Experiment

Tell Us Once → Show Me What You Know

Stop repeatedly asking users for information Defra already holds and reflect Defra’s current understanding back so people can confirm or improve it.

External userNext
KNOWREMEMBERUSEEXCHANGELEARNANTICIPATE

01

Problem to address

Users re-enter information while Defra’s existing records may be stale or wrong. This creates effort, errors and distrust, and it wastes the opportunity for service interactions to improve shared data quality.

02

What the evidence tells us

Farmers do not want to enter the same details repeatedly. People on the ground often hold better information than the central record. Showing people what Defra already knows, and letting them correct it, is a practical way to improve that record.

03

The proposition to explore

The idea

Show people the relevant information Defra already holds, let them confirm or challenge it, and reuse verified corrections rather than asking for the same information again.

04

Hypothesis

If users can see and verify what Defra already believes, they will do less repetitive data entry and high-value corrections can improve the shared record rather than remain trapped in a single transaction.

05

How it could work

Pre-populate relevant information from trusted sources. Show source/date where useful. Let users confirm, challenge or submit evidence for correction. Route changes through appropriate verification before propagating them to systems that rely on the data.

The exchange

Tell us once information exchangeYou, what Defra knows and Defra are connected by a two-way exchange. The user can correct, confirm and enrich information, which then updates what Defra knows.YOUWHAT DEFRAKNOWSDEFRACORRECTCONFIRMENRICH

06

Example: bring it to life

Instead of starting another application with blank fields:

Here's what we know about your farm.

  • these are your parcels
  • these are your agreements
  • these are the environmental constraints
  • this is the land cover we believe exists
  • these are your previous commitments

Then:

Is anything wrong?

The farmer corrects the boundary.

That correction doesn't merely fix the form.

It becomes evidence that improves the shared picture.

Users stop being passive data suppliers. They become participants in maintaining shared evidence.

07

AI opportunity

AI can explain why information is shown, detect inconsistencies and guide a correction. It is not essential to the core proposition, which is data reuse plus a governed correction loop.

08

Potential value

  • Lower user effort
  • Fewer contradictory records
  • Better trust through transparency
  • Continuous improvement of data through real interactions

09

Reduce uncertainty without overcommitting

Smallest meaningful experiment

Prototype one “What we know about your land” view using static data. Test whether farmers understand it, can identify errors and trust the correction process.

10

Open questions and risks

What we still need to learn

  • Must avoid showing data users are not authorised to see
  • Corrections need clear ownership and resolution times
  • Pre-population can create false confidence if source freshness is unclear