CHURN IS DEAD
The CSM Title Is Now a Liability. The Rebrand Everyone's Reaching For Won't Fix It.
11 min read · Team Design
Archive note: This issue predates the evidence ledger introduced in August 2026. Treat uncited benchmarks and examples as editorial analysis, not independently verified findings.
The 60-Second Version
Buyers didn't go defensive on "customer success" because of the word. They went defensive because the job quietly became a commercial one wearing a relationship costume, and they can smell it.
Renaming CSMs to Forward Deployed Engineers doesn't retire the credibility debt the role ran up. It just relocates the debt to the renewal, where the customer discovers your FDE has a number too.
This week: run a one-page audit that maps each customer-facing role to what the customer actually experiences before you draft the reorg memo.
Two candidates in two and a half months. That was the entire recruiting pipeline one AI company got against a req for "head of customer success," a story Assembly AI's Ryan Seams told from the stage at SaaStr AI 2026. Same job description. Same scope. Same money. They changed the title to "forward deployed engineer" and the pipeline filled overnight.
If you run a post-sales org, you have already forwarded that anecdote to someone. It confirmed a thing you'd been feeling for a year. The word "success" has gone radioactive with technical buyers, and here is proof that swapping it for a title borrowed from Palantir's deployment teams solves your talent problem in a weekend.
It does solve your talent problem. That part is real and I won't argue it.
What I want to sit with is the second half of the story, the half that doesn't fit on a slide. The candidate side of the funnel filled up. The customer side of the equation did not change at all. And the customer is the one holding the pen at renewal.
The rebrand is a tell. It tells you the role accumulated a debt nobody paid down. A new title is you refinancing that debt at a worse rate and hoping the balance doesn't come due before your reorg gets celebrated.
What actually broke
Here is the thing the discourse keeps getting wrong. It treats "customer success" as a branding accident, a word that aged badly, like calling a role "webmaster" in 2026.
It isn't a branding accident. The title was an accurate label. Somewhere between 2019 and now, the job it described changed underneath the word.
CS started reporting to the CRO. Net revenue retention became the only number anyone tracked at board level. Variable comp got bolted onto expansion. The role that was sold to customers as "your advocate inside the vendor" was now measured, promoted, and paid on how much more it could extract from them.
Buyers figured this out. Not because they read your comp plan, but because they sat across from enough CSMs whose "strategic check-in" always found its way to a seat count by minute forty. They learned that the person introduced as their partner had a quota with their logo on it.
That is what "customer success" now signals to a sharp technical buyer. Not incompetence. Not fluff. A disguised commercial motion pointed at them.
So when you rename the role, you are not fixing what broke. You are choosing a word the customer hasn't yet learned to distrust. They will learn. It takes about one renewal cycle.
A rename is what you reach for when you can't afford the thing the rename is describing.
The broken state, drawn from what I actually see
I run enterprise accounts. Let me tell you what the debt looks like from inside a real book of business, not a composite.
The pattern repeats across companies that sell per-seat or per-license software with expansion targets set by sales. The CSM inherits a large account after a handoff where the salesperson promised outcomes the product can't yet deliver. Week one is a fire drill: a P1 escalation that belongs to support, a renewal that was sandbagged into the CSM's lap, a champion who just left for a competitor.
The CSM's calendar is 80 percent reactive. The "proactive strategic advisor" motion the job description promised happens for maybe the top three accounts, on a good month.
Bobby Cooper's Retention Intelligence data, presented at that same SaaStr event, found that more than 50 percent of CSM activity has zero correlation with retention. Read that carefully. It doesn't mean CSMs are lazy. It means the org structure forces them to absorb work that belongs to support, to sales, and to product, and none of that absorbed work moves the number they're judged on.
Now layer the comp plan on top. This CSM has an expansion target. Every genuine relationship they build gets metabolised by the org into a lever. The moment the customer senses that lever, the relationship stops being a relationship and becomes a negotiation the customer manages carefully.
That is the handoff from trusted to managed. And it is the exact moment the credibility debt got borrowed. Not at the rename. Long before it.
When this org rebrands the role to FDE, nothing about that structure changes. The comp plan survives. The expansion target survives. The customer meets a person in an engineer's title with a salesperson's number, and the gap between the title and the incentive is the widest it has ever been. You didn't close the debt. You disguised it, which makes the discovery worse.
The Credibility Debt: a four-part audit
Before you send the reorg memo, run each customer-facing role through this. It takes an afternoon. It will change what you write.
1. The Commission Question. Does this role have a number the customer can smell? Not "does it have a quota on paper," but does the customer, in the room, feel the pull toward a commercial outcome? If the answer is yes, no title on earth hides it. The FDE title with a renewal target is a salesperson in a lab coat, and the customer will find the coat pockets. If you want the engineer title to mean something, the role either owns a technical outcome with no commercial pull, or it owns commercial with an honest name. The dishonest middle is where the debt lives.
2. The Deliverable Test. Can this person produce an outcome the customer could not have gotten from your sales team? This is the one most orgs fail. Palantir invented the forward deployed engineer to embed a person who wrote code against the customer's actual data and shipped a working thing. The title described a deliverable. When you borrow the title without borrowing the deliverable, you have an empty label. Ask the hard version: of the "value" your CS org claims to deliver, what fraction is shippable technical work versus narration of value that already existed? At most SaaS companies the honest answer is that the shippable fraction is small, and the rest is relationship management describing outcomes the product produced on its own. That's fine. It's just not an engineer.
