Dimension Labs/Causal Brief
Zendesk · Enterprise base
Dimension Labs · Causal Brief
Zendesk
Enterprise customer base · 80 accounts · eight feedback channels · October 2024 – March 2025
The one thing to know

Every account that left had warned us first, in the support queue, months early. The surveys were the last to know, or never knew at all.

Last year 29 of these 80 accounts left, and $5.5 million in revenue left with them. All 29 had written a warning into a support ticket before they went. Almost everyone complains in support, so volume alone is noise. Two specific sentences told you which complaints were real. Here is the whole chain, in one figure.

VOICESIGNALDRIVEREFFECTSTAKE6.5×more likely to leave$5.5Mwalks out the door80 accounts782 customer messagesOct 2024 – Mar 2025“We’re evaluatingalternatives.”the sentence that sorted themA promise unkept,the same bug twice,a rival in the room.THE CONTROLMeasured vs. who actuallyrenewed: 91% left vs. 14%.n = 23 · p < 0.001Two fifths of the book’svalue, gone in a year.
Read left to right: 80 accounts narrow to the one sentence that predicts a departure, to the operational cause, to a measured effect held against who actually renewed, to the revenue at stake. Every figure is computed from the customer messages; the comparison is the real renewal outcome.
$5.5M
in yearly revenue left with the 29 departing accounts.
Two fifths of the $13.7M book
25 of 29
accounts whose earliest warning was in the support queue, not the surveys.
Surveys never warned for 9 of them
6.5×
more likely to leave after an account wrote "we're evaluating alternatives."
91% vs. 14% · n=23 · p<0.001
The signal, not the noise

01Two sentences sorted the accounts that left from the ones that stayed. Ordinary complaints did not.

WHAT THE ACCOUNT SAIDSHARE THAT LEFTvs. accounts that didn’t36% portfolio avg“We’re evaluating alternatives.”23 accounts said it14%91%6.5× more likely to leave“You didn’t deliver what you promised.”23 accounts said it18%83%4.7× more likely to leave“It was fixed, then it broke again.”19 accounts said it26%68%2.6× more likely to leave“We’re looking at [a named rival].”18 accounts said it27%67%2.4× more likely to leave
Bold bar: the share of accounts that left, among those who wrote each sentence. Thin bar: the share that left among accounts that never wrote it. Every gap clears significance (the top two at better than one in a thousand). The phrase, not the volume of complaints, is the signal.

Sixty-seven of the 80 accounts grumbled in support about something. Listening for unhappiness, on its own, would have flagged most of the book. The accounts that actually left were saying something more specific.

When a customer wrote that they were weighing another vendor, or that something they were promised never arrived, the account left far more often than one that never used those words. The gap is not subtle, and it is measured against what really happened at renewal, not a sentiment score.

Put the two strongest together and they reach deep into the book. Thirty-nine accounts, carrying $6.65 million in yearly revenue, wrote at least one of the two sentences, and 72 percent of them left. That is the population a save team should be reading for, every week. One signal is worth setting aside: nearly every departing account had also written that the same problem failed again, but so had many that renewed, so on its own it flags too many to act on. It is a screen, not a separator.

"We're running a 60 day parallel evaluation of ServiceNow before we sign the renewal."
Expansion note · this account later left
Who saw it first

02The support queue saw it first for almost every account. For nearly a third, the surveys never saw it at all.

THE 29 ACCOUNTS THAT LEFT, BY WHERE THE WARNING CAME FIRST1694support first, then surveysupport onlysurvey firstSUPPORT WAS THE EARLIEST WARNING FOR 25 OF 29For the 9 “support only” accounts, the surveys and health programs never raised a flag at all.Across all 29, the first support warning landed a median of 205 days before the account left.
The 29 accounts that left, by which channel carried the earliest warning. Support was earliest for 25; the surveys were first for only 4; for 9 they never warned at all.

This is the part the current early-warning system gets backwards. For 25 of the 29 accounts that left, the earliest warning came in a support ticket, not a survey. The surveys were first in only 4 cases.

And for 9 of the 29, the survey and health programs never produced a warning at all, even though the account left and the support text had already turned negative. The playbook waits on the survey, and for a third of these accounts the survey simply never spoke.

The reason is structural, not a failing of any one team. A support ticket is written the moment something goes wrong. A survey goes out on a schedule, and a health score only moves once its inputs move. The earliest, most specific account of the problem is sitting in the support text, and almost all of it lives in one channel, shown below.

