Named as a systems archetype in Peter Senge's The Fifth Discipline (1990) and restated in John Sterman's Business Dynamics (2000).
Describes how two efforts competing for one resource drift apart because the early leader keeps winning the allocation, so the gap comes from the rule more than from merit.
Two product lines share one budget, and the early leader takes it all, though the other earns more per dollar
Each quarter the budget tilts toward whichever line led last quarter. At rest, line A's four-point head start becomes a runaway (R1) while line B starves (R2), though B returns more per dollar. Lower the tilt, or protect a floor for the trailing line, and Play again. Illustrative.
Line A's revenue, $M a quarter
Line B's revenue, $M a quarter
Quarter 16 of 16
- Line A's revenue, $M a quarter
- 89.7
- Line B's revenue, $M a quarter
- 11.0
- Budget share to A
- 100%
- R1, A's success
- A leads, so A gets more budget, so A grows and leads by more. No negative (−) links: an even count, so the loop reinforces.
- R2, B's decline
- Every point A gains is a point B loses, so B shrinks, which tilts the next quarter's budget further toward A. Two negative links make this loop reinforce. Two negative (−) links: an even count, so the loop reinforces.
- Budget share to A
- The allocation rule: an even split, tilted toward last quarter's leader. At full tilt, each point of revenue share above half earns two points of budget.
- Line A's budget
- Line A's part of the $10 million a quarter for development and marketing.
- Line A's revenue
- Revenue a quarter. Each budget dollar returns $1.50 over time; without spending, revenue fades 15 percent a quarter.
- Line B's budget
- What is left of the $10 million after line A's share.
- Line B's revenue
- Revenue a quarter. Each budget dollar returns $1.65, more than line A's, and fades at the same rate.
An example
An industrial software firm had two product lines: equipment monitoring, the older one, and maintenance planning, launched two years later. Each quarter $10 million for development and marketing was split by a rule the finance team had set: whichever line led on revenue the quarter before got the larger share, tilted by the size of its lead. Monitoring started at $52 million a quarter to planning's $48 million. Four years later monitoring brought in about $90 million and planning about $11 million, and the board asked whether planning should be shut down.
A strategy director pulled revenue per budget dollar first. Planning had returned more on every dollar it received, in every year. She ran the same four years under an even split, and planning passed monitoring early in the second year. The firm kept the planning line, split the budget evenly for two years, and began allocating on return per dollar spent instead of on last quarter's revenue.
Why it works
The archetype turns a small difference in starting position into a large difference in outcome, through a rule that looks fair. Rewarding last period's winner feels like backing merit, and each single allocation can be defended. The rule, though, feeds its own evidence: the line that gets more budget produces more results, which earn it more budget, and the trailing line's weaker numbers confirm the decision that caused them.
Drawing the two loops around one allocation shows where the result comes from. The test is to replay the history under another rule, an even split or one based on return per dollar. When the gap shrinks or reverses, the rule produced it, and the remedy is to change the rule or to protect a floor for the trailing option long enough to see what it can do.
The thinking behind it
Peter M. Senge Peter M. Senge named success to the successful as a system archetype in The Fifth Discipline (1990). Two activities draw on one limited resource, and the allocation follows past success, so the one that starts ahead receives more, succeeds more, and receives more again, while the other is starved. In the archetype's drawing, two reinforcing loops meet at the allocation, one lifting the leader and one pulling the trailer down, and the book's appendix sets it among the recurring structures a manager can learn to spot.
John D. Sterman John D. Sterman restated the archetype in Business Dynamics (2000), the standard system dynamics textbook, which covers stocks, flows, feedback loops, and delays and the archetypal structures they form. In that frame the allocation rule is a variable in the model like any other, which is what makes it possible to ask how much of the gap between two lines the rule itself produced. His beer game experiments, reported in Management Science in 1989, showed how readily managers misread the feedback their own decisions create.
How to run it
- Name the competitors and the shared resource. Product lines, regions, teams, or people, and the budget, time, attention, or talent they draw on.
- Write down the allocation rule. State exactly how the resource is split today and which past result drives the split.
- Measure return per unit given. Divide each competitor's results by what it received; the leader in totals may trail in yield.
- Replay the history under another rule. Recompute the last few periods with an even split or a yield-based rule, and compare.
- Decide whether the runaway helps. Concentrating on one option is sometimes right; choose it on purpose, with the yield numbers in hand.
- Change the rule if it hurts. Allocate on return per unit, on expected potential, or by a fixed split for a set period.
- Protect a floor for the trailing option. Fund it enough, for long enough, to show what it can do before deciding its future.
Where it goes wrong
The first failure is reading the outcome as proof of merit. The starved line looks weak, the favored line looks strong, and the allocation seems vindicated. Spot it when nobody can show results per dollar for both, and recover by putting yield next to totals in every budget review.
The second is overcorrecting into a permanent even split, which funds a weak option long after the evidence is in; set the protected period and its test in advance. The third is missing the archetype among people. The early star gets the best accounts and mentors, then the best ratings, while equally able colleagues never get the accounts that would show what they can do.
When to use something else
- Tragedy of the Commons. When many users draw on a shared resource with no allocation rule at all, and the resource itself is wearing out.
- Capital Allocation. When the task is setting the whole portfolio's split on expected returns, beyond two competing lines.
- Escalation Archetype. When two parties keep answering each other's moves, each raising the stakes, with no shared budget between them.
Where it sits in SOLVE eX
Go to the source
- The Fifth Discipline, 1990
- Business Dynamics, 2000