Numbers people usually beat visionaries because a spreadsheet looks like a fact. “Efficiency” often means moving work to customers, to another department, or to next year - credit concentrated, damage dispersed. If you define a role as its most visible task, you will fire the doorman and keep the hinge. If you define marketing as leads, you will get leads and lose the right to charge. Whatever you measure, you teach. Founders and families can look foolish for a few quarters and own the decade; listed firms that must win every 90 days often win them on the way down.
If that is already obvious, good. The rest of this is what to do on Monday.
Who Measures the Numbers People?
They grade everyone. Marketing. HR. Operations. The visionary with the inconvenient warehouse.
Who, exactly, measures their ROI?
And not by how much money they saved. That is the metric they already know how to play.
Who re-opens their decisions a few months or a few years later and asks:
- Did they save money, or move the headache to customers and chase some of them away?
- Did the cheap supplier actually save money, or buy you faults and hours?
- Did the headcount cut improve profit, or raise theft, burnout, and churn?
- Did they stop a bad investment or a good one that could not promise a return next quarter?
- Did they help the company see where to put more money, or only where to take it out?
Numbers people should not be scored only on cash extracted. They should be scored on the quality of the decisions they forced, the damage those decisions created, and the opportunities rejected because they would not fit in Excel.
The simplest check is the one that almost never appears in the deck:
Does the person who takes credit for the saving also take the bill for the damage?
If the answer is no, you do not have a control environment. You have a talent pipeline for people who are excellent at leaving.
The Hard-Hitting List
Print this. Put it in the next risk or strategy committee. Do not “note it.” Answer it.
1. Map every hero metric to the harm it incentivises
For each KPI that can move a bonus: what will a rational person neglect? Write the neglect down. If you cannot, you do not understand the metric.
2. Ban unpaired savings
No cost line without a named owner for customer, risk, quality, and next-year revenue effects. If that owner is “TBD,” the saving is not approved. It is a press release.
3. Re-define three roles before you automate them
Pick three candidates for “efficiency.” List the unpaid jobs inside the paid job (the doorman test). If the unpaid jobs have no new owner, you are buying an automatic door and a future review.
4. Split marketing into two ledgers
Performance that must convert now. Brand that must still be true in 24 months. Last-click does not get to murder the second ledger because it cannot see it.
5. Pair every target (Goodhart insurance)
Volume with quality. Close with retention. Speed with reopen rate. Unit price with total cost of failure. Growth with the customers you must not milk this quarter.
6. Put a delayed fuse on variable pay
A slice of bonus that only pays if the “win” is still a win after shrinkage, churn, rework, or brand damage would have shown up. If people scream, you have found the business model.
7. Demand one living, ugly, long bet
What are we building that currently makes the quarter look worse than a peer’s and that we will still defend in this room? If the answer is nothing, you are Blue Square in a good mood.
8. Review the numbers function like any other
Not “how much did you save.” How many good investments did you kill? How much pain did you export? Where were you wrong with a lag? This is not anti-finance. It is adult supervision.
9. Ask who is promised the decade
Founder involvement and family control correlate with longer clocks and, in the data, with better long-run returns. Listed boards have to manufacture that clock. Vesting, hold periods, and permission to be unfashionable are not culture posters. They are the product.
10. Run the museum test
Take your current KPI pack. Wind it back to Kodak in the late 1990s, Blockbuster in the mid-2000s, MTV when YouTube was a joke that was not a joke. Would these metrics have made the incumbents look healthy while the floor moved?
If yes, stop congratulating yourselves for being data-driven. You are on schedule.
The Direct Question
Are your current metrics setting you up to fall like Kodak, Blockbuster, or MTV?
Not “could disruption happen to us.” That question is cheap and everyone says yes.
The expensive question is: would our scoreboard have rewarded the people who delayed the response? because delay protected this year’s number, this year’s bonus, and this year’s story to the market?
If you cannot answer that without a speech, the metrics already answered it.
Visionaries are not always right. Numbers people are not always the villain. The villain is a system that pays for a clean quarter with a dirty decade, then calls the people who notice “not commercial.”
You can change the system. You cannot change it with another dashboard that measures the same clock in a prettier font.
- Why the numbers people almost always defeat the visionaries
- The efficiency report that hides the damage
- How to move your company’s work onto your customers (satire)
- The doorman fallacy
- Leads versus the brand you cannot count
- Goodhart’s Law
- The quarter versus the decade
- A KPI audit for boards
Relevant Sources
- Founder’s Mentality performance data — Bain & Company — https://www.bain.com/insights/founders-mentality-barriers-and-pathways-to-sustainable-growth/
- The Family 1000 — Credit Suisse Research Institute (2023) — https://www.allnews.ch/sites/default/files/20230321_credit_suisse_csri_family_1000_2023.pdf
- Goodhart’s Law — https://en.wikipedia.org/wiki/Goodhart%27s_law
- The doorman fallacy — Rory Sutherland, Alchemy — https://en.wikipedia.org/wiki/Doorman_fallacy
I work with boards that want this audit done properly: KPIs, incentives, and the decisions finance is quietly making for the rest of the firm. Contact me.