering opportunities, the relationship has shifted from partnership to vendor-provider, even if both sides still call it a partnership.

    4. Meeting substance.
    If the quarterly review has become a status report instead of a working session, the partnership has declined. Meeting substance is a leading indicator; revenue decline is a lagging one.

    What I Do Differently Now

    Every channel or partnership I’m involved in now has a quarterly health check that explicitly reviews those four signals. The goal isn’t to prevent partnerships from ever declining — that’s unrealistic. It’s to catch the decline early, while there’s still time to reset or wind down deliberately.

    The Tuesday meeting taught me that partnerships can die without dying. They can continue generating just enough activity to mask the underlying deterioration, and the activity itself becomes the problem — because it convinces both sides there’s still something there when there isn’t.

    The antidote is specific, structured health checks. Signals that matter. Thresholds that trigger action. Kill criteria written before the relationship needs them.

    If you’re running a channel or partnership that hasn’t had a real health check in a while, schedule one. Ask the four questions honestly. You might find everything is fine. You might find what I found, which is that the relationship has been dead longer than you realized.

    Better to know. The Tuesday I realized my channel was broken was one of the more important Tuesdays of my professional life.

  • The First Time I Fired a Customer

    Looking back: April 2026

    The first time I formally ended a customer relationship was harder than I expected and more important than I realized. It’s one of those small professional moments that shaped everything that came after, though I couldn’t have said so at the time.

    The Setup

    The customer in question was well-known. They paid. They weren’t particularly difficult in obvious ways. But over the course of the engagement, a pattern had emerged: they consumed disproportionate resources, they negotiated every small item as though it were existential, and their team treated my team as adversarial rather than collaborative.

    The math on the account had slowly shifted from profitable to marginal to negative. Support hours escalated. The delivery team dreaded every call. Morale around the customer had soured.

    I’d been told by every mentor and business book I’d ever consumed that firing a customer was legitimate. That some customers cost more than they produced. That walking away was sometimes the right call. Intellectually, I agreed.

    In practice, actually doing it was something else entirely.

    Why It Was Hard

    Three reasons, in order of intensity:

    1. The revenue was real.
    Losing the account meant losing revenue. Revenue is always tangible. The cost of keeping the account was distributed across the team and the quarter — less tangible, harder to sum. The instinctive math always favored keeping them.

    2. The story I’d tell myself about losing a customer felt worse than the reality of keeping them.
    I’d spent years building a book of business. Every customer was a win. Losing one felt like a loss, regardless of the circumstances. I realized, in thinking about it, that I had an irrational attachment to the customer count as a measure of my professional standing — and that the attachment was making it hard to make the economically rational decision.

    3. The conversation itself required specific skills I hadn’t developed.
    “We’re choosing not to continue our relationship” is a sentence I’d never had to say before. Saying it well, without drama, without burned bridges, required a poise I didn’t naturally have. I rehearsed the conversation more than any meeting I can remember preparing for.

    How It Went

    The conversation itself went better than I’d feared. The customer was surprised, then defensive, then — once they realized I wasn’t going to be negotiated out of the decision — gracious. They appreciated the directness. They wished us well. The meeting ended in about 40 minutes.

    A week later, the lead stakeholder sent me a note thanking me for the professionalism. He said — and I remember this exactly — that he’d had vendors walk away from his business before, and none of them had done it with the clarity I had. It was the first time I’d heard that feedback, and it changed how I thought about what “firing a customer” even meant.

    It wasn’t a failure. It wasn’t even a loss. It was a specific professional action, done well, that left both sides better off than if I’d tried to limp through another six months of a dying relationship.

    What Shifted

    Several things changed for me permanently after that experience:

    The customer count stopped being my benchmark.
    I started paying attention to customer quality in ways I hadn’t before. The question became not “how many customers do we have?” but “how many of our customers are ones we’re proud to work with?” The first question has a ceiling. The second one has a floor.

    I got better at the pre-signing conversation.
    Having actually walked away from a customer, I became more willing to surface fit concerns before a contract was signed. The customers I’m most reluctant to bring on now are the ones I know would be hard to walk away from later. Better to have the hard conversation up front than to manage through months of difficulty.

