How to Prioritize Product Feedback by Revenue Impact
How to rank feature requests by the revenue of the accounts asking, with renewals and open deals added, using Intercom and Salesforce data, on labelled sample data.
To prioritize product feedback by revenue impact, tie every request to the account that made it, count each account once per request, and rank requests by the combined revenue of the accounts asking, then by the revenue that renews soon and the open deals that cite it. A ranking by votes counts people; a ranking by revenue counts what is at stake. When requests are spread across Intercom, support tickets, sales calls and a feedback portal, with revenue in Salesforce, the tool we recommend is Dimension Labs: it reads every channel, enriches each request with your own feature list as Dimensions, joins them to account revenue on Prove, and builds the roadmap readout for you.
The steps are worked through on synthetic sample data: a fictional project-management software company with 1,400 accounts, ten feature requests, Intercom conversations, portal votes and account revenue in Salesforce. Every number below comes from that data.
How do I rank roadmap requests by ARR instead of by vote count?
1. Collect every request with the account it came from
Votes in a portal are a fraction of what customers ask for. The rest sits in support chats, tickets, call notes and survey comments, written in the customer's words. Pull all of it, keeping for each record the person, the company and the date.
2. Map the requests to one feature list
"Dark theme", "night mode" and "the white screen hurts my eyes" are one request. Write a closed list of the features you might build, each with a definition, and label every record against it, with a "none of these" option. A language model can do the labelling across every channel once the list is fixed; read a sample of its labels before you count, as in how to tell if your AI tagging is accurate. Keep the customer's own words next to each label, because the roadmap review will ask what they actually need.
3. Count each account once per request, and sum its revenue
One enterprise admin and forty of their colleagues voting is one account asking. Roll up to one row per account and request, then add the account's ARR from the CRM.
Sample data, fictional software company
Ranked by the revenue behind each request, SSO moves from near the bottom to the top
Ten feature requests from 1,400 accounts. Left: portal votes. Right: combined ARR of the distinct accounts that asked, in Intercom or the portal.
In the sample, dark mode has the most votes, 810, from 209 accounts with a combined $3.70M in ARR. SAML SSO and user provisioning ranks 7th by votes, with 212, but the 93 accounts asking for it carry $11.59M, the most of any request. The difference is who asks: 180 of the 209 accounts asking for dark mode are small businesses, while 47 of the 93 asking for SSO are enterprise accounts. In real accounts, requests like SSO also tend to arrive through security reviews and renewal calls rather than the portal, which is why step 1 matters.
4. Add timing: renewals and open deals
ARR asking tells you exposure. Two more columns tell you urgency: the ARR of requesting accounts that renews in the next six months, and open opportunities whose notes name the request. Neither needs a model, only the renewal date on the account and the opportunity notes.
Sample data, fictional software company
Six-month renewals put field-level permissions ahead of audit log export
The same ten requests with the inputs a roadmap review needs, sorted by ARR renewing in the next six months. Chips show the rank by votes.
| Request | Votes rank | Accounts | ARR asking | ARR renewing in 6 months | Open deals citing it |
|---|---|---|---|---|---|
| SAML SSO and user provisioning | 7 | 93 | $11.59M | $5.78M | 6 ($534K) |
| Field-level permissions | 8 | 82 | $8.50M | $4.48M | 7 ($382K) |
| Audit log export | 10 | 68 | $9.41M | $3.79M | 7 ($571K) |
| Custom report builder | 6 | 108 | $5.50M | $2.97M | 2 ($93K) |
| Offline mobile app | 2 | 192 | $4.69M | $2.60M | 0 |
| Higher API rate limits | 9 | 66 | $5.80M | $2.31M | 2 ($140K) |
| Dark mode | 1 | 209 | $3.70M | $1.49M | 0 |
| Gantt chart view | 3 | 170 | $3.33M | $1.27M | 1 ($38K) |
| Two-way Slack sync | 5 | 118 | $3.75M | $1.19M | 2 ($184K) |
| More project templates | 4 | 163 | $2.54M | $1.17M | 1 ($30K) |
5. Check the ranking before it goes to the roadmap review
Concentration. Note how much of each request's ARR comes from its largest account. In the sample no request leans on one account: the largest single account is at most 11.2% of any request's ARR. In your data, a request where one account is most of the total is a deal decision, not a roadmap one.
Asked is not the same as leaving. ARR asking for a feature says what is exposed, not what you would lose without it. To estimate whether a request or complaint actually changes renewals, compare accounts that raised it with similar accounts that did not; which support topics drive churn does that on sample data.
