July 7, 2026
The function looked stronger on paper than it ran.
Field Notes · No. 5
The leader describes what should happen. The CSM describes what does. The gap between those two answers is the most reliable signal I've found in this work.
In addition to building post-sale motions from the ground up, I've inherited them. In one instance, I moved into this seat at a company that had, by any documentation standard, a complete operating model. A lifecycle map with named stages and entrance and exit criteria. A value realization motion with verified moments of truth at key points in the year. A health score rubric. A comp plan tied to outcome delivery, not just activity. On paper it was robust and more mature than many.
I asked one of the CSMs what triggered the value review conversation with a district. She told me that it was supposed to happen mid-year. I asked when the last time was that any automation had actually prompted it.
She said she wasn't sure it ever had.
I then asked leadership. Their answers were confident: we run a structured value review in November/December. Standard.
The gap between what the leader describes and what the CSM experiences is the most reliable signal I've found in this work. It tells you more about where a function actually is than any number on a dashboard, any bit of process documentation, or any single conversation. Most scoring conversations or diligence interviews never surface it, because they start with the wrong person.
Why self-assessment inflates the score
When you've built a motion, you can't see where it breaks. You're patching the gaps in real time, adjusting in the moment, covering for what a newer CSM doesn't know, catching the account the system flagged wrong. You don't experience those patches as gaps. You experience them as judgment calls. The motion feels complete because you're completing it.
A leader detailing their own function will score it higher than where it actually runs, and not dishonestly. They believe the score. The problem is they're scoring the motion they intended, not the one that fires consistently in practice.
The CSMs scoring the same function score it lower. Because they're the ones in the room when it doesn't fire. They know which steps get skipped when it's October and the book is full. They know which Salesforce fields get filled in after the call, from memory, because there wasn't time during. They know that the verified milestone is often a guess made from a call they can only half-remember.
The gap between those two scores is not an alignment problem or a morale problem. It's diagnostic information. It's telling you exactly which parts of the well-intended motion exist only in your internal wiki.
Two questions that cut through it
There are two questions I use when I want an honest read.
The first: would this step still happen if the person who built it, and the person currently running it, were both gone?
This isn't a hypothetical. CSMs turn over. The implementation lead who held the onboarding motion leaves in March. The VP who designed the value framework is in a transition. When the next wave of districts needs to go live, the people who knew what "done" looked like are gone.
A step that only happens because someone remembers to do it isn't a system. It's a habit that belongs to a person. When the person leaves, the habit leaves.
The second: if I scored this dimension against what actually happened last cycle, not what should happen, but what did happen, in the accounts I didn't personally touch…where does it land?
That distinction matters. The playbook is not the evidence. The evidence is what fired, what didn't, and what the team can actually point to from the last twelve months that shows the motion ran. In many companies, this is harder than it sounds. The documentation is often thin. Most of the evidence lives in the heads of people who may not be there anymore or who don't have capacity to share.
What the gap looks like in practice
The most common pattern: a function that runs well in the accounts a senior CSM handles, and on instinct everywhere else. Most of the time, the senior CSM isn't running the motion. She's compensating for what the motion doesn't do. Three years of partner relationships, institutional memory, judgment about which districts need what and when. She fills the gaps the system leaves, and the accounts look clean.
Leadership looks at her book and concludes the system is working. They point to her as evidence that everyone should be able to manage this way. But her accounts aren't proof the system works. They're proof she does. When you put a newer CSM on the same book with the same playbook, you find out what the system actually produces without someone to fill in what it misses.
What honest scoring looks like
Score what fired. Not the playbook. Not the good accounts. What happened across the full book, last cycle, when the motion wasn't being held together by memory and senior expertise.
Ask the question separately to the people managing the function and the people running it. Not in the same room, and not at the same time. The gap between those two conversations almost always lands on one or two dimensions. Those are the dimensions where the motion lives in the design but not yet in the practice. They're the ones worth moving first, because closing that gap doesn't require a rebuild. You're simply closing the distance between what was built and what actually fires.
What would the CSMs closest to your book say about the one dimension your leadership team is most confident in? The gap between those two answers is where the diagnostic starts.
June 29, 2026
Most post-sale orgs aren't broken. They're uneven.