3. The Handoff Scar. Every place the customer got passed from "trusted" to "sold to" is a place the debt was borrowed. Map them. Sales to onboarding. Onboarding to CSM. CSM to renewals. Renewals back to sales for expansion. Each handoff where the tone shifts from advocacy to ask leaves a scar, and the customer remembers every one. Gainsight's new "Handoff Analyst" agent, launched at Pulse this year, exists because the sales-to-CS handoff is where enterprise relationships go to die. That's a real pain worth automating. But automating the reconstruction of goals doesn't repair the trust that snapped when the customer felt the handoff. The scar is relational, not informational.
4. The Repayment Motion. You don't earn a new title by announcing it. You earn it by shipping the technical outcome first, then letting the customer name what you are. Deliver first, get named second, monetise third. If you invert that order, if you claim the engineer title before you've shipped an engineer's outcome, you're spending credibility you haven't earned and the balance comes due at renewal.
Deliver first, get named second, monetise third. That's the whole repayment plan in six words. Print it above the reorg draft.
What a real FDE org actually costs
Here is the part the rebrand crowd skips, because it's the expensive part.
A true forward deployed engineer is not a retitled CSM. It's a person who can sit inside a customer's environment, understand their data model, and build or configure a working outcome against it. That person is loaded expensive. SaaStr's own operating guidance and the market for the profile put it well north of what a relationship-first CSM costs, which is exactly why most orgs can't afford to make everyone one.
So they don't. They ration. The real FDEs go to accounts above some ACV or headcount threshold, and everyone below the line gets the title without the substance, or gets a Slack channel and an automated check-in.
The disciplined version, the one that actually works, is the one every complex AI deployment SaaStr runs well shares in common: hire one strong FDE, embed them with your top handful of accounts, document exactly what they do during deployment, systematise it into a playbook, then let CSMs or implementation specialists run the proven process. The CSM's job becomes maintaining the relationship after the deployment is genuinely solved, not heroically pretending to solve deployments they can't.
Notice what that model does to the four audit dimensions. It gives the FDE a real deliverable. It removes the commercial pull from the technical role. It repairs the handoff scar because the person who shows up actually ships something. And it earns the title in the right order.
The counter-pattern, mass-converting your CSMs into pseudo-engineers overnight, tanks both roles. You lose the relationship coverage you had and you don't gain the implementation depth you claimed. You get a book of business full of people with an engineer's title, a salesperson's quota, and a CSM's toolkit, which is three jobs and zero credibility.
The two markets you're trading between
The rebrand wins one market and loses another. The recruiting market and the customer market are not the same market, and a title behaves differently in each.
In the recruiting market, "forward deployed engineer" is a promotion. It reads as technical, senior, adjacent to product. Candidates who wouldn't touch a CS req apply. Your funnel fills. Real win.
In the customer market, the title makes a promise. It says: this person ships. When the person doesn't ship, the title becomes a liability that the plain word "CSM" never was, because "CSM" promised less and therefore disappointed less.
Most orgs running this reorg right now are about to win the recruiting market and lose the customer market, and they'll call the whole thing a transformation. The recruiting win shows up this quarter, visible, celebrated, attributed. The customer loss shows up two renewals later, diffuse, hard to trace, by which point the person who ran the reorg has moved on with the win on their record.
That lag is why credibility debt is dangerous. Like all debt, the person who borrows it is rarely the person who repays it.
Do this before you rename anything
1. Run the audit on paper. Take every customer-facing role and score it against the four dimensions. Be honest about the Deliverable Test. If the role can't produce an outcome the customer couldn't get from sales, no engineer title will survive contact with the customer.
2. Find your handoff scars. Walk one real account's journey from signed contract to first renewal and mark every point the tone shifted from advocacy to ask. Those are your debt origination points. Fix the worst one before you touch a title.
3. Decouple the number from the technical role, or keep the honest name. If you want an FDE title to mean something, that role cannot carry a renewal target the customer can feel. Pick one. Commercial with an honest name, or technical with a real deliverable. Never the dishonest middle.
4. Hire one real FDE and watch what happens at renewal. Before you convert the org, embed a single genuine forward deployed engineer with your top accounts and measure the renewal outcome against your CSM-covered accounts. Let the data tell you whether you're buying a deliverable or a costume.
5. Let the customer name the role. Ship the outcome first. When a customer refers to your person as "basically our engineer," you've earned the title. Put that in the reorg memo instead of the other way around.
The honest version of this transformation is harder and slower and more expensive than the rename. It requires you to build a deliverable, not borrow a word. It requires you to touch the comp plan, which nobody wants to touch. It requires you to pay down the debt instead of refinancing it.
But here's what I'd ask the leader with the reorg memo half-written. Your candidates went defensive at "customer success" for a reason that had nothing to do with the word. What makes you think your customers will stay comfortable with the new word once they learn what it's incentivised to do?
Rename the role if you want to fill the funnel. Just don't confuse a full funnel with a paid-down debt. The customer keeps the ledger, and the customer collects at renewal.
By Kuber Sethi · All issues · Subscribe