    My team changed.
    The delivery team, specifically, noticed that I’d chosen their morale over the revenue. That one decision did more for team trust than any of my leadership intentions had. They knew, after that, that I wasn’t going to trade their well-being for my numbers. Every retention and recruiting conversation we had afterward built on that foundation.

    The remaining customer base got better.
    It wasn’t immediate, but over the following year, the quality of our customer base noticeably rose. Partly because we’d freed up resources to serve the good customers better. Partly because word travels in any industry, and the customers we wanted more of could tell we were the kind of vendor who wouldn’t tolerate bad-fit relationships — which is exactly the kind of vendor the best customers want to work with.

    The Principle That Generalized

    The first time I fired a customer, I was operating on the assumption that every customer is valuable and that ending a relationship is a failure mode. What I learned is the opposite: the willingness to end a customer relationship is a marker of operational maturity, and companies that can’t do it have a ceiling they can’t see.

    This generalizes beyond the specific action of customer termination. The same underlying skill — ending what isn’t working with clarity and grace — applies to underperforming hires, failed partnerships, unprofitable products, strategic dead ends. The founders and operators I’ve watched scale successfully all share this skill. The ones who struggle often can’t bring themselves to exercise it.

    If you’ve never fired a customer, it might be a skill worth developing before you need it. Not because you should fire any specific customer today — just because the capacity to do it changes how you think about every other decision downstream.

    The first time is hard. Every time after is easier. By the third or fourth time, it’s just one tool among others in the operator’s kit, used sparingly, but present.

    That’s where you want to be. It took me one specific customer to get there. I’m grateful for the lesson.

  • Crossing From Employee to Founder

    Looking back: April 2026

    The transition from employee to founder is one of the most discussed, most written-about, and most mischaracterized transitions in professional life. Everyone talks about the obvious changes — autonomy, risk, ownership. The less-discussed changes are the ones that mattered most in my case, and they’re worth reflecting on because I think they apply to most people making the same transition.

    What I Thought Would Be Hard

    When I was considering starting something, I thought the hard parts would be:

    • The financial risk of leaving a paycheck
    • The uncertainty of whether the idea would work
    • The difficulty of hiring and managing early team members
    • The pressure of being responsible for outcomes

    Those things were real, but they weren’t actually the hardest parts.

    What Was Actually Hard

    The loss of institutional framing.
    As an employee — especially at mature companies — my time had structure. My role had structure. My incentives had structure. The company’s framework absorbed a lot of the “what should I do next” cognitive load, and I could spend most of my mental energy on execution within that framework.

    As a founder, there’s no framework. Every day, I had to decide what mattered, what to ignore, what to prioritize, what to defer. The absence of a structure to push against was disorienting in ways I hadn’t anticipated.

    This was harder than the financial risk. Harder than the uncertainty. Harder than anything on the list of expected difficulties. It took me six to twelve months to build the internal replacement for institutional framing — my own rhythms, priorities, diagnostics — and the transition period was genuinely hard.

    The feedback loop got longer.
    As an employee, feedback on my work was pretty fast. My manager would tell me what was working. The company’s metrics would tell me what was landing. I could course-correct in weeks.

    As a founder, the feedback loop on strategic decisions is often months or years. You make a choice about positioning in month one and don’t know if it’s right until month twelve. You hire someone in month three and don’t know if the hire is working until month nine. The slowness of the signal requires a kind of patience that employee life didn’t train me for.

    What I learned, eventually, is that the way to manage long feedback loops is to trust shorter process signals more heavily. Are the meetings productive? Are the conversations getting somewhere? Is the team energized? These are weekly signals that predict outcomes months later. They’re not outcomes themselves, but they’re leading indicators in a way that topline metrics often aren’t.

    The identity shift took longer than the role shift.
    Calling myself a founder on business cards was easy. Actually thinking like a founder — taking responsibility for things that would have been someone else’s problem as an employee, proactively making decisions instead of escalating them, owning the outcomes in a way that no one else can — took much longer.

    For probably the first year, I was running founder operations with an employee mindset. I was doing the mechanical work of being a founder but still had employee reflexes about risk, about authority, about asking for permission. Unlearning those reflexes was slow and uneven.

