Sign in
Set up for SaaS

AI for insights and analytics

AI analytics for SaaS

Blended net revenue retention is 103 percent, and the newest cohort, 38 percent of ARR, is at 89 because accounts that took more than sixty days to reach their first report retain at 71. AI analytics reads the ARR, usage and CRM files, and answers the head of customer success: where is retention failing, and which renewals are unprepared.

Ask it in your own words

These are the questions SaaS companies ask. Each one maps to a measure our tools compute from your files. The AI model chooses the measure and explains the result; the arithmetic is done by our code, and every total is checked.

The questionThe measure behind itWhat comes back
What is net revenue retention by cohort, not blended?Net revenue retention by cohortNRR by purchase cohort with ARR weight, and gross retention beside it.
Which renewals have low seat utilisation?Seat utilisation before renewalActive seats over licensed seats for accounts renewing in the window, by ARR.
Where is the whitespace, and does it add up?Whitespace reconciled to ARRPotential by account against current ARR, reconciled so the sum matches the plan.
Which accounts have had no touch?Untouched accounts, by ARRAccounts with no logged activity in the period, by ARR and renewal date.
How long does a new account take to reach value?Time to valueDays from contract to the first value milestone by cohort, with retention beside it.
Is every renewal in the next two quarters owned?Renewal calendar coverageRenewals due with an owner and a plan over renewals due, by month.

A conversation, with the figures

Every figure below was computed by a tool from the rows and checked before it was shown. The lines sum; the percentages match; each line opens to the rows that make it.

Where is net revenue retention failing?

ARR lost to churn and contraction in the last four quarters is $3.5m across 3 cohorts, against expansion of 3.1 million. 2024 cohort accounts for $1.9m, 54% of the loss.

2024 cohort$1.9m54% of the total
2023 cohort$1.1m31% of the total
2022 cohort$500k14% of the total
The 3 lines sum to$3.5m0 unexplained

What is different about the 2024 cohort?

3 groups within it account for $1.9m of the $1.9m lost in the 2024 cohort, 100% of it. Accounts that took longer than sixty days to reach their first report retain at 71 percent; those under thirty days retain at 108.

Slow to first report, over 60 days$1.2mretention 71%; 41 accounts
Seat utilisation under 40% at renewal$500k18 accounts
No touch in the 90 days before renewal$200k9 accounts
These 3 are$1.9m100% of 2024 cohort

What does customer success change?

Onboarding, for the newest customers first: a thirty-day first-report target, an owner per account, and a renewal calendar that flags low utilisation ninety days out. NRR by cohort goes into the board pack beside the blended figure.

Customer successPut every 2025 account on a 30-day first-report planThis quarter
Account managersContact the 18 low-utilisation renewals nowThis month
Head of customer successReport NRR by cohort to the boardQuarterly
What is net revenue retention by cohort, not blended?Which renewals have low seat utilisation?Where is the whitespace, and does it add up?Which accounts have had no touch?

The measures behind the answers

Each measure has one formula, one source and one meaning. They are computed per rep and cohort and in total, and every one carries an identity that must hold before it is shown.

MeasureFormulaFromWhat it tells you
Net revenue retention by cohortARR now from customers active a year ago ÷ their ARR then, by start-year cohortSubscription or billing systemWhich cohorts grow and which shrink
Gross revenue retentionAs above, capped at each customer's prior ARRSubscription systemThe floor under the expansion
Seat utilisation before renewalActive seats ÷ contracted seats, against the curve for months since start; accounts under the curve within 6 months of renewalUsage export; contract fileDownsells and cancellations, ahead of time
Whitespace reconciled to ARRPotential seats and products − held, valued at the account's prices; ARR + whitespace = stated potentialSubscription system; account dataRoom to grow, by account
Expansion pipeline against whitespaceOpen expansion pipeline at accounts ÷ whitespace at those accounts; whitespace with no pipelineCRM; whitespace tableWhether reps work the gaps the data found
Untouched accounts, by ARRARR of accounts with no two-way contact within cadence ÷ total ARRCRM; subscription systemRevenue nobody is talking to
Time to valueDays from contract to the usage milestone that predicts renewalUsage export; onboarding recordsFirst renewals at risk from the start
Renewal calendar coverageARR renewing in 180 days with an owner, a plan and a recent contact ÷ ARR renewingContract file; CRMRenewals drifting to the date
Expansion by sourceExpansion ARR from seats, products, usage and price, separatelyBilling with line detailWhether growth is customers buying more or being charged more
ARR concentrationTop ten customers' share of ARR; largest customerSubscription systemDependence

Each measure is worked through, with the export it comes from and what to drop, in Customer base KPIs for SaaS sales teams.

What the AI model does, and what our tools do

The AI model chooses the measure

From your question and the measures declared for SaaS, the model picks the one that answers it, and the period and comparison the question implies.

Our tools do the arithmetic

Deterministic code reads the rows, computes the measure, and checks the identities below. The same question on the same data gives the same answer, every time.

The AI model explains, and cites

The AI model writes the sentence around the result, naming the rep or account behind it. It states no figure that is not in the result, and every figure links to its rows.

The identities that must hold

ARR movementOpening ARR + new + expansion − contraction − churn = closing ARR
CohortsCohort ARR sums to total ARR
SeatsContracted = active + inactive assigned + unassigned
WhitespaceAccount ARR + whitespace = stated potential; account ARR sums to company ARR

What it reads

The exports SaaS companies already produce. Column names are mapped once and the mapping is reused. A file is the way in; scheduled delivery and connections to your systems come with the plan, and every source is listed here.

  • Subscription or billing system
  • Subscription system
  • Usage export
  • Contract file
  • Account data
  • CRM
  • Whitespace table
  • Onboarding records
  • Billing with line detail

Who owns each answer

An answer is a list with an owner and a cadence, or it is a chart nobody works.

MeasuresOwnerCadence
Untouched accounts; renewal calendar coverageAccount managers; head of customer successWeekly
Utilisation before renewal; time to valueCustomer successWeekly to monthly
Expansion pipeline against whitespace; expansion by sourceChief revenue officerMonthly
Retention by cohort; concentrationChief revenue officer; financeQuarterly

Questions SaaS companies ask about AI analytics

Why by cohort rather than blended?

Because a blended 103 percent can be 120 in an old cohort and 89 in the newest, which is 38 percent of ARR. The blended figure hides where the problem is and when it started. NRR by cohort is computed from the ARR file by purchase month, and the ARR weight is shown so nobody mistakes a small cohort for a trend.

How does whitespace reconcile to ARR?

Potential per account is estimated from the sizing you provide, and the tool checks that whitespace plus current ARR equals the potential and that the total matches the plan. A whitespace figure that does not reconcile is reported as such, not quietly rounded.

Can it read our CRM and usage data?

Yes: the ARR file by account and month, the usage export with active seats, the CRM activity log and the renewal dates. Account names can be replaced with identifiers before upload. The tool joins them by the account key you map once.

Does it predict churn?

No. It shows the facts that precede it: utilisation, time to value, touches and cohort retention, per account, with the renewal date. The customer success lead decides which accounts to act on; no model produces a score.

See it on your data

Bring a few thousand rows. The data map opens next, every column mapped once, and the first question is answered in minutes. Free, in your browser, no account.