Field Notes · No. 4
Most post-sale functions that look broken are only uneven. One or two weak spots get told as a verdict on the whole. The rebuild that follows is the expensive mistake.
When the gross retention number is down or the renewal forecast has slipped more than once, there's pressure in the room to do something big and visible about it. Without a clear root cause, it's easy for a dramatic narrative to take hold: the function is in trouble and needs a dramatic move. A reorg, a rebuild, a new operating model, sometimes a new leader.
Most of the time the big visible move is the wrong one. The function typically isn't broken. It's just uneven. And a dramatic change doesn't fix the uneven part. It shakes the foundation under the parts that were working, and pushes the recovery even further out.
Underneath the miss is usually a function that did lots of things well and needed work in one or two areas. The onboarding motion was solid, the CSMs were good operators, the tech stack further along than expected. And then one dimension — the team couldn't articulate value well enough, or there was no real coverage math behind the assigned book — sat at the bottom of the operational maturity curve and dragged the visible numbers down with it.
The leadership team experienced the bad outcome and reached for a hyperbolic explanation. The actual problem was narrow and fixable. They just couldn't see it, because outcome metrics average everything together. They rob you of context and tell you nothing about where the function actually breaks.
An uneven function fails where it's weakest, not on average across the board.
One weak spot becomes "nothing works"
There's a move that turns an uneven function into a broken one, and it happens in the telling. One dimension is genuinely failing, and the story told about it grows to swallow the rest. The renewal misses get described as "we have no retention motion." The single CSM who can't connect the product to a superintendent's priorities becomes proof that "the team can't sell value." A specific failure turns into an assessment on the whole function.
Beware of absolutes. That absolute is almost always wrong, and it costs you. It erases everything the function does well, which is most of what it does, and it makes the dramatic move sound reasonable…if nothing is working, tearing it all out is defensible. The likely truth, that six things work and one doesn't, points at a smaller fix that's harder to rally a room around.
The narrow diagnosis takes operator judgment. Operating dimensions don't carry equal weight, and they don't fail independently. When I'm scoring a post-sale function, I'm not trying to grade all of it. I'm trying to find one or two gaps that are holding the rest back. Moving those is worth more than moving the other eight, and usually cheaper, because the strong parts are already there to build on the moment that gap closes.
Why the gaps cluster in K-12
The operating dimensions aren't a checklist of unrelated capabilities. They're a chain, and in K-12 EdTech they run in a particular order.
Lifecycle definition feeds value realization. You can't revisit the outcome a district was sold if there's no stage in the lifecycle where that conversation is supposed to happen. Value realization feeds advocacy. A superintendent who can't name what they got is not going to co-present at a conference or take a reference call. And advocacy, as I argued in an earlier note, is the actual sales pipeline in this market. Referenceable districts are how the next districts get sold.
So a weak lifecycle dimension or handoff motion, way upstream, eventually shows up as a thin pipeline, far downstream, and the company misreads it as a broken marketing function.
The academic year stretches the distance between the gap and the symptom. The weak onboarding-to-value handoff happens in September. The renewal it impacts shows up the following spring. The thin referenceability it causes isn't felt until the sales team comes up short two cycles later. By then the pain is three steps and a year from its cause, which is why leadership keeps reaching for "the whole thing is broken." They're staring at a symptom and can't see back up the chain to the gap that produced it.
Sometimes the gap isn't in the post-sale motion at all. A product that requires significant configuration to sustain — one where implementation depends on a single champion walking teachers through it every August — shows up as a usage problem by October. By renewal in April, the account looks like an onboarding failure. The actual gap was in how much effort the product demanded from the district, and it was baked in before the CSM ever took the handoff. The post-sale team looks like the one who lost the renewal. The product team never connected the friction they built to the outcome that followed.
The question worth sitting with
The question worth bringing to evaluate the strength of a function isn't "is it good or bad." That framing produces both the false comfort and the panicked rebuild. The sharper question: which one or two dimensions, improved, would free the others? What evidence supports this beyond the renewal number?
Most teams are guessing. They feel the bad outcome and assign it to the most visible dimension, or to the whole thing, or a single person. The honest version of this work scores the function evenly enough to find the one place that's holding the rest back, fixes that, and leaves the working parts alone.