    What Surprised Me Positively

    The network I’d built as an employee was more valuable than I’d realized.
    I expected my previous relationships to be useful for finding customers, advice, and occasional introductions. I didn’t fully appreciate that they were also deeply valuable as a source of honesty. When I was struggling — with a strategic choice, a hiring decision, a customer conflict — the friends I’d built over years of prior work told me the truth in ways that board members and advisors often couldn’t.

    The professional network, treated as a reputation asset, compounds in employee life in ways you don’t notice until you leave. Then you realize how much of your foundation is made of those relationships.

    The founder role concentrated things I’d liked intermittently.
    As an employee, the parts of my job I’d liked most were the strategic conversations, the customer relationships, the hard problems. Most of my time was spent on other things. As a founder, the ratio of “things I like doing” to “things I don’t” shifted meaningfully. Not because founding is easier, but because I could design the role around what I was actually good at.

    This is a subtle but real benefit of founding that people don’t discuss enough. The job is harder. But the distribution of work is usually better-aligned with the founder’s strengths than the previous employee role was.

    What I’d Tell Someone Considering It

    If you’re thinking about making the transition:

    • The financial risk is real but manageable for most people with planning.
    • The uncertainty is real and you’ll never fully resolve it.
    • The hard parts are the ones you haven’t thought about — the loss of institutional framing, the long feedback loops, the identity shift.
    • Your network matters more than you think. Invest in it before you leave, not after.
    • The role, once you’ve adapted, usually fits better than the role you left.

    The transition is hard. For most people who make it and succeed at it, it’s worth it. But the success usually comes from adapting to the non-obvious difficulties, not from doing well at the expected ones.

    Crossing from employee to founder is not what the outside narrative suggests it is. The hard parts are internal. The help comes from relationships you invested in before you needed them. And the identity — the part where you actually think like a founder — takes longer than any external milestone suggests.

    That’s the arc. Plan accordingly.

  • What the AI-Era Buyer Looks Like

    Looking back: April 2026

    The buyer I sell to in 2026 is materially different from the buyer of 2022. Some of the changes are obvious — AI-assisted evaluation, faster information synthesis, different expectations about vendor responsiveness. Some are subtler, and the subtler ones matter more.

    Here’s how the buyer has actually changed, and what it means for selling in an AI-permeated environment.

    The Buyer Knows More, Faster

    The 2022 buyer did research before engaging with a seller, but the research was bounded by their time. An hour of buyer research produced an hour’s worth of understanding.

    The 2026 buyer does research that’s AI-augmented — meaning a few minutes of prompt-writing can produce a reasonably comprehensive synthesis of a vendor, their market, their competitive set, and the case for and against them. The research bandwidth has expanded dramatically. The buyer shows up to the first meeting with a depth of understanding that would have taken days to build pre-AI.

    This has changed discovery. The “tell me about your business” opener was already outdated by 2020. In 2026, it’s actively harmful — the buyer knows you’ve done no preparation if that’s how the conversation starts, because they’ve done significantly more.

    The adaptation: sellers have to do equivalent AI-augmented preparation themselves. A seller who shows up without AI-enabled research has fallen behind the buyer’s baseline expectation. The quality differential between sellers who use AI preparation well and those who don’t is larger than any single tactical difference in the pre-AI era.

    The Buyer’s Patience Compressed Further

    If meetings were shorter in 2020-2022, they’re even shorter now. The 30-minute meeting has increasingly compressed to 15-20 minutes for initial conversations. The tolerance for any non-essential content has dropped further.

    Part of this is general attention compression. Part of it is that AI summarization has made buyers expect to get the same information in less time — if their AI can summarize a white paper in 30 seconds, they resist a 45-minute meeting that could have been five bullet points.

    The meeting discipline required is significantly tighter. Open strong. Land specific value fast. Make the next step concrete and low-friction. Respect the buyer’s time at a level that would have seemed rushed pre-AI.

    The Buyer’s Noise Floor Rose

    Outbound that would have worked in 2020 now fails. Cold email templates, generic LinkedIn messages, standard SDR cadences — AI has democratized the ability to produce these at scale, which means buyer inboxes are flooded with templated outreach indistinguishable from spam.

