CHURN IS DEAD
The Two-Vendor Trick: How CS Platforms Quietly Became Org Charts You Can't Edit
11 min read · Operations
Archive note: This issue predates the evidence ledger introduced in August 2026. Treat uncited benchmarks and examples as editorial analysis, not independently verified findings.
In 2024, Totango and Catalyst merged into one company. Two years on, they still ship as two distinct products under one logo, with two data models, two onboarding paths, two histories of how a "health score" gets calculated.
If your CS team runs on one of them, here's a question worth sitting with before your next renewal: do you actually know which of the two your daily workflow lives in? And if your answer is "does it matter" — that's the whole problem, named in one shrug.
This isn't a Totango story. Gainsight has acquired and bolted on enough products over the years that its data model carries the geology of every deal it ever closed. ChurnZero, Planhat, Vitally each grew from a different first principle about what a customer success motion even is. None of them are neutral. None of them ship empty.
Every one of them ships with an opinion about how your team should work. And the day you adopted it, that opinion quietly started becoming your operating model.
You think you bought a system of record. You bought a system of constraint.
The thing nobody re-examines at renewal
Walk through a renewal cycle for a $200K CS platform. The CS Ops lead pulls usage stats. Someone confirms adoption is "healthy." Finance asks if there's a cheaper tier. Procurement negotiates 8% off. The VP signs.
Nobody asks the only question that matters: how much of how we work exists because we chose it, versus because the tool made it the easiest path?
That question never gets asked because the people who could ask it have stopped seeing the tool. It's furniture now. It's the room they work in, not an object in the room. You don't audit the floor you're standing on.
So you re-sign. And in re-signing, you re-buy a thousand small decisions you never actually made. The vendor's definition of "at-risk." The vendor's idea of what a lifecycle stage is. The vendor's notion of which fields are worth filtering on, which silently became your segmentation strategy.
That's not renewing a tool. That's renewing an org chart you've never been allowed to edit.
Tenancy, not ownership
Most CS leaders believe they own their platform. They own a login and a renewal obligation. The platform owns the shape of the team.
Here's the test that exposes it. List your last five operational decisions about how your CS team works. How you define a healthy account. How you tier your book. What triggers an intervention. What a CSM is expected to do in week one of a new account. How you measure success.
Now check each one: could you change it tomorrow without filing a vendor ticket, waiting on a Professional Services scope, or being told "that's not how the platform handles it"?
Every decision that fails that test isn't yours. It's the vendor's, and you're renting it back from them every year.
The share of your operating model that exists because the tool ships that way, rather than because you chose it, is your Vendor-Default Ratio. You can't read it off a dashboard. You compute it by running the test above across your core operating decisions and counting how many you can't change without permission.
Most leaders have never run it. Which means most leaders are paying rent on their own strategy and have it filed under "software."
The Frozen Layer
The place where this calcifies is what I call the Frozen Layer: the band of your operating model that has hardened into vendor defaults so completely that you've stopped experiencing it as a choice.
It freezes in four places. Here's where to dig.
Definitions: who decided what "healthy" means?
I sat with a CS Ops lead at a data-integration company who walked me through their health score with real pride. Login frequency, feature breadth, support ticket volume, all weighted, all green across the enterprise book.
Then I asked what login frequency meant for their product. It turned out their product was an integration layer. The whole point of it working well was that nobody logged in. It ran in the background, moving data between systems. A customer logging in daily was a customer whose pipeline was breaking.
Three of their largest accounts were green. All three had admins who'd stopped touching the tool because it was finally stable, which the score read as engagement decline, which their playbook routed to "educate the user." Meanwhile the actual risk — that a stable, invisible product is the easiest line item to cut when a CIO goes hunting for spend — was a signal the platform had no field for.
The health score wasn't measuring their customers' health. It was measuring the vendor's generic assumption that logins equal love. They'd inherited a definition built for a seat-based SaaS product and applied it to an integration product where the same signal meant the opposite thing.
That's the Definitions Layer. The most dangerous frozen decisions are the ones that arrive disguised as obvious. "Healthy means engaged." "At-risk means declining usage." "Enterprise means over X seats." These feel like physics. They're product defaults, written by people who never met your customer.
They rebuilt the score around "is the integration moving data without errors" and added a renewal-exposure flag for stable-but-invisible accounts. It took a quarter and a fight with the platform's scoring logic. The point isn't the fix. The point is they'd run for three years inside someone else's definition of their own business and called it a strategy.
Workflow: which motions are right, and which are just easy?
Tools have grain. Some paths are smooth and some cut against the wood, and over time a team migrates toward the smooth ones whether or not they're the right ones.
That's mostly harmless until you notice what it costs. The motion your tool makes easiest becomes the motion your team runs by default, and that default becomes "how we do CS here," and eventually someone writes it into the onboarding deck for new hires as if it were a deliberate philosophy.
The tell: ask a CSM why they run a particular play in a particular sequence. If the honest answer is "because that's the order the screens are in," you've found a frozen workflow. The platform's information architecture became your operating procedure, and nobody chose it. It just had the lowest friction.
