Quick summary
A customer feedback loop, the short answer: it is the process from a customer telling you something to that same customer hearing what you did about it. It runs in five steps: collect, acknowledge, route, act and close.
Closing the feedback loop: the loop is closed only when the customer is told the outcome: the bug is fixed, the feature shipped, or you decided not to build it, and why.
Where it breaks: at the handoff. The fix ships in Linear or Jira, the ticket in your help desk never hears, and the customer finds out from the release notes, or not at all.
Customer Feedback Loop at a Glance
| Step | What happens | Usual owner | Where it lives | How it fails |
|---|---|---|---|---|
| 1. Collect | Feedback from tickets, chat, calls and surveys lands in one place | Support | Help desk | Feedback stays in Slack threads and inboxes |
| 2. Acknowledge | The customer is told it landed and what happens next | Support | Help desk | Silence, or a reply that promises a date |
| 3. Route | The actionable part goes to the team that can act on it | Support, tier 2 | Help desk to tracker | Pasted into Slack, retyped in the tracker |
| 4. Act | Engineering or product fixes, builds or declines | Engineering, product | Issue tracker | The decision never makes it back |
| 5. Close | The customer hears the outcome | Support | Help desk | The bug ships and nobody tells the ticket |
Before you redesign anything, check three things:
Answer these first
- Can you list every customer who reported a given bug? If the answer lives in someone's memory, the loop cannot close.
- Does support find out when an engineering issue is done without asking? If someone has to check the tracker, it will be checked late.
- Who sends the closing reply? "Support" is not an answer. The ticket owner is.
In this article
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
What Is a Customer Feedback Loop?
A customer feedback loop is the full circle from a customer giving you feedback to that customer hearing what happened to it. Intercom describes it as the process between a customer providing feedback and a brand acting on that feedback and telling the customer. The last part is what makes it a loop rather than a funnel: without it, feedback goes in one end and nothing comes out the other.
The reason it matters is trust. In Microsoft's 2017 State of Global Customer Service Report, 77% of respondents said they have a more favorable view of brands that ask for and accept customer feedback. Only 52% believed most brands actually take action on it. That gap is the open loop: customers are willing to tell you, and about half of them doubt it goes anywhere.
For a B2B SaaS company the loop has a specific shape. The customer talks to support, in a help desk such as HubSpot. The fix happens in engineering, in a tracker such as Linear or Jira. The loop has to cross from one tool to the other and back again, and the trip back is the part nobody owns. Our customer feedback management guide covers the wider process of collecting and categorizing feedback. This post is about getting the loop closed.
The 5 Steps of a Customer Feedback Loop
- 1
Collect: one place for every channel
Tickets, chat, calls, shared Slack channels, sales notes and surveys all produce feedback. Collecting means it lands somewhere searchable and attached to a customer, usually a ticket in your help desk, not a screenshot in a DM.
- 2
Acknowledge: tell them it landed
Reply the same day with what happens next. "We've reproduced this and logged it with engineering" is a good acknowledgment. "We'll have this fixed by Friday" is a promise you don't control. If you can't commit to a date, don't give one.
- 3
Route: get it to the team that can act
Questions stay in support. Bugs go to engineering with steps to reproduce. Feature requests go to product with the customer and the use case attached. The key move here is linking: one issue per problem, and every ticket that reports it linked to that issue, so the issue knows who asked.
- 4
Act: fix, build or decide
Engineering fixes the bug, product builds the feature or decides not to. All three are outcomes worth telling the customer about. "Not planned" is a decision, not a failure to act.
- 5
Close: tell the customer the outcome
Reply to every customer linked to the issue, the day it ships or the day the decision is made. Name what changed and what they should do now. This is the step that turns feedback into trust, and the one most teams skip.
Most teams are strong on step 1 and weak on step 5. That's not because anyone thinks closing the loop doesn't matter. It's because step 4 happens in a tool support doesn't watch, so step 5 depends on someone remembering to check.
