What is a Pareto frontier?
A Pareto frontier is the set of options, among many, that are not beaten on all counts by any other option. An option dominates another when it is at least as good on every objective and strictly better on at least one. Dominated options can be discarded without regret; the survivors form the frontier, named after the economist Vilfredo Pareto. On the frontier, every move is a genuine trade-off: improving one objective costs you another.
Dominance in one table#
Four project options, scored on two objectives:
| Option | Cost | P(success) | Verdict |
|---|---|---|---|
| A | 10k | 55% | On the frontier (cheapest) |
| B | 14k | 70% | On the frontier |
| C | 15k | 62% | Dominated by B (costs more, succeeds less) |
| D | 22k | 85% | On the frontier (best odds) |
C is simply a mistake: B beats it on both axes. But nothing beats A, B, or D on both axes at once, so the frontier is {A, B, D}, and choosing among them is no longer analysis but appetite: how much success probability is worth how much cash?
That division of labour is the concept's whole value. Computation removes the objectively wrong options; judgement picks among the defensibly different ones. A tool that pretends to pick the single "optimal" point for you is hiding a value judgement inside the maths.
Reading a frontier#
Plotted (cost across, success up), a frontier's local shape tells you where you are:
- Steep: small extra spend buys large probability. Underspending here is the expensive mistake.
- Flat: diminishing returns; you are buying comfort, not odds.
- The knee: the bend between the two, where most risk appetites land.
Frontiers for project plans#
Applying this to planning requires two ingredients. The plan must declare its decisions (the option space: spending tiers, scope levels, runway choices) as first-class objects rather than scattered assumptions (decisions vs tasks). And each option must be evaluated honestly, which for an uncertain plan means a full Monte Carlo forecast per option, with the same random draws across options so differences reflect the decisions rather than luck.
Topolog's Pareto tab does exactly this: it sweeps every combination of the plan's declared decisions, evaluates each against the objectives you choose (cost, drawdown, end balance, duration, probability of success), and highlights the non-dominated set. The essay on spend versus success walks a worked example, and the Pareto docs show the surface.
The portable takeaway, even without tooling: before debating which option to pick, check whether some options are simply dominated. A surprising fraction of project arguments dissolve when option C turns out to be beaten on every axis by option B, and what remains is the real conversation: where on the frontier your risk appetite lives.