The Workflow Layer is where your team's habits and the tool's grain become indistinguishable. Untangling them is uncomfortable, because half the time the easy path was also the right one and you can't tell which half until you look.
Segmentation: are your tiers a choice or a side effect?
How do you tier your accounts? Most leaders will give you a strategic-sounding answer about value, potential, and risk.
Then watch how the tiers are actually built in the system. They're built on the fields the platform lets you filter and group on. ARR band, because there's a field for it. Industry, because there's a dropdown. Renewal date, because it's structured data.
The dimensions that might actually predict who deserves what level of attention — stakeholder fragility, executive sponsor strength, how deeply the product is wired into a critical workflow — those rarely have clean fields, so they rarely make it into the segmentation. You tier on what's filterable, not on what's strategic, and the gap between those two is a decision the vendor made for you.
The Segmentation Layer is where your tool's schema quietly writes your coverage model. You staff your highest-touch motion against the accounts the platform makes easiest to group, and call it strategy. It's data-model convenience wearing a strategy costume.
Exit: how much survives a migration?
Here's the layer that makes the other three real. If you offboarded your CS platform in 90 days and migrated to a blank one, how much of your operating model would survive?
Not the data. The data migrates. I mean the thinking. Your definition of health, expressed cleanly enough to rebuild in any tool. Your tiering logic, written down somewhere that isn't a saved view. Your intervention triggers, documented as decisions rather than as automations nobody can explain.
For most teams the honest answer is: not much. Strip away the platform and you'd discover that a large share of "how we run CS" was never an articulated operating model at all. It was a set of configurations, and the reasoning behind them left the building with the implementation consultant two years ago.
The Exit Layer is the truest measure of your Vendor-Default Ratio. The portion of your operating model that couldn't survive a migration is, by definition, the portion the vendor owns. You've just never had to look at it because you've never tried to leave.
Why this bites hardest right now
For years a frozen platform was an annoyance, not a crisis. Your model could afford to be rigid because the problem it solved was stable. You were retaining customers, and the schema was tuned for retention. Green meant safe.
That's no longer true. The industry-wide data says gross retention is holding while net revenue retention is falling. Customers are staying and not expanding. The problem has moved from "are they leaving" to "why aren't they growing," and those are different questions that demand different definitions of a healthy account, different segments, different motions.
Which means the exact moment your operating model most needs to change is the moment your frozen platform is least able to let it. You need to re-define health to catch the renewed-but-shrinking account. You need to re-tier around expansion potential rather than retention risk. You need new triggers.
And every one of those changes runs straight into a schema that was hardcoded for the problem you used to have.
A frozen platform is a tax you don't notice until the day you need to turn the wheel and find it welded straight.
The audit, before you sign
You don't fix a frozen layer by switching vendors. You'll just freeze again in a new schema. You fix it by separating your operating model from your tool, on paper, before you renew. Run this:
1. Write your definitions down, tool-free. Define healthy, at-risk, and enterprise in plain language, the way they should be true for your product and your customer. Do it on a blank page. Then compare that page to what your platform actually computes. Every gap is a frozen definition you've been renting.
2. Trace five workflows to their origin. Pick five things your CSMs do every week. For each, ask why it runs that way. Sort honestly into "because it's right" and "because it's the path of least resistance." The second pile is your Workflow Layer, and it's bigger than you'd like.
3. Rebuild your tiers from strategy, not from fields. Write the dimensions that should determine coverage. Now check which ones your platform can actually segment on. Where strategy and schema disagree, your schema has been winning, and you've been calling the result a tiering model.
4. Run the 90-day exit test on paper. For each core piece of your model, ask: would this survive a migration to a blank tool? Count what wouldn't. That count, over your total, is your Vendor-Default Ratio. Don't guess at a percentage. Compute it from the list.
5. Renew against the gap, not the discount. Take the frozen items into your renewal conversation. Ask the vendor, specifically, which of them you can change yourself and which require Professional Services. The answer tells you exactly how much org chart you're re-buying, and at what price.
The goal isn't a Vendor-Default Ratio of zero. Some defaults are fine. A platform that imposes nothing imposes no leverage either. The goal is to know your number, so that when you sign for another year you're choosing the constraints instead of inheriting them.
The Totango and Catalyst question I opened with isn't really about Totango. It's about the gap between believing you run a tool and the tool running you, and how invisible that gap stays until someone makes you look.
Most CS leaders will renew this quarter without running this audit. They'll get the 8% discount and the warm feeling of a clean procurement cycle. And they'll re-sign for another year inside a definition of "healthy" they never wrote, a tiering model their data fields chose, and a workflow their CSMs follow because of the order of the screens.
Before you renew, find out which decisions in your CS function are actually yours. My guess is fewer than you'd be comfortable putting in writing. The discomfort is the point. That's the number you've been managing a team by without ever being allowed to see it.
Kuber
By Kuber Sethi · All issues · Subscribe