Closing the Feedback Loop: What Counts as Closed
Closing the feedback loop means the customer who gave the feedback has heard the outcome. Not the team, not the changelog, the customer. Three outcomes each need their own reply:
A bug is fixed
Tell them it is fixed, what they need to do (refresh, re-run the export, nothing), and thank them for reporting it. Reply the day it ships, not when someone notices.
A feature shipped
Tell the people who asked for it before you announce it to everyone. Link to the docs and say how it solves the use case they described. They will be your first and best users of it.
You decided not to build it
Say so, and say why. "We are focusing on X this year" or "this conflicts with Y" is an honest answer. Suggest a workaround if one exists. A clear no closes the loop. Months of silence does not.
One test for whether your loop really closes: pick the last bug engineering fixed that a customer reported. How long after the fix did that customer hear about it? If nobody knows, the loop is not closed by a process. It is closed by luck.
Fixing the bug is engineering's job. Telling the customer is the loop's job, and it only happens if someone owns it.
Closed Loop Feedback: The Inner Loop and the Outer Loop
The term closed loop feedback comes from customer experience programs, most famously Bain's Net Promoter System. Bain splits the work into two loops, and the split maps neatly onto support and engineering.
The inner loop
Bain describes it as giving frontline employees real-time customer feedback, routed to the employees who affected that customer, so they can learn from it and follow up. It is fast and personal: one customer, one reply. In a SaaS company this is the support team answering the person who raised the ticket.
The outer loop
Bain describes it as the loop for problems the frontline cannot fix alone: track trends in the feedback, investigate root causes, launch changes, and communicate the results to employees and customers. In a SaaS company this is the bug or feature request that needs engineering.
The outer loop is where SaaS teams lose customers. The inner loop is in support's hands. The outer loop runs through engineering's tracker, on engineering's timeline, and its final step, communicating the result back to the customer, needs a bridge back to the ticket.
Customer Feedback Loop Examples
Three examples of a customer feedback loop in a B2B SaaS company, one per outcome. They are illustrations, not customer stories.
A bug report
A customer writes in that CSV exports cut off at 10,000 rows. Support acknowledges it, reproduces it and creates an engineering issue with a bug report that has steps to reproduce. Over the next week two more customers report the same thing, and their tickets are linked to the same issue rather than creating two more. The fix ships on a Tuesday. On Tuesday, all three customers get a reply.
A feature request
An account asks for SSO with their identity provider. Support logs the request against the account and links it to the existing feature request, which now shows twelve accounts asking. Product schedules it for next quarter, and support tells each account it is planned without promising a date. When it ships, those twelve accounts hear first, with a link to the setup docs.
A declined request
A customer asks for an on-premise version. Product decides against it. Support replies within the week: not planned, here is why, here is how other customers with the same security requirements handle it. The ticket closes with an answer, not a "we'll pass it on."
Closing replies don't need to be long. Here is a short one you can adapt:
Hi {first name},
Good news: the export issue you reported on {date} is fixed. Exports now include every row, however large the file.
You don't need to do anything. If you re-run the export from {page}, you'll get the full file.
Thanks for reporting it. Your steps to reproduce got it fixed faster.
{your name}
Our bug report reply email templates cover ten cases, from the first acknowledgment to "works as designed" and "won't fix, with a path forward."
Customer Feedback Loop vs Product Feedback Loop
The two terms overlap, and it helps to keep them apart. A customer feedback loop covers everything a customer tells you, including billing complaints, praise and questions. A product feedback loop is the slice that implies a change to the product: bugs and feature requests. It runs through product and engineering, and it is what most "close the loop" advice is really about.
In B2B SaaS the two meet at the ticket. The customer talks to support, so the customer feedback loop starts and ends in the help desk. The product feedback loop runs through the tracker. Closing the first one depends on the second one reporting back. Our guide to managing feature requests from support to engineering covers the request side in more depth.
The Most Expensive Mistake: Closing the Ticket Before the Loop
The most common way to break a customer feedback loop is tidy-looking: support escalates the bug, sets the ticket to closed or waiting, and moves on. The ticket metrics look great. Then the fix ships three weeks later, and nothing is left open that would remind anyone to tell the customer.