Isolate the gap.
June 12, 2026
Frameworks aren't the product. Systems are.
Field Notes · No. 3
You can author a framework in an afternoon. Building the system that makes people actually run it takes more reps than the calendar gives you. That distance is the hardest part of this work.
A few years ago I rebuilt an onboarding process to be far more prescriptive. The old version left too much to the individual, and I'd watched that produce wildly different launches depending on who happened to be running them. So I tightened it. Named the steps, sequenced them, defined what "done" looked like at each stage, took the guesswork out. It was, I'll say without false modesty, a good framework.
People still ran their own play.
Not out of defiance. The experienced CSMs had a way that had worked for them before and trusted it more than the new motion. The newer ones reverted to it the first week they got busy. The prescriptive process I'd built was clearly better on paper, and on paper is roughly where it stayed. I had authored a framework. I had not built a system. And I've come to think the distance between those two things is the hardest distance in this work, harder than the strategy, harder than the framework itself.
A framework is one decision. A system is a hundred repeated behaviors.
You write a framework once, in a moment of clarity. A system asks people to do something differently every week, indefinitely, usually when they're busy and the old way still mostly works. That's the gap, and it's not an intelligence gap. It's a behavior-change gap.
Three things live inside it, and I've been caught by all three. The first is the one I most underestimated.
Nothing in the environment forced the new behavior. The framework described what should happen. It did not specify what would trigger each step, or who was accountable when it didn't fire. That is the actual line between a framework and a system: a framework describes what should happen; a system specifies what triggers it and who owns the miss. A motion that runs on memory and goodwill loses to the inbox every time, especially when the cost of doing it the new way is today and the payoff is a renewal eight months out. I had given people a better map and changed nothing about the terrain they were walking. So they walked the path they already knew.
The second: a framework is complete in the author's head and full of holes in everyone else's hands. When I built that onboarding motion, I was unconsciously filling every gap with my own judgment, so it felt finished to me. Handed to someone with less context, all the gaps I'd been silently covering became places to improvise, and people improvise back toward what they already know.
The third: it takes more reps than you think, and more than the calendar gives you. Behavior doesn't become a system until it's been done enough times to be the default. People needed to run the new motion through several launches before it would stop feeling like extra work. We didn't get that many clean reps before the next thing changed.
The mirror
Here's the part that reorganized how I think about this work. The thing I was struggling with internally is the exact thing our customers were struggling with when they tried to adopt our product.
When a district buys your platform, you hand them a framework too. A best-practice implementation model. A PD day. A usage rubric, a rollout plan, a "here's how the high-performing schools do it" deck. And it dies inside the building for precisely the reasons my onboarding process died inside my team. The teachers have an old way that still works. The new way costs effort now and pays off later. Nothing in the school day forces the change. The instructional coach who championed it transfers to another building in July. And changing the behavior of an entire staff takes more than the one school year you get before summer resets the momentum and a new cohort of teachers arrives in August who never saw the original rollout.
A district that "isn't adopting" usually isn't resisting you. They're running their own play, for the same human reasons my CSMs ran theirs. Low adoption is rarely a motivation problem. It's a behavior-change problem that nobody resourced as the multi-year effort it actually is.
Once you see that, two things change. You stop being frustrated that the district hasn't adopted the framework, and you start building the system that makes adoption happen on their calendar: the triggers, the owners, the reinforcement across more than one academic cycle. And you extend yourself the same understanding internally. The CEO frustrated that CS "isn't running the playbook" and the vendor frustrated that the district "isn't adopting" are making the identical error. Both confused authoring the framework with producing the behavior.
Why this is the whole game in K-12
In K-12 EdTech the behavior-change tax compounds, because almost everything that matters runs on the academic year and the academic year gives you one rep at a time. One onboarding spike. One renewal window. A teacher gets a single school year of practice before summer wipes the momentum. So a system, yours or your customer's, needs two or three cycles to become muscle memory, and a single departure, on either side, can reset the clock to zero.
This is why a framework looks more built than it is, and why the gap stays hidden until it's expensive. You can present a clean diagram in a board meeting and have very little that actually fires against it. The diagram is real. The motion underneath it is not. And you don't find out which parts were real until October, when the district that signed in April still isn't fully live and you go looking for whose name was on the step that didn't happen.
