River
Y CombinatorBacked by Y Combinator
FREE TEMPLATE

Customer Feedback to Product Template

Two documents and three sheets that rank every theme by the real ARR behind it, not by how many tickets mention it.

Free download  ·  No account needed

Corvent Systems · Q1 2026 feedback register

Dark mode has the most mentions. It has the least revenue behind it.

ThemeAccountsTotal ARRRank by mentions
Auto-escalate idle request, 48hrs5$281,5003rd
Custom fields on intake form3$210,0004th
Bulk tag records9$145,0002nd
Dark mode14$52,0001st

Rows sorted by revenue exposure, the sum of ARR across every distinct account that asked. Sorted by mention count instead, dark mode leads and auto-escalate finishes third, roughly backwards from what a renewal or an expansion actually depends on.

Page one for how to pass customer feedback to product already agrees that revenue should outrank volume: attach the ARR of the requesting accounts, not just a vote count, and sort the backlog on that. Several guides walk through the formula. None of them check that number against the register before a new ticket gets logged, and none follow every historical requester through to notification. A theme closes with one message to whoever happened to be in the room, and after that the record of who else ever asked simply stops mattering.

Corvent Systems ran four themes through its register last quarter. Ranked by mentions, Dark mode leads with 14 accounts and Auto-escalate an idle request past 48 hours sits third with 5. Ranked by revenue exposure, the order inverts: Auto-escalate is first at $281,500 and Custom fields on the intake form is second at $210,000. Bulk tag records is third at $145,000, and Dark mode, the mention-count leader, finishes last at $52,000, just under a fifth of what the five-account theme carries.

Because the intake step checks Frequency and Revenue Exposure before anything is logged, that gap is caught the moment Solveig re-raises it, as a missed notification rather than a second backlog item. Four of the five accounts had already been told; only the smallest one slipped through, which is also the account a mention-count view would rank last. An account that goes quiet in this register for months afterward is the kind of signal a churn review is built to test against a whole cohort, not this one theme alone.

The same register, three views, one caught re-ask

Request Register, Frequency and Revenue Exposure, and Status Feedback Loop from the pack.

Request Register

Corvent Systems, one row per occurrence.

DateSourceAccountARRTheme
2026-01-06TicketAshgrove Veterinary Partners$12,000Bulk tag records
2026-01-12TicketBrannock Freight Co$64,000Auto-escalate, 48hrs
2026-01-14CallCorvindale Insurance Group$70,000Custom fields
2026-01-19TicketWexford Municipal Services$118,000Auto-escalate, 48hrs
2026-01-28Lost dealThistlewood Manufacturing$75,000Custom fields
2026-02-10TicketSolveig Dental Group$22,500Auto-escalate, 48hrs
2026-02-18TicketPetrescu Logistics$31,000Auto-escalate, 48hrs

Every occurrence is its own row, tagged with the account, its ARR and a theme, so revenue exposure can be summed from real rows rather than estimated later.

Frequency and Revenue Exposure

Same register, rolled up by theme and sorted by revenue exposure.

ThemeAccountsTotal ARRStatus
Auto-escalate an idle request past 48 hours5$281,500Shipped Mar 2
Custom fields on the intake form3$210,000Planned
Bulk tag records9$145,000Under review
Dark mode14$52,000New

Sorted by total ARR, not by the accounts column; sorted by accounts instead, Dark mode would sit first and Auto-escalate third.

Status Feedback Loop

Every account that has ever asked for a Shipped theme, and whether they have been told.

AccountARRNotifiedNotified dateNotes
Wexford Municipal Services$118,000Yes2026-03-04First pass
Brannock Freight Co$64,000Yes2026-03-04First pass
Amsterhall Retail Partners$46,000Yes2026-03-04First pass
Petrescu Logistics$31,000Yes2026-03-04First pass
Solveig Dental Group$22,500Yes2026-03-20Missed in first pass, caught on re-ask, not logged as new

Solveig re-raised the same request 18 days after ship, unaware it already existed; the register caught it as a notification gap the same day instead of opening a second backlog item.