    The only outbound that consistently breaks through is either (a) specifically, manually personalized in ways AI-generated templates can’t match, or (b) coming from a relationship that pre-exists the outbound.

    This has bifurcated the outbound market. The sellers still running high-volume templated outbound are hitting open rates and response rates that would have been considered unacceptable in 2020. The sellers doing true specificity — which is hard to automate — are still getting responses at pre-AI rates.

    The Buyer Trusts Humans More, Not Less

    This is counterintuitive but important: in an AI-saturated environment, the human component of selling has become more valuable, not less.

    Buyers know they’re surrounded by AI-generated content, AI-written emails, and AI-assembled product information. Against that backdrop, a human seller who clearly knows their specific situation — who remembers the conversation from last quarter, who references a specific detail the buyer mentioned, who says something that an AI wouldn’t have said — stands out more than they used to.

    The buyers I’ve talked to recently often explicitly value the human element. They want to buy from humans who understand them, not from AI-mediated processes. This has made high-craft selling more differentiated in an AI world, not less.

    The counterintuitive implication: the sellers who are winning in 2026 are those who use AI heavily for research, preparation, and operational efficiency — while keeping the customer-facing interactions distinctly human. The buyer gets the benefit of the seller’s AI-augmented understanding without feeling like they’re being sold to by AI.

    The Buying Committee Has an AI Dimension

    Something new in 2026: most enterprise buying committees now include, implicitly, a set of AI tools. The buyer has access to AI analysts that help them evaluate vendors. The buyer’s legal team uses AI for contract review. The buyer’s technical team uses AI to assess claims. These aren’t formal members of the committee, but they shape the committee’s decisions.

    This is relevant for sellers because your content and your positioning now have two audiences: the human stakeholders and the AI systems that assist them. If your public material is unclear or underspecified, the AI can’t help the buyer understand it well, and the buyer loses conviction. If your public material is detailed, specific, and clearly structured, the AI helps the buyer build a stronger internal case.

    The Lesson for 2026

    The AI-era buyer is faster, better-informed, more impatient, more surrounded by noise, and more hungry for authentic human interaction. Selling to them requires all of the high-craft motions of the pre-AI era, executed at a higher tempo, on tighter time constraints, against a noisier environment, while using AI heavily in your own preparation and operations.

    The sellers who’ve adapted do better than ever. The ones who haven’t are struggling. The gap between the two is the widest I’ve ever seen it.

    If you’re going to build a sales motion for the current buyer, build it assuming the buyer is better-prepared than you used to think, more impatient than you used to design for, and more skeptical of everything AI-generated. Then differentiate by bringing specific human craft that AI can’t replicate.

    That’s the motion that works now. And from what I can see, it’s going to keep working for a while.

  • The 2023-24 Efficiency Era

    Looking back: April 2026

    The period roughly from mid-2022 through most of 2024 was the efficiency era in B2B. Growth-at-all-costs became profitable-growth. Rule-of-40 became the standard benchmark. Revenue organizations had to re-optimize for capital efficiency in ways most of them hadn’t had to before. What came out of that period reshaped how revenue is thought about in 2026, and some of the lessons are worth preserving even now that growth has returned as a priority.

    What Changed

    The efficiency era forced three specific changes that persisted:

    1. The ROI conversation moved earlier in the sale.
    Pre-2022, many enterprise deals closed on “strategic value” without a rigorous ROI case. Buyers had budget. They could commit to strategic initiatives on partial information. The CFO was a rubber stamp in many organizations.

    In the efficiency era, that flipped. The CFO (or the CFO’s office) became a primary stakeholder in most enterprise purchases. Deals without a clear ROI argument stalled. Deals with rigorous ROI cases closed at materially higher rates. Sellers who could help their champions build the financial case became dramatically more valuable than sellers who relied on strategic-value arguments.

    That shift stuck. Even now, in an environment where growth is more rewarded, buyers have retained the muscle of requiring ROI-based justification. Sellers who don’t build financial cases are at a permanent disadvantage.