In the last two notes I argued that referenceable customers are your real sales engine, and that retention follows from outcomes a district can name. Both of those are things you have to produce reliably, account after account, year after year. You cannot produce them with a framework. A framework gets you one good outcome, from one good person, in one good year. The reliability is the system, and the system is the actual product, not the slide that describes it.
A question to sit with, honestly, is not "do we have a framework." Almost everyone does. It's: what have we actually done to change the behavior the framework asks for, the triggers, the ownership, the reps across more than one cycle, and would the motion still fire next year if the person who built it, and the person who champions it, were both gone?
The honest answer to that, for most functions, is where the real work starts.
June 5, 2026
Customers don't buy products. They buy outcomes.
Field Notes · No. 2
A district never bought your product. They bought a result they could defend to their board. They'll leave the moment someone offers a straighter path to it.
I once watched a district leave us while every dashboard I owned was green.
Logins were up. Feature adoption sat right where we wanted it. The CSM had a standing call the district never moved. On paper, a screaming success. And then, at renewal, the curriculum director told us they were leaving for a competitor whose product, feature for feature, was arguably behind ours.
It took me a while to understand what had happened, because I was looking at the wrong thing. I was measuring whether they used the platform. They were measuring whether their teachers could sit down and intuitively get to the information that mattered, fast enough to act on it before the moment passed. We had given them a tool they used well. The competitor had given them an easier way to get where they wanted to go.
A district logging in every day is a district still trying to get somewhere, and if they never arrive, the logins won't save the renewal. They never wanted the software. They wanted the place it was supposed to take them, and that's what they'll be judging you on. Side note: if we've spoken, you've heard me use the drill bit / hole analogy on this one, borrowed from another CS leader.
The job underneath the metric
But "the outcome" in K-12 is trickier than it looks, because the outcome a district writes into the contract is rarely the one that actually keeps them. Procurement buys a number: a score, a proficiency target, a compliance box that has to be checked for a state report. The real job sits underneath it. A teacher trying to see progress in a kid who's behind. A superintendent who needs something defensible to stand in front of the board and say this is working, this was worth the money.
When you serve the procured metric and miss the real job, you get an account that looks healthy and isn't. The box gets checked. The story the superintendent needs never shows up. That's a false green, the most dangerous reading on the board, because it tells you to relax about the account you should worried about.
The expensive mistake
This is where a lot of companies make an expensive mistake. They decide that because the outcome is what matters, customer success must own the outcome, all of it, manufacturing included. So they load the entire result onto post-sale and wait.
Then a district leaves, and CS takes the blame: for a path sales oversold before the contract was signed, for a path product never built, or for a path product did build but filled with so much friction that adoption stalled and the outcome eroded on the way. Even when the outcome failed upstream, the bill often hits the post-sale team.
You cannot manufacture in isolation an outcome whose upstream inputs you don't control. Sales sets the expectation for what's being bought. Product determines whether the path to it exists, and how much friction sits in the way. Post-sale operates at the end of that chain, and too often gets graded as if it were the whole chain.
Earned retention vs. inertia
Which brings me to the number most leaders give full focus: retention.
Not all retention is the same kind. Some of it is earned. A district renews because you helped them get somewhere they cared about, and they'd do it again. Some of it is inertia. They renew because it's February, ripping out a platform mid-year is a nightmare, and nobody has the appetite for the fight. Both show up as the same line on the same chart. They are not the same asset.
Inertia can masquerade as a moat when it's really a delay. Loyalty in this market runs to the outcome, not to you. The moment a peer credibly offers the same progress by a straighter path, inertia converts to loss. That's how you lose an account that renewed three years running.
Sticky isn't the same as durable.
Durable retention is retention earned on recurring outcomes delivery. It's the kind that survives a competitor with a better pitch, because the district already has the result and the story to go with it. In the last Field Note I argued that renewed isn't the same as referenceable. This is the layer beneath it. Stickiness is what you have before someone tests it. Durability is what's left after they do.