What's in the pack

01

Rank by Revenue, Not Mention Count

The method itself: every theme sorted by the real ARR of the distinct accounts that asked, not by how many tickets say the same thing, so a loud, low-revenue theme stops beating a quiet six-figure one.

02

Request Register sheet

One row per ticket, call or lost-deal mention, tagged with the account, its ARR and the theme it belongs to, so revenue exposure can be summed from real rows instead of an estimate.

03

Frequency and Revenue Exposure sheet

Every theme rolled up into a distinct account count and total ARR, sorted by revenue rather than count, with status tracked from New through Shipped or Declined.

04

Theme Summary doc

The narrative version of the top themes by revenue, with two or three verbatims per theme attributed to the account and dated, never paraphrased into one line.

05

Request Format for Product doc

The standard handoff for a theme ready for engineering: the problem behind the feature ask, the account list with ARR and renewal timing, and effort left blank for product to fill in.

06

Status Feedback Loop sheet

Every account that has ever asked for a theme now Shipped or Declined, and whether each has been told. A slipped-through request gets caught here as a missed notification instead of turning into the kind of complaint an escalation has to fix.

How it works

  1. 1

    Send tickets, calls and lost-deal notes

    Whatever you have; River pulls out every distinct request and tags it with the account, its ARR if known, and a theme.

  2. 2

    River builds the Request Register

    One row per occurrence, checked against Frequency and Revenue Exposure first, so anything already Shipped routes to a notification instead of a new item.

  3. 3

    Themes roll up by revenue exposure

    Distinct accounts and total ARR per theme, sorted by the dollar figure rather than the count, so the two rankings can be compared side by side.

  4. 4

    Get the summary and the handoff doc

    Theme Summary with dated verbatims for the top themes, and a Request Format for Product ready to send, revenue exposure stated as a number, not a tier.

Frequently asked questions

Is this template free?

Yes. Two Word documents and three CSV sheets, downloaded as a zip, no signup. River is the optional half: it reads your tickets, calls and lost-deal notes, sorts each into a theme, and checks the register before logging anything as new.

Why does revenue exposure beat a mention count?

Because a mention count rewards whoever is loudest. In Corvent Systems' own register, dark mode drew 14 accounts worth $52,000 combined; auto-escalate drew 5 accounts worth $281,500. Sorted by mentions, dark mode wins. Sorted by the revenue actually behind each theme, it finishes last.

What happens when a theme that already shipped gets asked for again?

It gets checked against Frequency and Revenue Exposure before it is logged, not opened as a new item. On the worked example, Solveig Dental Group re-raised a shipped request eighteen days after ship; the intake check caught it the same day as a missed notification, not a duplicate.

How does a shipped fix reach the customer, not just the backlog?

Status Feedback Loop lists every account that ever asked, so notification goes to all of them, not just whoever asked most recently. If the same account already gets the monthly usage report, a shipped fix is one more line worth adding before it goes out, not something the champion has to notice on their own.

Isn't a duplicate ticket just a minor annoyance?

Not to the account filing it. Repeat contact for the same issue is the single largest driver of the customer effort that erodes loyalty. Nothing a team does to delight the customer once they're already on the ticket outweighs that effort. Catching the re-ask the same day it happens is cheaper than losing the account over it.

Does closing the loop actually change retention, or is it just a nice gesture?

It measures. A field experiment at a financial services firm found that proactively closing the loop with dissatisfied customers cut defections by 40 percent. The largest gains showed up among customers who were never even the ones contacted. An account that stops asking for anything is itself the kind of quiet signal a portfolio-wide health review is built to catch.

How does this fit next to the quarterly business review?

The QBR reviews one account each quarter, evidence-graded, in its own meeting. This register runs continuously across every account, tickets and calls included, and a request still open at review time can carry straight into that meeting's open-issues discussion rather than restarting from a blank page.

Rank feedback by revenue, not by volume

Send your tickets, calls and lost-deal reasons. Get a register sorted by the ARR actually behind each theme, and a loop that closes back to every account that asked, not just whoever asked most recently.

Edit with AI