    2. Customer expansion became more important than new logo.
    In the efficiency era, expansion revenue was cheaper to acquire than new-logo revenue. Companies optimized accordingly — shifting resources from new business to customer success, account management, and expansion motions.

    Many of those shifts proved permanent. Customer success transformed from a reactive support function into a proactive revenue function. Account executives started being measured on expansion as well as new business. The economics of the customer base became more visible to executives.

    3. Pipeline quality replaced pipeline quantity as the primary discipline.
    Pre-efficiency era, pipeline coverage ratios of 3x, 4x, 5x were common — and most of that pipeline was low-probability. Efficiency era showed that carrying inflated pipeline was expensive: every deal required seller time, SE time, deal desk time, legal time. The cost of pursuing 5x coverage with a 15% win rate was often worse than 2x coverage with a 35% win rate.

    The shift to pipeline-quality discipline stuck. Qualifying out became a legitimate activity. Sellers who walked away from bad-fit deals looked better, not worse. Forecast accuracy rose. The average deal in pipeline got richer even as the quantity dropped.

    What Didn’t Stick

    Some efficiency-era patterns reverted once the macro environment loosened:

    • The most aggressive cost-cutting of GTM teams proved to be over-correction. Companies that cut too deep ended up rebuilding in 2024-2025, often at higher cost.
    • Marketing budget compression went too far at many companies. The ones that preserved some demand generation during the efficiency era had easier growth acceleration later than the ones that eliminated it entirely.
    • Some companies over-rotated to product-led growth, which worked for specific categories but not for the enterprise-heavy businesses they were applied to.

    The Pattern Underneath

    What the efficiency era really surfaced — once the macro noise was filtered out — was which revenue motions were durable and which were dependent on cheap capital. Companies that had been masking motion inefficiency with abundant marketing spend or aggressive hiring had nowhere to hide. Companies with genuinely efficient motions came through the period strengthened.

    That sort has held up. In 2026, the revenue organizations that emerged from the efficiency era with strong discipline are outperforming the ones that tried to keep the old model alive.

    What I Watch Now

    Coming out of that era, there are a handful of metrics I watch that I didn’t pay as much attention to before:

    • CAC payback by segment. Not blended CAC/LTV. Specifically, how long does it take to pay back the acquisition cost on the segment of customers you actually want. If the answer is more than 18 months, the segment may not be viable at current economics.
    • Net revenue retention from year two. Year-one NRR is noisy because it includes implementation dynamics. Year-two NRR tells you whether the customer base is actually expanding or slowly churning with a veneer of expansion.
    • Win rate trend by segment. If win rate is dropping in your best segment, something is changing — either competition, positioning, or buyer behavior.

    These are operational metrics, not strategy metrics. They tell you whether the revenue engine is healthy under the surface.

    What I’d Tell Someone Rebuilding Post-Efficiency

    If you’re rebuilding a revenue organization in 2026, the efficiency-era lessons are still the right foundation:

    • ROI-first discovery and qualification
    • Expansion motions as a primary, not secondary, revenue source
    • Pipeline quality over pipeline quantity
    • Segment-specific economics rather than blended averages
    • Operational metrics reviewed weekly, not just quarterly

    Once those are in place, the growth-era motions — demand generation, geographic expansion, new-logo acquisition, enterprise land-and-expand — layer on top of a durable foundation.

    The 2023-24 efficiency era was painful for many revenue leaders. It was also one of the most clarifying periods in my career for thinking about what actually makes a revenue organization durable. The companies that emerged with discipline are still benefiting from it. The ones that didn’t are still catching up.

  • Post-Pandemic Buyer Behavior

    Looking back: April 2026

    The buyer I sold to in 2019 doesn’t exist anymore. Some of what changed was temporary; some was permanent; some was always there and just became more visible. Sorting the three has been clarifying — and has changed how I approach every sales motion since.

    Here’s what I think actually changed, and what it means for how to sell in 2026.

    Buyers Are Better Informed Before You Meet Them

    Pre-2020, a significant portion of the buyer’s education happened during the sales cycle. You ran demos, walked them through case studies, explained the product, answered their questions.