So here's where I land. Customer success owns naming the outcome and orchestrating the organization toward it. It does not own manufacturing that outcome in isolation. The job of the function is to refuse to let the outcome go unnamed and unowned. It says, out loud and early: this is the result this district is actually buying, here is the path to it, here is who has to deliver each piece. Then it holds the whole engine to that.
That's a much harder job than driving adoption. It's also the only version of the job that's honest about what post-sale can and can't control.
Retention follows outcomes. It never precedes them. A district stays because you delivered something they can name, not because they logged in, not because they liked the CSM, not because it was February.
So the question I'd sit with, for your top accounts: can your team name the specific outcome each district would say they're buying, in the district's own words, not yours? Not the contract metric. The job underneath it.
If you can't, that's the gap. It's the first thing the self-assessment is built to find.
May 30, 2026
In K-12, your customer base is your sales team
Field Notes · No. 1
I learned this the way you learn most things that matter — by watching a deal close that I had nothing to do with.
A district in a neighboring state signed. No demo request from a campaign, no SDR sequence, no booth scan from a conference. They signed because their curriculum director had called another curriculum director two states over and asked one question: "Does this actually work?" And the answer was yes.
That call closed the deal. Everything marketing did that quarter was, at best, a warm-up act for it.
If you've sold into K-12, you already know this in your bones. It's a relationship-dense, word-of-mouth-driven market. Superintendents talk to superintendents. Curriculum directors sit on the same state panels, attend the same conferences, share the same regional consortia, and text each other before they'll text a vendor. The buying committee for your next deal is, more often than not, sitting in a room full of people who already use you — or already left you.
The base is the top of the funnel
So here's the thing most growth-stage EdTech companies get backwards: they treat the customer base as the end of the funnel. Something sales closes and hands off, and post-sale is left to defend until renewal. But in K-12, the base is the top of the funnel. Your referenceable customers are your most productive sales reps, and they don't carry quota. They carry credibility.
You can't buy your way past it
You cannot buy your way past this. I've watched companies pour money into demand gen to fix pipeline coverage, and it almost never works — because the problem was never awareness.
I've watched the same companies invest in more sales reps to go hunt more business, while holding the line on — or quietly cutting — the delivery arm that produces the references those reps need to close. You end up paying to fill the top of the funnel from two directions while starving the one part of the business that makes the funnel convert. The reps work harder against colder ground.
It's a brutal way to grow.
Referenceable beats renewed
And "referenceable" is a higher bar than "renewed." Plenty of districts renew out of inertia, switching costs, or because nobody had the appetite to rip out a platform mid-year. A renewal tells you they stayed. It does not tell you they'll vouch for you.
So if you're a leader looking at a retention rate you're proud of, I'd ask you to sit with one question: what portion of that number is actually referenceable? Of the districts renewing, how many would pick up the phone and sell on your behalf?
Because that's the real signal. It's the customer saying, in their own words, to a peer who trusts them, that the outcome was worth it.
A reference is downstream of an outcome
That last part is what people miss. A reference is downstream of an outcome. Not downstream of a relationship, not downstream of a great QBR, not downstream of a CSM the district happens to like. A district will only put its own reputation on the line for you if you helped them deliver something their community cared about — student growth, recruitment and retention, an operational headache you took off their plate. The warmth is nice. The outcome is what they'll repeat on the phone.
Which means the chain runs in exactly one direction, and it runs through the entire GTM function:
Outcome → referenceable customer → pipeline → growth.
And every link traces back upstream. Sales sets the expectation for what outcome the district is actually buying — or sets the wrong one. Product determines whether the path to it exists. Post-sale gets the district there, or surfaces what's blocking the way. The reference at the end is the whole engine's report card, not one team's.
If your new business has stalled and you're questioning where to invest, my first question is about something else: Can you name ten districts that would take a cold reference call this week and sell on your behalf? Not ten that renewed. Ten that would advocate.
If you can't, the leak isn't at the top of the funnel — it's in the base. And the base is built post-sale, one delivered outcome at a time.
Let's stop treating referenceability as a nice-to-have. Treat it as a measurable output of the entire GTM engine — engineered, tracked, and owned across sales, marketing, product, and post-sale, the same way you'd track pipeline or NRR. Because in this market, it is pipeline.
Your customer base is your sales team. The question is whether you're managing it like one.
