How Much Does a Statistical Consultant Actually Cost?

The honest answer is "it depends," but that's not a useful answer, so here's what it depends on. Statistical consulting is priced in two main ways: hourly and project-based (a fixed quote for a defined scope of work). Independent consultants and small practices commonly bill somewhere in the $100–$300/hour range depending on seniority and specialization; larger firms and litigation-support statisticians can run well above that. But hourly rates alone tell you almost nothing about what a project will actually cost, because the real driver isn't the rate — it's the scope.

What actually drives the price

• How defined the question is. "Run a mixed-effects model on this clean dataset and produce three figures" is a fundamentally different job than "help us figure out what we can even claim from this data," even if both take a few hours.
• Data condition. Clean, well-documented data with a clear codebook costs less to work with than raw exports from three different systems that need reconciling before analysis can start.
• Novelty of the method. A standard regression is faster to run and defend than a custom simulation, a novel psychometric model, or a method that needs its own justification in a methods section.
• Deliverable format. A slide deck of results costs less than a peer-review-ready methods section, a full reproducible code package, or a document written to survive a funder's own statistical reviewer.
• Timeline. Compressing six weeks of work into one doesn't just cost more — at some point it isn't possible without cutting a corner somewhere, and a consultant who says so is doing you a favor.

Why a fixed quote is usually better than an hourly estimate

Hourly billing puts the schedule risk on you. A fixed quote for a clearly scoped deliverable puts it on the consultant, which is exactly where it belongs if they're the one estimating the work. The tradeoff is that a fixed quote requires an actual scoping conversation first — which is why a real consultation before any quote, not a generic rate card, is the only way to get a number you can plan around.

What to ask before you compare quotes

Two quotes that look far apart often aren't quoting the same thing. Before comparing price, make sure you're comparing scope: does the number include revisions after you see a draft? Does it include a methods write-up your funder or journal will actually accept, or just output? Is reproducible code part of the deliverable, or an extra? A lower number that doesn't include the part you actually need isn't a better price.

If you want a number instead of a range, tell us what you're working on and we'll give you a scoped quote, not a rate card.

One email whenever we publish something new. No spam, unsubscribe anytime.