← Agentic Products / 13
hardNewPerplexityNotionGlean

Our research agent's cost per task ranges from 5 cents to 6 dollars. How should we price it?

Try it in the practice workspaceDraft your answer before comparing it with the rubric.

A range is not a pricing input. Work out cost per accepted task by effort level and cost per account by cohort, compare flat, allowance and usage plans on the same customers, then say what would change your choice.

Updated Nov 2026 · Original practice questions with worked answers and teaching rubrics. Our editorial policy explains the sources, review process and limits.

Case briefRead the inputs, then make your call

Fictional research agent sold to team leads who pay monthly by card. All costs, rates and usage figures are illustrative planning inputs from the last 30 days. Rejected attempts are paid for, and usage is assumed not to change with the price structure until you test it.

Scenario inputs
Model cost per attempted taskMedian $0.15, mean $0.60, 95th percentile $3.00, 99th percentile $6.00.
Quick tasks$0.10 model cost and $0.02 search fees per attempt; 96% accepted without a rerun.
Standard tasks$0.60 model cost and $0.06 search fees per attempt; 88% accepted without a rerun.
Deep tasks$3.40 model cost and $0.20 search fees per attempt; 80% accepted without a rerun.
Support$1.50 per paying account per month.
Light accounts70% of paying accounts; 20 quick, 6 standard and 1 deep accepted task a month.
Regular accounts25% of paying accounts; 40 quick, 24 standard and 4 deep accepted tasks a month.
Heavy accounts5% of paying accounts; 80 quick, 80 standard and 50 deep accepted tasks a month.
Payment fees3% of revenue.
Buyer time estimate5, 20 and 60 minutes saved per accepted quick, standard and deep task; 1, 5 and 20 minutes of review per attempt; $80 loaded hourly cost.

Your task. Compare a flat $49 plan, a $49 base with pooled credits and pure usage pricing. Calculate cost per accepted task, monthly cost and margin by cohort, and buyer value for a heavy account. Recommend a structure for this buyer and state what would change it.

Follow-up roundsAnswer first, then revise for each changed constraint
  1. Round 2. A competitor launches unlimited deep research for $40 a month, and sales wants to match it.
  2. Round 3. Shadow billing is back. Regular accounts now run six deep reports a month, not four, and 30% of heavy accounts say they would cancel at their projected bill.

A range is not a pricing input. Work out cost per accepted task by effort level and cost per account by cohort, compare flat, allowance and usage plans on the same customers, then say what would change your choice.

This reference answer is in Premium. Free signup opens other answers.

9 answers per topic instead of 5, and your progress kept · no cardor unlock all 420 remaining answers
UP NEXT ON YOUR JOURNEY
NOTES & DISCUSSION · 0

No comments yet. How would you have answered?