Jobcentres kept asking what jobseekers had already done, so the questions were easy to fudge and impossible to act on

the problem

A recurring check-in process only verified past compliance, giving staff no lever to actually change what happened next

background

what everyone would do

The standard jobcentre check-in asked a backward-looking compliance question — had claimants completed at least three job-search activities in the past two weeks — treating the appointment as an audit of past behavior, on the assumption that verifying compliance after the fact would keep claimants accountable.

what they saw

The Behavioural Insights Team saw that a backward-looking question gave the advisor nothing to actually do — it only confirmed whether a box could technically be ticked, with no way to shape what the claimant would do differently going forward. Replacing the question with a forward-looking commitment, specific plans for the next two weeks tied to the claimant's real routine, converted the appointment from a passive audit into an active lever on future behavior.

the move

The pilot replaced the backward-looking compliance question with a forward-looking, written commitment device: claimants stated exactly what they would do over the next two weeks, discussed it with an advisor, and returned to report on and revise that specific plan rather than simply confirming a generic activity count after the fact.

why it works

Having claimants state specific job-search activities they would do over the coming fortnight, tied concretely to their actual daily routine and spoken aloud to an advisor, functioned as a classic commitment device — people follow through more reliably on an explicit, specific plan tied to real triggers than on a vague general intention. At the next appointment, the advisor had something concrete to check against and discuss, giving them a genuine lever to identify where a claimant was struggling and adjust the plan, rather than just confirming a generic activity count. Because each appointment produced a new specific commitment rather than a retrospective checkbox, the intervention compounded across appointments, building an increasingly concrete and personalized job-search routine over time, which is what closed the gap between stated intention and actual action that a backward-looking audit never could.

the payoff

In the initial trial, jobseekers who received the new commitment-device approach were 15-20% more likely to be off benefits within 13 weeks of signing on; a subsequent stepped-wedge controlled trial scaled the intervention to 88,033 participants across 12 Jobcentres (following a 2,383-person pilot at a single site) and confirmed a significant increase in the rate at which claimants exited benefits.

where it breaks

The mechanism depends on the advisor having the time and skill to engage meaningfully with each claimant's specific plan rather than rushing through it as a new formality — a commitment device delivered as perfunctorily as the old compliance question would lose the psychological weight that makes it work. It also requires the stated plan to be genuinely achievable and tied to the claimant's real circumstances; a plan disconnected from actual constraints like childcare, transport or health would fail regardless of how specifically it was stated, since commitment devices work by leveraging intent a person can actually act on, not by manufacturing intent that doesn't match reality. And the scaled trial's own data shows the effect, while real and statistically significant, was a percentage improvement in exit rates, not a guarantee for every claimant — some of the hardest-to-place jobseekers needed additional expressive-writing and resilience-building support layered on after eight weeks, suggesting the basic commitment-device mechanism alone isn't sufficient for claimants facing more than a simple planning gap.

what came after

The Behavioural Insights Team trained roughly 25,000 Job Advisors on the approach, and the commitment-device format developed at Loughton was rolled out to every Jobcentre in the United Kingdom, becoming a standard reference case in behavioural public policy for how reframing a compliance check as a forward commitment can outperform monitoring alone.

references

1. [1]The Behavioural Insights Team Update Report 2013-2015 (Section 1: Economic Growth and the Labour Market — the Loughton/Essex Jobcentre trial)The Behavioural Insights Team, 2015bi.team
2. [2]There should be greater public involvement in deciding what is a legitimate 'nudge'LSE British Politics and Policy Blog (London School of Economics), 2013blogs.lse.ac.uk

keep it

same kind of clever

• 1911 · Aalsmeer Flower Auction (Royal FloraHolland)Aalsmeer sells its dying flowers by racing a price downward instead of bidding one upward
• 1911 · Roald Amundsen's South Pole expeditionAmundsen marked his supply depots sideways to the route, not along it, so a lost party could still walk into one
• 1841 · P.T. Barnum (American Museum)Barnum cleared his overcrowded museum by making the exit look like an exhibit