Two fixes. First, keep the ticket in a waiting-on-engineering status while the issue is open, so the queue reflects work that is not finished. Second, make the issue's completion something support sees on the ticket without checking the tracker. The second one is a tooling problem, and it is the one most teams solve with a Slack message and a good memory.
How to Run a Customer Feedback Loop in HubSpot and Linear
If your support team runs on HubSpot Service Hub and your engineers work in Linear, here is a setup that closes the loop on every bug and request.
- 1
Collect in HubSpot
Route email, chat and forms into the help desk so every piece of feedback is a ticket attached to a contact and company. If you run surveys, HubSpot's customer feedback surveys (NPS, CSAT and custom) are part of Service Hub Professional and Enterprise.
- 2
Add a waiting-on-engineering status
Add a pipeline status such as "Waiting on engineering" so escalated tickets stay open and visible. Our HubSpot ticket automation guide covers pipelines and statuses on each plan.
- 3
One Linear issue per problem, every ticket linked
Create the Linear issue from the first ticket, and link every later ticket about the same bug to that issue instead of creating a new one. The issue then carries the full list of customers to tell. With Linear's Customer Requests turned on, each account can also show on the issue as a request, so engineering sees how many customers are asking. Our Linear customer requests guide explains how that feature works.
- 4
Bring the status back to the ticket
Make the issue's status visible on the HubSpot ticket, as a property you can filter on, so a saved view of "tickets whose issue is done" becomes support's to-do list.
- 5
Reply to every linked customer the day it ships
The ticket owner sends the closing reply, then closes the ticket. That is the moment the loop closes.
HubSpot has no native ticket sync with Linear, which is why steps 3 and 4 are usually done by hand. This is the job IssueLinker does:
Create or link from the ticket
A card on the HubSpot ticket creates a Linear issue or links an existing one, so the second and third customer join the first issue. It can also add the ticket's company to the issue as a Linear customer request.
Status back on the ticket
The Linear issue's status is written to properties on the HubSpot ticket, so you can filter, report and build views and workflows on it.
Comments both ways
Notes on the ticket and comments on the issue sync in both directions, with loop prevention, so engineering's "fixed in the latest release" reaches support without a Slack ping. With file copying turned on, screenshots in those comments open for people without a login to the other tool.
An email when the issue closes
Optional status emails tell each ticket's HubSpot owner when a linked issue is completed or canceled, for every ticket linked to it. The owner knows to send the closing reply the same day.
Teams that want the ticket to move on its own can also set status rules that move a ticket to a pipeline stage when its issue reaches a Linear status. They are off by default. Our Linear HubSpot integration guide compares every way to connect the two tools.
Close the loop between HubSpot and Linear
IssueLinker creates the Linear issue from the ticket, syncs comments both ways and writes the issue status back to the ticket, so support knows the day a fix ships. $39 per month flat, unlimited users, any HubSpot plan, 14-day free trial.
How to Tell if Your Customer Feedback Loop Is Closing
You don't need a dashboard to start. Track three numbers for a month:
Time to close the loop
For each issue completed this month that had customer tickets linked, the time between the issue being done and the last customer being told. The aim is the same day.
Loops closed
Of the customers linked to completed issues, how many got a closing reply. Anything under all of them is a list of people who think nothing happened.
Open escalations
Tickets in "Waiting on engineering" with no linked issue. Those are loops that can never close, because nothing will ever tell support they are done.
Who This Guide and IssueLinker Are Not For
This guide is about the loop between support and engineering in a B2B software company. If your feedback program is mostly NPS and CSAT surveys, follow-up calls and account management, a customer experience platform built around Bain's inner and outer loops may suit you better.
IssueLinker is for teams that run support on HubSpot Service Hub and build in Linear, Jira or GitHub. It is not for teams whose support runs on Zendesk, Intercom or Front, since Linear connects to those natively, and it does not sync deals, contacts or custom fields between tools.


