A good Web3 PR agency has direct, recent relationships with the reporters who cover your beat, can name the specific journalists and publications where your story fits, and will show you coverage it earned organically, not just sponsored placements. A bad one will dazzle you with case studies, stall on references, and start charging a retainer before it has written a single pitch. The checklist below takes roughly 90 minutes to run and will save you a wasted three-month contract.
I run fractional PR for Web3, AI, DePIN, and cybersecurity founders, which means I spend a lot of time doing the evaluation work founders should have done before they hired the agency they are calling me to replace. The pattern is consistent: the founder liked the deck, trusted the logo wall, and did not ask the questions that would have revealed whether the team had real access or was reselling hustle and hope. This guide is the framework I use when founders ask me to vet an agency before they sign, and the internal checklist I apply to my own positioning when I pitch against them.
Start with the access-vs-noise test
PR firms in Web3 exist on a spectrum from genuine media operators to press release distributors with a CRM and a prayer. The fastest way to locate an agency on that spectrum is to ask one question in the first call: name the three CoinDesk journalists you have placed in the last 90 days, what the stories were, and whether the placement was earned editorial or paid/sponsored.
If the answer comes quickly and specifically, you are talking to someone with access. If the answer hedges toward "we have relationships across the ecosystem" or pivots to reach statistics and distribution numbers, you are looking at a noise operation. CoinDesk, Cointelegraph, Decrypt, The Block, and Blockworks are the tier-1 editorial outlets in the space. Any agency billing at $15,000 a month or more that cannot name a recent earned placement in at least two of them is selling you something other than media coverage.
The six questions that expose what a Web3 PR agency actually does
Beyond the access test, these are the questions I run through every agency evaluation. They are designed to surface what the team does operationally, not what it promises strategically.
1. Who writes the pitches and who sends them?
At many agencies, a senior operator sells the account and a junior account executive runs it. The senior person's relationships and credibility do not transfer. Ask who will be on your account day-to-day, what their specific beat coverage looks like, and whether you can speak to them before signing. If the person pitching the work cannot tell you the editorial focus of the three reporters they would pitch first for your story, the person running your account is not the person who has access.
2. What does your coverage report look like, and how do you distinguish earned from paid?
Agencies that mix paid placements, sponsored articles, and press wire pickups into a single coverage report are obscuring their actual earned editorial rate. Ask for a sample report and look for the distinction. A sponsored CoinDesk piece costs $15,000 to $50,000 depending on placement. If that is appearing in the same column as a Decrypt editorial, the real earned coverage ratio may be far lower than it looks.
3. Who are your current clients and are any of them competitive with us?
Conflict is real in a small industry. An agency running three Layer-1 infrastructure projects simultaneously has an incentive to distribute reporter attention across them. The same reporter will not write four competing stories on similar protocols in the same quarter. Ask directly whether the agency has clients in your category and how it manages that. Good operators have a clear answer. Agencies that deflect are hiding a conflict.
4. What is your narrative process before you start pitching?
This one separates PR agencies from comms operators. A narrative-first shop will not pitch a single reporter until it has a clear story architecture: who the founder is, what the protocol does that is genuinely new, what the tension or problem is in the market, and why this story matters now. An activity-first shop will start issuing press releases and pitching on week one before any of that exists. Ask what the first four weeks look like. If it is more about output volume than story clarity, you are buying noise.
5. What happens if we get no coverage in month one?
This question reveals the contract structure and the agency's honest expectations. Legitimate operators will tell you that month one is mostly onboarding, narrative development, and relationship-warming, and that the first pitches go out in weeks three and four with meaningful coverage expected in month two onwards. Agencies that promise immediate pickup and then hide behind "editorial timelines are unpredictable" when month one is empty have set wrong expectations to close the deal. Ask what the commitment is and what the escalation process looks like.
6. Can we speak to two current clients at a similar stage?
References in PR are almost universally positive because clients who had a bad experience have already left and are not in the agency's reference list. Ask for clients at a similar fundraise stage or market cap, in a similar vertical, and in a similar news cycle. Ask those clients specifically: what did earned editorial coverage look like month over month, did the agency's pitches evolve as the story developed, and would you re-sign today? The answer to that last question is the one that matters.