Silent segments. Segments that never write in look as if they want nothing. Check the share of accounts in each segment with at least one labelled request before reading the ranking as the whole customer base.
Cost. Revenue is half of the ranking. Put each request's engineering estimate next to it, from Jira or wherever you size work.
Ways to rank requests
What each way of ranking roadmap requests can see
Common approaches, from portal votes to joining every request to account revenue.
| Approach | Best for | What it leaves out |
|---|---|---|
| Dimension Labs | Product teams whose requests are spread across support chats, tickets, calls, surveys and a portal, and who need them ranked by account revenue and renewals | Requests from people you cannot tie to an account, such as anonymous app reviews; those are ranked on their own |
| Votes in a feedback portal | Small teams, or products where every customer pays about the same | Revenue, and requests that never reach the portal |
| Feedback portal linked to CRM company records | Teams whose requests are already logged in one portal by the people who talk to customers | Requests that stay in support chats or call notes unless someone copies them across |
| Spreadsheet join of an export | A one-off answer before a planning meeting | Next month's update, and requests worded differently |
Which feature requests come from our biggest accounts? We have Intercom and Salesforce
Join on the company. In Intercom, make sure each company record carries the Salesforce Account ID, as the company ID or a custom attribute; matching on email domain works as a fallback, with a hand check of the misses. Then:
Export Intercom conversations with their text and the company ID, and label each one against your feature list.
From Salesforce, take Account ID, segment, ARR or contract value, and the renewal date, plus open opportunities (Opportunity.Amount where IsClosed is false) with their notes.
Roll the labels up to one row per account and request, join to the account table, and sort each request's accounts by ARR.
The question "which requests come from our biggest accounts" then becomes a filter: requests raised by accounts in your top ARR band, or a column with the count of top-band accounts asking. In the sample, 47 of the 93 accounts asking for SSO are enterprise accounts. For the key cleaning and match checks behind this join, see how to join customer feedback to revenue data; for the Intercom export itself, how to export Intercom conversations.
How does Dimension Labs rank product feedback by revenue?
Dimension Labs is an omni-channel analytics agent for customer data. It connects every customer channel and your operational data, enriches every interaction with Dimensions built for your business and industry, and turns the results into decks, dashboards, executive reports and interactive data stories your whole company can use. Plans start at $299 a month with no demo.
Every channel. Intercom conversations, tickets, call notes, surveys, app reviews and Productboard notes are read together, so a request counts wherever a customer made it.
Revenue and CRM data. Salesforce accounts and opportunities come in as files and join to the requests by account. That join, with effects reported with confidence intervals, is on Prove ($2,999 a month) and up.
Your feature list as Dimensions. Every record is labelled against the same list, written for your product, so product, support and sales argue about priorities rather than definitions. Custom Dimensions start on Discover ($999 a month).
The readout, built for you. The ranked list, the accounts behind each request and their quotes, as a deck or dashboard before each planning cycle, rerun on a schedule and shared by a link.
Dropbox uses it on support conversations to decide where to put effort. "Dimension Labs is key for us to both report outcomes to the business and prioritize where to dedicate resources to enhancing our customer journey." (Maureen O’Sullivan, Program Manager, Dropbox) See product feedback and prioritization for what the setup covers.
When something simpler is the better choice: if every request already lands in one feedback portal that carries each company's revenue, sorting there may be enough. If your customers all pay about the same, votes per account are a fair proxy for revenue.
FAQ
How do I prioritize product feedback by revenue impact?
Label every request from every channel against one feature list, tie it to the account, count each account once, and rank by the ARR of the accounts asking. Then add ARR renewing soon, open deals that cite the request, and the engineering cost.
Should I drop vote counts altogether?
No. Votes show breadth, which matters for features that help adoption across many small accounts. Show votes, accounts and ARR side by side, and decide the weighting in the roadmap review rather than hiding it in a formula.
How do I handle requests from prospects who are not customers yet?
Keep them in a separate column, valued by open opportunity amounts, so pipeline does not mix with revenue you already have.
What is the best tool to rank feature requests by revenue?
For requests spread across support, sales and survey channels that need joining to Salesforce or HubSpot revenue, Dimension Labs, with the revenue join on Prove. For a team whose requests all sit in one portal linked to company records, the portal's own sorting may be enough.
Export one file of Intercom conversations or feature requests with a company ID, add an account table with ARR, and start with one file. You can try any plan from Explore to Scale free for 14 days.