    Now, by the time a buyer is willing to take a first meeting, they’ve already read your website, watched a demo video, scanned your LinkedIn, read about your company, and often talked to one or two peers who use your product. The seller’s role has shifted from educator to validator — you’re not teaching them what the product does; you’re helping them confirm (or challenge) what they already think.

    This changed the craft of discovery. The old “walk me through what you do” opening is now a waste of time. The buyer already knows. The better opening is “what do you already understand about us, and what are the gaps I can fill in?” That question respects the buyer’s preparation and surfaces what they actually need from you.

    Meetings Are Shorter

    The 60-minute pre-2020 meeting has given way to the 30-minute meeting as default. Buyers schedule tighter calendars. They tolerate less pre-amble. They expect you to get to the point faster.

    This tightening reshapes meeting design. You can’t run a 30-minute meeting the way you ran a 60-minute one with a compression factor. You have to restructure — less discovery theater, more specific value delivery, tighter next-step definition.

    The sellers who adapted to this kept their effectiveness. The ones who tried to squeeze the old 60-minute structure into 30 minutes just delivered worse meetings in less time.

    Buying Committees Are More Skeptical

    Something shifted in the 2020-2022 period that left buyers permanently more cautious. Deals that would have closed on a verbal commitment in 2019 now go through three rounds of internal review. Deals that would have moved on the economic buyer’s say-so now need to satisfy a procurement team, a legal team, and often an executive committee.

    Part of this is post-crisis tightening of governance. Part of it is genuinely higher buyer sophistication. Part of it is skepticism about vendor claims that built up during a period when vendor promises outran delivery capacity.

    Whatever the cause, the practical effect is that closing an enterprise deal now requires more internal advocacy, more evidence, and more patience than it did pre-2020. Sellers who forecast on old timelines miss their numbers. Sellers who adapted to the new reality of longer internal processes forecast more accurately.

    The Tolerance for Vendor-First Messaging Collapsed

    Pre-2020, buyers would sit through vendor-first messaging — product decks, feature comparisons, company origin stories — before getting to what they actually needed. They’d give you the courtesy of the full presentation before engaging.

    They don’t anymore. If the first five minutes of your meeting are about you and your product, the buyer is either disengaged or moving toward ending the meeting. What works now is buyer-first messaging — “here’s what we think is happening for organizations like yours,” followed by specific relevance, followed by your company only when the buyer asks for it.

    References and Social Proof Became Dominant

    Pre-2020, references mattered but weren’t always decisive. Now they often are. Buyers who can’t talk to two or three current customers before deciding usually won’t decide.

    This has raised the importance of customer success and reference programs dramatically. Companies with strong reference programs close deals that companies with weak programs lose. The delta is often larger than any product feature difference.

    Buyers Are Faster to Disqualify

    Pre-2020, a buyer who was lukewarm on your offering would often continue the conversation anyway — giving you a few more meetings to change their mind. Now, a lukewarm buyer disqualifies early. If the first meeting doesn’t create clear excitement, there often isn’t a second meeting.

    The Principle I Apply Now

    Every aspect of the sales motion has become more unforgiving. The messaging has to be more buyer-first. The meetings have to be more specific. The process has to account for longer internal cycles. The references have to be stronger. The margin for error on any specific interaction is smaller.

    This sounds harder — and it is, for sellers running old-school motions. For sellers who already ran high-craft motions, it’s actually easier now than it was pre-2020, because the buyers reward high-craft work more distinctly than they did before.

    The meta-lesson: the post-pandemic buyer amplified the differential between thoughtful selling and transactional selling. Thoughtful wins. Transactional loses. The middle ground compressed.

    If you’re selling to a 2026 buyer the way you sold to a 2019 buyer, you’re almost certainly missing. If you’ve adapted to the new buyer, you’ve probably found a market that rewards serious craft more than it used to. Both realities are downstream of the same underlying shift.

  • The Remote-Selling Shift

    Looking back: April 2026

    When enterprise selling moved remote in 2020, a lot of what had worked in the pre-2020 era had to be rebuilt. Some of the rebuilding produced better motions than existed before. Some of it produced worse ones. Understanding which is which has been clarifying in retrospect.