WHERE THE RISK LANGUAGE SHOWS UPmessages carrying a warningSupport tickets229Product feedback51Onboarding survey24Renewal-health survey23QBR notes23CSAT survey20Expansion notes19NPS survey18
Messages carrying risk language, by channel. The support queue carries roughly ten times what any single survey does, and it carries it sooner.
The honest part

03The health score was not blind. It was late, and it cried wolf.

OF THE ACCOUNTS THAT…29 leftthe score had said28 flagged at risk, but late1 it called healthy51 stayedthe score flagged these too38 also flagged at risk13 quiet
Left: of the 29 that left, the score flagged 28, but later than the support text did. Right: of the 51 that stayed, it also flagged 38. The score sees risk; it does not separate it.

It would be easy, and wrong, to say the dashboard missed these accounts. It did not: it marked 28 of the 29 departing accounts at risk before they left.

The problem is two sided. It marked them later than the support text did, and it marked too much. It also called at risk on 38 of the 51 accounts that went on to renew. A warning that fires on three quarters of your healthy accounts is hard to act on. The support phrasing is both earlier and more selective, which is why it belongs at the front of the early-warning system, not behind the score.

The other side

04The accounts that stayed were just as specific, about the value they were getting.

WHAT THE ACCOUNTS THAT STAYED CREDITEDaccounts (of 51)Agents handling more39Work taken off the team37Automation that deflects tickets23A fast, smooth setup17Handling more volume11Better satisfaction scores7
What the renewing accounts credited, by how many named it. Every one of the 51 named at least one measured result; the strongest are operational, not soft satisfaction.

The same reading that surfaces the leavers surfaces the loyalists, and just as clearly. All 51 accounts that renewed named a concrete, measured result they credited to the platform.

They did not simply sound pleased. They pointed at agents handling more, work taken off the team, a number that moved. That specific, results based praise is the asset worth protecting, and the place to ask for a reference or an expansion.

"Our scheduling accuracy improved a lot. Agents are better staffed at peak and overtime is down 18 percent."
Support ticket · this account renewed
What to do

05Point the early warning at the support queue, and act while the months of lead are still there.

The signal already exists, unread, in the support text. Three moves turn it into retained revenue. Sizing is deliberately conservative: it assumes you save only one account in seven that you reach in time.

1

Watch for the two sentences, not the falling score.

Flag any account whose support text says it is weighing another vendor, or that a promise was not kept. These two phrases reach 39 accounts and $6.65 million in revenue, and 72 percent of the accounts that wrote them left. Saving one in seven of that group is worth roughly a million dollars a year in kept revenue.

Owner: Customer Success operations, working from the support queue.
2

Open the save when the sentence appears, not at renewal.

For these accounts the first support warning landed a median of seven months before they left, and for nearly a third the surveys never warned at all. Route a flagged account to a save owner the day the phrase lands, while there is still time to change the outcome, rather than waiting on a survey that may never come.

Owner: the account's success manager, with an executive sponsor for the largest accounts.
3

Turn down the volume on the health score, and turn up the specificity.

The score flags three in four accounts that go on to renew, so it cannot stand alone as the trigger. Use it as background, and let the specific support phrasing decide who gets a save motion. Keep the "same problem failed again" pattern as a watch list, not an alarm.

Owner: Revenue Operations, in the health model.

How we know this, and what we don't

This brief reads 782 customer messages from 80 enterprise accounts across eight feedback systems (support tickets and chat, the NPS, CSAT, renewal and onboarding surveys, the quarterly business reviews, the expansion notes, and the product feedback), from October 2024 to March 2025. Only messages written by the customer were read; replies from Zendesk staff were not. Every "more likely to leave" figure compares the share of accounts that left among those who wrote a given sentence with the share among those who did not, measured against the real renewal outcome on each account, with a test of statistical significance. The renewal outcome is fact, not a prediction. The timing figures are account level: for each of the 29 accounts that left, the earliest support warning (a competitor or alternative named, a promise unmet, a problem fixed then broken again, a high-effort unresolved issue, or an escalation) was compared to the renewal date and to the earliest survey or health warning.

What to hold lightly. The "same problem failed again" phrase appears in nearly every departing account but also in many that renewed, so it screens rather than separates. Exactly one departing account ($54,000) was rated healthy by the score and still left; that single account is noted, not leaned on, and the larger gap is that the surveys often produced no reading at all. The retained-base result categories overlap, so they are read as the distinct-account view, not as a sum. Account revenue figures come from the account records; the public figures on ownership, pricing, the market, and retention benchmarks are cited from public reporting and are never mixed with the figures drawn from these messages.

Dimension Labs · Causal Brief · ConfidentialPrepared June 2026