The creative sprint is the default unit of work in most agencies. Three weeks. Four people. One brief. A deck at the end. Every modern marketing org runs on them. Almost none of them price them honestly.
This essay is the honest math.
The anatomy of a sprint
A standard creative sprint is, in our experience, roughly the following:
- Kickoff week. Brief review, stakeholder alignment, audience and offer framing, competitive audit. Mostly strategist and creative director time.
- Concept week. Creative pair explores directions, lands 2–3 concepts, produces the first round of comps.
- Refinement week. Stakeholder review, revisions, adaptation to platform and format specs, production of the final deck or asset set.
Small sprints take 60–80 person-hours. Large sprints take 200+. The average across the agencies we've worked with sits around 230–260 person-hours for a single meaningful campaign sprint.
The line-item math
Let's reprice a typical mid-size campaign sprint at 240 person-hours, distributed as follows:
| Role | Hours | Blended rate | Line cost |
|---|---|---|---|
| Strategist | 60 | $145/h | $8,700 |
| Creative director | 45 | $175/h | $7,875 |
| Copywriter | 70 | $110/h | $7,700 |
| Art director | 70 | $115/h | $8,050 |
| Project manager | 30 | $95/h | $2,850 |
| Overhead / tools (15%) | — | — | $5,276 |
| Total cost to the agency | 275 | — | $40,451 |
Numbers vary by geography — a sprint priced identically in Mexico City is roughly 35–45% cheaper; in London, 15–20% more. But the internal structure is consistent: strategy and creative direction are the smallest hour buckets and the highest-cost per hour, and the two creative roles (copy + art) account for more than half of the sprint's total cost.
The client sees one number. The agency sees three.
The client sees a retainer line or a fixed-fee deliverable: “Q2 campaign sprint — $58,000.” They don't see the internal composition. They can't tell that the $58K is covering the strategy time, the craft time, and a margin cushion against rework.
The agency sees three numbers and usually tracks only one of them:
- Cost to deliver — the $40K above, burdened with opportunity cost of the senior operators assigned.
- Cycle time — three weeks from kickoff to final. This is where timelines slip and where margin gets eaten by rework.
- Cost per production-quality artifact — rarely calculated, because nobody wants to compute it. A sprint that produces one hero deck and six platform variants is a sprint whose artifacts cost between $5K and $40K each, depending on definition. The range alone tells you the unit isn't stable.
The unit of measurement that matters
The most useful number an agency operator can hold in their head is cost per production-quality variant at shipping standard. This is not the cost of the hero concept; it's the cost of the asset the client actually runs in market. For most agencies, honest versions of this number look like:
- 2019: ~$2,000 per variant, with platform reuse ratios of 3–4x
- 2023: ~$3,200 per variant, with reuse ratios dropping to 2–2.5x as platform requirements diverged
- 2025: ~$3,800 per variant, and clients now expecting 50–80% more total variants per campaign than in 2019
This is the real spread agencies are living in: roughly 2x the per-variant cost of 2019, driven out of the same retainer pool. The factory side of the P&L is compressing from both ends.
Why clients won't pay more to fix it
Agencies reach for price increases first. The response is almost always the same: the client benchmarks against three competitors, finds one willing to match the old number, and moves a portion of the work. The compression continues.
The real lever isn't price — it's the math underneath it. A sprint that produces 10 production-quality variants at $4,000/variant loses to a sprint that produces 30 at $1,400/variant, even when the second sprint is billed at a higher retainer. The second agency is running a different operation; the client is paying for outcomes, not for craftsmen.
The honest recommendation
If you run a creative agency in 2026 and haven't computed your cost per production-quality variant in the last six months, compute it before the next retainer review. If the number is above $3,000, your operation has a structural problem that headcount won't solve. If the number is below $1,500, the question changes: you have the unit economics; now figure out what to sell alongside it so the judgement track gets priced, not just the factory.
Either way: the worst position is not knowing the number.