wurqurBook a scoping call

Consent lapses at six months. The data goes the same day, and the Wurqurs make sure no one has to remember.

Two Wurqurs took that off four people at an HR software company, 250 to 300 times a month.

HR software company · 60 people · six systems

The clock

An HR software company carries a very large signup base, and most of those people forget they are in it. Consent still has to be managed for all of them, and consent expires. When the six-month mark comes round and it has not been renewed, that person’s data has to go that day to keep holding it lawfully.

That is 250 to 300 requests a month, each one meaning find this person across six systems, a minute or two on every record. And the whole thing ran on somebody remembering the deadline was coming.

Four people were losing about a day a week each to it.

They asked us to automate the whole cycle

The answer was no. Some of that cycle is judgment, and judgment is the part you leave exactly where it is. We walked the manual flow instead, step by step, and came back with a plan for the parts that should be automated and a line drawn around the parts that should not.

What we found, and did not touch

Their process was fine and we replaced none of it. What had stopped scaling was around it: the volume, the scatter, and the fact that how a request moved lived in the heads of the people moving it. A new joiner would have taken months to absorb that, and they were hiring.

What we built

Two Wurqurs, and one screen for the people. Rights Handler owns the request. Evidence Clerk owns the proof that it happened.

Rights Handler

Finds the person across every system, routes the decisions, carries out what is agreed, meets the deadline.

  • task
  • audit-entry

Evidence Clerk

Collects the proof, assembles the pack, finds the gaps before an auditor does.

And a screen, which is ours

Every item waiting on a decision sits in one queue, the record already assembled beside it, and a second view shows what has been resolved. Deciding used to mean going to find the context. Now the context is what you are looking at.

What changed

the work in one request

30 minutes

2–3 minutes

4 people

a day a week each, back

1,500 hours

a year, off the team

~$52,000

a year

What used to cost them $4,375 a month now runs on two Wurqurs for $318/month.

Where the humans stayed

Every decision about what may lawfully be erased. The Wurqurs find the person, assemble the record, carry out what is agreed and produce the proof. They do not decide, and the screen exists so that deciding is the only work left in it.

What it connects

Six systems they already ran, one place for the proof, and one screen that is ours.

where the person is

  • Ticketingwhere the request arrives
  • PostgreSQLthe application database
  • Salesforcethe customer record
  • Stripebilling history
  • Google Workspacemail and documents

what runs

  • Rights Handlerfinds the person and acts
  • Evidence Clerkrecords and proves
  • The review screenbuilt by us, where the decisions get made

where the proof goes

  • A dedicated audit logevery action, timestamped, separate from the systems it describes
  • The original ticketthe evidence package, generated on completion

What it took

Four weeks. Two to build, half a week to deploy, and a week running on real cases with the old way still in place.

The same idea, somewhere else

A checklist somebody used to remember, turned into a gate that runs every time.

  • Checks & evidence

Two Wurqurs, Pre-flight Check and Evidence Clerk, run the checklist somebody used to remember.

~$12,000 saved a year for a B2B SaaS company

See how
All eight case studies

Every company has a version of this clock.

Tell us the job. We will tell you whether a Wurqur can hold it.