AI Clean Up Prediction

I wonder if there is an equivalent of this for the Consumer Insights + Intervention Design field?@Team1001Stories @rorysutherland @dilipsoman pic.twitter.com/GvRGbQZ2RI

— Behavioural Science Club (@BehSciClub) May 26, 2026

'I wonder if there's an equivalent of this for the solution sales field?'

Is your sales organisation tracking your AI usage volume?

Or is it starting to come out the other side, and rowing back on their AI push?

I read about Goodhart's Law being proven yet again across businesses.

Bosses recording how much AI was being used by staff. Prizes for the most on-board.

And yes, that metric became the purpose. "When a measure becomes a target, it ceases to be a good measure". People soon gamed their scores. AI took over. Needless tasks being fed into the machine.

Then subsequent clean-up necessary means tasks actually took longer, with no discernible improvement in quality.

One advocate of this measuring said formulising AI use in this way was like wandering around a building site and looking for sawdust on the floor. No sawdust, no wood being cut, no work being done.

Nice imagery. Completely misplaced metaphor though.

We are not software engineers.

Yet in our realm, it also seems AI deployment's wild west days are coming to an end.

Human-in-the-loop mandates, limiting automation batch sizes and strict data janitors assigned are but three counters I've read emerging from Sales departments.

Those heady hours around the Millennium spring to mind. Unbelievable valuations for countless seemingly implausible dotcoms. The question clanged among my peers; 'but how are they going to make money?'

Where's the productivity boost?

We have brutal stats should we choose.

Margin increase. Close rates sharpened. Cycles shortened.

It's around these kinds of movements we must trace our own AI use impact.