Methodology
A benchmark is only as good as what's behind it.
Alex pairs conversational intelligence with a governed calculation engine built on independent market data. Here is what stands behind every verdict.
The data foundation
Observed market expenditure
Alex's benchmarks are grounded in over $8 billion of real marketing expenditure observed across global, regional and local engagements — what work actually cost, not what rate cards claim it should.
Independent rate benchmarks
Hourly rate distributions by market, agency type, agency calibre and role seniority, gathered independently of any single agency group and adjusted for inflation and data recency.
Deliverable delivery models
For each deliverable in our taxonomy, structured models describe the team that typically delivers it — functions, seniority mix and effort — at high, medium and low complexity.
A disciplined taxonomy
Around 85 deliverable types across Strategy, Ideation & Concepting, Master Production and Individual Assets. A shared language for scoping that makes proposals comparable in the first place.
How an evaluation works
Understand the scope
Alex reads your brief and the agency's proposal — documents, spreadsheets, pasted email — and extracts the deliverables, markets, quote lines and assumptions.
Map to the taxonomy
Each piece of work is mapped onto the approved deliverable taxonomy and a complexity level, driven by your campaign's ambition. What doesn't belong (retainer admin, media buying, pass-through costs) is identified and excluded honestly.
Run the governed engine
A controlled calculation engine prices the scope from delivery models and market rate data, applying configured adjustments for geography, reach, regulation, agency calibre and economies of scale. The AI never invents a number — every figure comes from the engine.
Score the confidence
Every result carries a confidence level derived from data coverage, recency, market fit and how cleanly your scope mapped. Thin data widens the range — it never silently sharpens it.
Advise
Alex compares the quote to the benchmark, flags what looks reasonable and what deserves challenge, and drafts the exact questions to put back to your agency.
Governed, audited, honest
Deterministic where it matters
The commercial calculation is governed, versioned and testable. Identical inputs produce identical benchmarks — reproduced against manually worked reference cases.
Auditable end to end
Every evaluation stores its full calculation trace — the data versions, mappings and adjustments behind the number — available for authorised review.
Directionally honest
Marketing services pricing has genuine variance. Alex gives you a defensible range and states its assumptions, rather than false precision to the pound.
Independent by design
Alex's underlying market data is held independently of any agency group and is never disclosed in evaluations — the foundation of a benchmark both sides of the table can trust.
See it on your own proposal.
Start an evaluation