How to Automate User Interviews

Eight of the right people, and the study can start

User Interviews is where a research team goes when it needs those eight, and needs them by Thursday. It works two ways round. Recruit sources participants from the company's own panel; Research Hub runs a panel of your own, made of your customers, your users and the people who opted in on your site.

What lives in it is a study: the screener that decides who fits, the sessions on a calendar, the incentive each participant is owed, and the participants themselves, with a history of what they have already been asked to sit through.

The people using it are researchers, designers and whoever holds research ops, which in most teams is a researcher with a second job. They work to a deadline set by a product decision that will be made with or without the research, which is why recruiting is the part that hurts. Everything else in a study can move. The sessions cannot.

Most of research ops is cross-checking two calendars

The work around a live study is small, constant and impossible to batch.

Tomorrow's sessions have to be checked against the video tool and the calendar, because a session with no room link is a session that will not happen. Someone has to see who has signed the consent form and who has not. A no-show has to be recorded while anyone still remembers it. Incentives have to actually go out, because a participant who was not paid will tell twenty people. And the study is at target or it is not: if nobody notices that it filled, recruiting carries on and the budget goes with it.

Then the reporting. What ran last week, what is booked, what it cost, written up for a product team that has no login and would not open the dashboard if it did.

User Interviews

Sessions ready before the day starts Tomorrow's sessions matched against the calendar and the video tool, with missing room links and unsigned consent forms listed for you.

Recruiting that stops at target A study reaching its session count flagged the moment it happens, so nobody keeps recruiting into a project that is already full.

A weekly picture without the write-up Completed sessions, next week's bookings and what they cost, tallied into the update the product team will actually read.

Screeners and quotas work inside the study, not around it

The platform automates a great deal of this already, and a team doing it by hand should fix that before anything else.

Screener surveys carry advanced question types and skip logic, so responses sort themselves before a researcher reads them. Quotas keep a recruit balanced and update as people move through the study. The scheduling tool books sessions against rules you set, and incentive distribution can be handled for you rather than by somebody buying gift cards. Research Hub adds a participant CRM with dynamic segments, contact rules that stop the same willing people being asked every month, opt-in forms and email automation, plus a participant API for teams with engineering time to spend.

The ceiling is not the feature list. It is that all of this happens inside User Interviews, and a study does not. The call is in Zoom or Teams. The consent form is in the signing tool. The calendar the researcher lives in is Google or Outlook. Findings go to Dovetail or Notion, questions arrive in Slack, and the spend answers to finance. Nobody gave User Interviews those logins.

The recruiting picture arrives before the standup

Read the calendar, the video tool and the signing app in one pass each morning, and the study starts accounting for itself.

Tomorrow's sessions, checked against the calendar and the video tool, with any missing a room link or a signed consent form listed for the researcher before the day starts. That list is worth more at eight in the morning than at two.

A study hitting its target session count, flagged the moment it happens, so nobody is still recruiting for a project that filled on Tuesday. A no-show noted against the session while the room is still open. Last week's completed sessions, this week's bookings and what they cost, tallied into the update that currently gets written by hand.

Participants whose rest period has just reopened, gathered into a list a researcher can look at when a study needs people like them. Incentives still owed on sessions completed a week ago, listed rather than released, because that is the quiet way a panel loses its goodwill.

The line matters here more than in most software. Gathering runs unattended. Every decision about a person, taking them onto a study, turning them away, releasing what they are owed, waits for the researcher. A participant is not a record.

The decisions about people stay with the researcher

Which leaves the collecting, and the collecting is what eats the morning: two calendars, a signing tool, a video platform and a study page, opened one at a time before anyone has spoken to a user.

WebRun is an agent that works a real Chrome browser, signed in as the researcher. It opens the study, the calendar, the signing tool and the video platform, reads what is on screen, and puts the mismatches in front of a person. It reaches the tools your team already uses, with no integration or API key.

It runs on your schedule in your own private environment, and you can watch a run and stop it. Nothing that accepts a participant, turns one down or releases an incentive happens without you: those are drafted and left. Each workflow below names what it opens.