TuringCorp

DeciderExamples › Business

Recorded run · Business · 2026-09-15

Two products, one team of six. One retains well and grows slowly; the other grows fast and users leave after a month. Keep both, or consolidate?

Two candidate answers were submitted. Decider picked one and reported how close the call was. The reasoning below is the run's own output, published verbatim.

Option A

Keep both and split the team. Each product keeps its own momentum, and we revisit the split in a quarter.

Option B

Consolidate on the one that retains, and put all six on it. One product gets real investment instead of two that each have three people.

SUGGESTION: B
Confidence: 73.3% · recorded 2026-09-15

Consolidating on the product with strong retention is the better choice. Retention is the more reliable signal of product-market fit and durable value; slow growth is usually an optimization or distribution problem, while month-one churn is a structural flaw that more acquisition cannot fix and would only scale a leaky bucket. With only six people, splitting into two three-person teams dilutes execution and risks starving both products, leaving the retaining product under-invested and the churny product still structurally broken after a quarter. Keeping both preserves optionality, reversibility, and some learning from the fast-growing product's acquisition channel, and it defers a hard decision; but that optionality comes at the cost of focus and likely mediocre trajectories. Putting all six on the retaining product gives it real investment to compound from a healthier base, and it is higher-return to fix growth for a product users already keep than to try to repair retention for a product that loses users. Therefore, consolidate on retention rather than split resources.

How to read this

The confidence value is calibrated against outcomes, not a vibe: on JudgeBench, judgements Decider rates 90%+ were correct 99.6% of the time, and the 80–90% band 94.0%. The lowest band flags near-ties where either choice is defensible — an answer, not a failure. Full tables: the product page.

This page is a recorded run published as a worked example. It is not a live call, and nothing here is charged. Raw judgements behind our numbers are published at api.turingcorp.net.

Try your own two options All Business examples

Prefer the other side? Option A is right there above — the point of a calibrated reading is that you can tell a decisive call from a coin flip.