Thinking · November 2026 · Barking Cat series
Why do our people resist the investments we approve?
Because people back what they are rewarded for, and every investment changes what someone is rewarded for. A new channel, tool, model or partner moves something between people: sales, credit, workload, budget or control. The people who lose by it resist, without needing to say so, and from where they sit they are right to. A business case that assumes everyone affected wants it to work has left out the people it depends on.
The resistance is not bad faith. It is a sensible response to how the business pays people, and it lasts until the pay changes.
Rewarding one thing while hoping for another
Steven Kerr named the pattern in a 1975 paper in the Academy of Management Journal, “On the Folly of Rewarding A, While Hoping for B”. His argument was that organisations hope for one behaviour, reward another, and are then surprised to get what they rewarded. Fifty years on, it describes how a retail investment can lose its return before the technology has done anything wrong.
What an investment moves, and what resistance looks like
The table sets out four of the moves. The point is that each one has a loser inside the business, and a business case can be approved without naming them.
| What the investment moves | Who can lose | How the resistance shows |
|---|---|---|
| Sales or credit, to a new channel or partner | The team paid on the old channel’s sales | Customers are not referred; the new channel is described as a rival |
| Workload, so that one team saves time and another adds steps | The team doing the extra steps | Data arrives late or incomplete; work-arounds appear |
| A decision, from a person to a model or a rule | The person who used to make it | Recommendations are overridden or ignored |
| Budget, from a supplier or a team to something new | The supplier or team that loses the budget | The new thing is judged harder than the old one ever was |
Resistance of this kind is quiet. Nobody votes against the investment. It simply takes longer to adopt than the case said, and the return arrives late, smaller, or not at all.
What the delay costs
The cost is easy to work, with invented numbers. A case books £1.2m a year of benefit over three years, £3.6m in all, from month one. If the people it depends on take twelve months to come round, the same three years deliver £2.4m. A third of the case has gone before anything else goes wrong, and none of it shows as a failure: the investment was delivered, and the people around it carried on doing what they were paid to do.
A business case can be written, approved and launched without anyone asking who in the business is paid less if it works. I have yet to meet anyone who campaigned for a smaller bonus.
What a fix has to do
Any fix has three jobs. It has to reward people for the outcomes they influence, so that helping the investment no longer costs the person who helps. It has to be simple enough for the people affected to explain to each other. And it has to be affordable, because crediting the same sale or saving twice costs real money.
The best time to agree it is before launch. After launch, the same agreement has to be negotiated with people who have already spent months on the other side of it.
In my own record, the clearest case was a retailer whose store colleagues were paid commission on their own till: agreeing how the stores would share in online sales took a year and three costed options. The same mechanism runs through the rest of this series, from agency targets to pricing teams.
What to do before the next approval
For each investment, write down what it moves and who loses by it: sales, credit, workload, budget or control. Then write down what the case assumes those people will do about it. If any of them will be rewarded less, the case needs either a way for them to share in it, agreed before launch, or a line for the months it will take them to come round.
The cheaper middle route, where a permanent change to pay is too big a decision, is a temporary one: for the first year, the people affected share the credit for the new channel or tool, and the arrangement is reviewed at the end of it.
How you will know
The measure is adoption against the plan: the share of the expected use, referrals or decisions the investment is actually getting, month by month. The guardrail is the affected team’s own results, so that a fix which buys adoption by hollowing out the team’s core work does not count as a win. Read it at three months and again at twelve.
The decision
The decision for the board is whether the people an investment depends on are rewarded for making it work, and it belongs in the approval. The question for Monday: who in this business is rewarded less if our next investment succeeds?
This rests on one operator’s record and a fifty-year-old paper. It cannot tell you how long the argument takes in your business, only that it is cheaper to have it before launch.
Sources
- Steven Kerr, On the Folly of Rewarding A, While Hoping for B, Academy of Management Journal, vol. 18, no. 4, 1975, pages 769 to 783 (citation as given by Talking About Organizations, episode 57)
Related: The Barking Cat problem; Why does the business report success the profit line does not show?; Why did our approved project never go live?.
Back to Thinking