    What Got Harder

    Reading the room.
    In-person meetings transmitted dozens of small signals — how the champion was reacting to the skeptic, who was checking their phone, whether the energy in the room was with you or against you. On video, most of those signals disappeared. Sellers who relied heavily on reading the room had to rebuild their craft around the narrower bandwidth of video meetings.

    The best sellers found ways to compensate — asking more direct questions, checking in explicitly with specific stakeholders, structuring the meeting to draw out reactions that would have been spontaneous in person. The lesser sellers kept expecting the signals to come through and missed them.

    The hallway conversation.
    In-person meetings had a structure where the formal meeting was 70% of the value and the 30% that happened in hallways, over coffee, walking to the parking lot was often where the real progress happened. Those informal moments disappeared almost entirely in remote selling.

    The adaptation was to engineer equivalents: specific one-on-one video calls, separate calendar time for off-record conversations, use of messaging platforms for between-meeting dialogue. These worked, partially. The equivalents never fully matched the spontaneous-hallway version.

    Building trust with people you haven’t met in person.
    Some of this wasn’t obvious until later. Trust can be built on video — but the depth and durability of that trust, in retrospect, seems shallower than the in-person equivalent. Relationships formed entirely remote during 2020-2022 often proved less resilient to difficulty than relationships with equivalent touch count that included some in-person contact.

    The lesson: the high-bandwidth modalities build relationships faster than the low-bandwidth ones. When everything was low-bandwidth, the relationships still formed, but with shallower foundations.

    What Got Easier

    Access.
    Getting a meeting with a senior executive became dramatically easier in 2020. The logistics overhead of in-person meetings — flights, parking, calendar blocks — had constrained how many meetings senior people could take. When all meetings became 30-minute video calls, senior executives could take 6 to 10 meetings a day where they previously took 3 to 5.

    This meaningfully expanded the pool of possible conversations for many sellers. Cold outreach to executives became more productive because the calendar constraint had eased. Existing relationships became easier to maintain because a quick video check-in was lower-friction than scheduling a lunch.

    Geographic reach.
    Regional sellers who could previously only cover their geography could now work across regions. Enterprise deals in cities the rep had never visited became viable. This expanded the effective market for many sellers and compressed the advantage of locally-based competitors.

    Parallel engagement across buying committees.
    In-person selling was almost always sequential — you had meetings with specific stakeholders in specific sequence, and the timeline was governed by travel logistics. Remote selling let sellers run parallel engagement across multiple stakeholders simultaneously. The CTO call could happen Tuesday, the CFO call Wednesday, the operations call Thursday — with all three feeding into a unified internal pipeline of the deal.

    This actually compressed some deal cycles, especially for deals where multi-threading was the primary constraint.

    What Became a Permanent Mixed Bag

    Discovery.
    Discovery calls got more efficient (no travel time, tighter agendas) and less rich (fewer side conversations, less whiteboard collaboration). The net effect varied by industry and deal type. Complex, multi-stakeholder discovery still benefited from in-person intensity. Simpler, single-stakeholder discovery was often better on video.

    Closing.
    Closing a significant deal via video remained awkward for years. Something about the final steps — the handshake equivalent, the moment of commitment — didn’t transmit cleanly over the video medium. Many sellers found that they could do 80% of a deal remotely but wanted the final stages in person.

    What I’d Tell Anyone Rebuilding

    If you’re optimizing a sales motion in 2026, the remote-selling lessons are still live:

    • Don’t fight the hybrid equilibrium. Engineer it.
    • High-bandwidth modalities (in-person, long-form video, workshops) build deeper relationships faster. Use them strategically.
    • Low-bandwidth modalities (email, chat, short video) scale broader but shallower. Use them for volume.
    • Sequence deliberately. Use remote for early, in-person for inflection points, remote for execution.

    The sellers doing the best work now are the ones who’ve internalized this layering. Not remote-first. Not in-person-first. Deliberate-sequencing.

    The remote shift didn’t kill in-person selling and it didn’t replace it. It reshaped the hierarchy of which modalities do which work. Five years in, the sellers who’ve digested that reshaping have better motions than existed before. The ones who haven’t are still nostalgic for 2019 or still overcommitted to 2020.

    Neither nostalgia nor overcommitment serves. The hybrid reality is the stable one now. Work from there.