Power Law Performance
Power Law Performance challenges traditional reporting by intentionally removing the majority of wins, so only above-average results stay in view. What remains is too small a sample to scale on, so we install signals as early in the sales cycle as possible to proxy high-value optimization feedback and maximize our revenue return ratios.
Volume and value run in opposite directions.
Check the book before choosing the metrics. If a minority of accounts make up a large majority of the revenue, the book follows a power law. In these revenue models, optimizing for volume-based metrics (cost per lead, cost per sale, cost per anything) will naturally optimize for lower-value accounts.
Get the newsletter: new work on scaling value instead of volume, as we publish it.
Beat the volume KPI, lower the average.
In 1970 the economist George Akerlof described the mechanism in “The Market for ‘Lemons’”: when buyers can’t tell quality apart at the point of sale, the market prices everything at the average, and the above-average sellers quietly exit.1 His example was used cars. Later work found the same adverse selection in insurance, where buyers know their own risk better than the insurer,2 and in lending, where banks can’t tell safe borrowers from risky ones.3 Michael Spence showed how the better side gets noticed: by sending a signal that is costly to fake.4 Akerlof, Spence and Joseph Stiglitz shared the 2001 Nobel prize in economics for this work.5
That research is about consumer markets and the data buyers and sellers hold about each other. Power Law Performance applies the same mechanism to customer acquisition: judged on volume, an acquisition system can’t see value, so it doesn’t buy any.
Volume-based optimizations are naturally biased against value. Improved performance numbers compound negatively: an expanding market with shrinking margins.
- George A. Akerlof, “The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism,” Quarterly Journal of Economics 84, no. 3 (1970): 488–500. Journal, Wikipedia.
- Michael Rothschild and Joseph Stiglitz, “Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information,” Quarterly Journal of Economics 90, no. 4 (1976): 629–649. Journal, Wikipedia.
- Joseph E. Stiglitz and Andrew Weiss, “Credit Rationing in Markets with Imperfect Information,” American Economic Review 71, no. 3 (1981): 393–410. JSTOR, Wikipedia.
- Michael Spence, “Job Market Signaling,” Quarterly Journal of Economics 87, no. 3 (1973): 355–374. Journal, Wikipedia.
- The Nobel Prize in Economic Sciences 2001, awarded to George A. Akerlof, A. Michael Spence and Joseph E. Stiglitz “for their analyses of markets with asymmetric information.” nobelprize.org.
More in the research library: more than 100 works on adverse selection, signals, power laws and the measures teams manage to.
Optimizing for Power Law
Label the book, then read it backwards.
Split the book in half on the average, then split each half again: four tiers, A to D. Then trace the tiers back through the funnel to the earliest signal that predicts them.
- 1
Identify Above Avg.
Tier every account from the statements it produces, on the same rule every month, and mark the ones running above the average.
- 2
Apply Attribution
Attribute each tier back to the sources, campaigns, pages and reps that produced it, to find the early traits that lean toward the top tiers.
- 3
Implement Signals
Put those traits to work as early in the funnel as they show up: in targeting, in lead scoring and in routing to sales.
- 4
Optimize
Move spend toward what produces above-baseline accounts, read R2m as it comes in, and repeat every statement cycle.
The earlier in the funnel a signal can be identified and anchored, the bigger its impact on value.
Aim at value, and the revenue target follows.
It sounds backwards. The way to exceed a revenue target isn’t to chase the revenue number harder. Find the early signal, aim marketing at it, route the accounts that can carry the book to sales, and let each team do its job on the right metrics. Results beat the baseline the plan assumed, margins go up, and the target is reached from underneath.
That’s what ABPRO_ is for: reading investment through to revenue and back, every statement cycle, and flagging what runs above and below baseline while there’s still time to move the money.
The Power Law Performance newsletter
Scale value, not volume.
Ongoing updates on the work we’re putting out: where volume KPIs pick the wrong winner, the signals that sort value from volume, and how to put them in your own data.