ProductSignal
Early access
Changelog

What changed

New in ProductSignal, in the order it shipped. Bigger releases get a note here; everything else lands quietly.

Live studies

Until now a study was a batch: recruit, interview, read the answers, close. A live study keeps going. People are invited as they hit the moment you care about, the interviewer talks to them, and new evidence lands on the topic page as each conversation ends.

Open any study and switch it to live from the header. You set a weekly cap on interviews so the cost stays predictable, and you can pause it at any point without losing the evidence so far.

Recruit from inside your product

Drop the in-app prompt on the screen where the behaviour happens. It asks for two minutes, opens the interview in a side panel, and closes itself if the person says no. Nothing is shown twice to the same user.

Evidence lands as interviews finish

Each finished interview adds its quotes to the topic page, tied to the person who said them and the company they work for. The newest evidence sits at the top and is marked for a week.

Alerts

Point a live study at a Slack channel and it posts each new piece of evidence, with who said it. Live studies are available on Team and Enterprise plans starting today.

EU hosting

Workspaces can now run entirely in the EU. Transcripts, evidence and the models that process them stay in Frankfurt, and nothing leaves the region for processing.

What stays in region

Everything a user says: the transcript and every quote drawn from it. Billing and account details are held separately and contain no interview content.

EU hosting is included on Enterprise and available as an add-on on Team. The DPA has been updated to match; existing customers will receive the new version this week.

Start from a question, not a script

A study no longer needs a written discussion guide to begin. Tell the interviewer what you want to learn in a sentence and it drafts the guide, then adjusts its follow-ups in each conversation.

Get started

Choose Start from a question when you create a study. Write the question the way you would say it to a colleague. The draft appears in a few seconds.

Edit as much or as little as you like

The draft is an ordinary guide. Reorder or rewrite any question before you send it.

Try it before you send it

Every study now has a preview interview you can take yourself before recruiting anyone. It uses the same interviewer and the same guide, and it is not counted against your plan.

AI outreach now starts from each user’s feedback history

We added AI-generated outreach on March 12, 2026, so Product Signal can contact users with messages based on their actual feedback history.

What shipped

  • AI can draft the first outreach email automatically for a selected user
  • The message references the user’s previous feedback directly in the opening note
  • Outreach is tailored to that specific person instead of using a generic survey template
  • Once a user responds, the AI can continue with follow-up questions to go deeper

Why it matters

Generic surveys usually ask broad questions with no context. That makes them easy to ignore.

This new outreach flow starts from what the user has already said. The first email carries their own feedback back into the conversation, which makes the ask more relevant and more personal. The goal is to improve response rates by making the outreach feel specific from the first message.

What comes next

This creates a better discovery loop:

  • Higher-quality outreach rooted in real customer context
  • More follow-up questions without extra manual work from the team
  • Stronger evidence as every reply adds more detail to the customer’s history

Productboard now feeds the customer graph

We added Productboard as a feedback source in Product Signal on February 21, 2026.

What shipped

  • Productboard notes can now be imported directly into Product Signal
  • Feedback from Productboard is linked into the same customer graph as signals from other sources
  • Companies, people, and pains stay consolidated instead of fragmenting by tool
  • The intelligence layer now has a broader evidence base for search, synthesis, and follow-up

Why it matters

Product teams rarely have a single source of truth for customer evidence. Useful feedback lives across call notes, feedback tools, and follow-up conversations.

Adding Productboard means Product Signal can build a more complete map of what each customer is telling you over time. Instead of reviewing isolated exports, teams can see feedback from different systems consolidated around the same accounts, contacts, and recurring pains.

What comes next

This sets up the next layer of customer intelligence:

  • Stronger cross-source consolidation as more signals land in the graph
  • Clearer company and contact histories built from multiple feedback channels
  • Better synthesis on top of a fuller map of customer evidence