Four red flags that predict a wasted retainer
These are the patterns I have seen consistently across the agencies that founders call me to replace. None of them are fatal in isolation, but two or more together is a strong signal to walk.
| Red flag | What it signals | How to test |
|---|---|---|
| No earned vs. paid distinction in coverage reports | Padded metrics, real editorial rate is low | Ask for a sample report, count sponsored vs. organic |
| Senior team sells, junior team runs | Access stays with the seller, not the account | Ask who will be on your account and meet them pre-sign |
| Guarantees placements in specific tier-1 outlets | Either sponsored or a lie; editorial is never guaranteed | Ask if the placements are paid or editorial |
| Narrative process is "we'll start pitching immediately" | Activity-first, no story architecture | Ask what the first 30 days look like specifically |
A fifth one worth naming: agencies that cannot explain the difference between a KOL campaign, an editorial pitch, and a press release distribution are selling you a bundle of activities with no strategic coherence. Each of those things does a different job. An operator who runs them all the same way does none of them well. If you want to understand what the most common Web3 PR agency mistakes look like in practice, that piece goes deeper on the patterns.
The model-fit test: agency vs. fractional vs. in-house
Before you evaluate any agency, the more fundamental question is whether an agency is the right model for your current stage. The answer depends on three variables: budget, narrative maturity, and launch cadence.
| Model | Best stage | Typical cost | Tradeoff |
|---|---|---|---|
| Full agency retainer | Post-Series A, active launch cadence, multiple verticals | $15K–$45K/mo | Broad team, but access diluted across a large book |
| Fractional senior operator | Pre-launch, seed to Series A, narrative-building phase | $5K–$12K/mo | Senior access and focus, limited bandwidth |
| Launch sprint | Specific milestone: mainnet, raise, product launch | $15K–$40K flat | Concentrated effort, not sustainable for ongoing narrative |
| In-house PR hire | Series B+, established brand, ongoing content volume | $120K–$180K/yr fully-loaded | Institutional knowledge, but thin media relationships at hire |
The founders I work with most often come to me after a bad agency experience at the seed or pre-launch stage, where they were paying $15,000 to $20,000 a month for an agency that had a book of 25 clients and assigned a junior associate to their account. The fractional model exists precisely for that stage. If you want the full breakdown of how the models compare in practice, the fractional vs. agency comparison covers it.
What the coverage report should actually look like
Ask any agency you are evaluating to show you a coverage report from a current client (redacted if needed). What you want to see is not a list of publication names. You want to see the reporter's name, the story angle, whether the placement was earned or paid, and the rough timeline from pitch to publication. That timeline tells you something important: a two-week pitch-to-publish cycle in an editorial outlet means the reporter knew the team well enough to move fast. A six-week cycle for a single placement is usually a signal that the relationship was being built from scratch on your dime.
The outcomes I hold myself to: for RARI Chain, 11 tier-1 placements in 24 hours around mainnet. For MANTRA Chain's $11M raise, a CoinDesk exclusive with a Middle East RWA angle that ran before the wire. For Gaia AI, a Forbes "Stripe for AI agents" framing that led into Decrypt and Benzinga, followed by a six-podcast tour. Those results are not normal, and no honest operator should promise them as a baseline. But they are the benchmark that lets me tell founders what a strong campaign actually looks like versus an average one.
Conflict and reporting: what to put in the contract
Two things that rarely make it into the initial conversation but should be in the contract before you sign.
First, conflict of interest. Get a written list of current and recent clients in your vertical and a commitment that the agency will not onboard direct competitors during your engagement without disclosure and your consent. In a small industry where the same twelve reporters cover the same beats, a conflict of interest is not just awkward, it is a structural reason your pitches will not land.
Second, reporting cadence and format. A good agency will commit to a weekly update on pitch status, reporter responses, and story development, plus a monthly coverage report that distinguishes earned from paid. Agencies that resist committing to a reporting format are agencies that do not want to be held accountable for output. Put the format in writing and make it a condition of the retainer. If you want to see how agencies get reviewed in the wild, the Web3 PR agency reviews on Clutch for 2026 surfaces some useful patterns, though read those reviews with the same skepticism you would apply to any self-reported metric.
How to use this checklist before you sign
Run the six questions in the first call. If the answers are specific, current, and honest about limitations, schedule a second call and ask for the reference introductions and a sample coverage report. If the answers are vague, redirect to logos and testimonials, or promise placements in outlets the agency cannot demonstrate it has earned before, stop there. The first call is the agency at its most prepared and most motivated to impress you. If it cannot answer basic access questions on the first call, it will not build the narrative architecture your launch needs once you have signed and the incentive to impress has passed.
The broader landscape of who the legitimate players are, and how they compare on the dimensions that matter, is in the best Web3 PR agencies for 2026. Use this checklist alongside that list, not instead of it. The list tells you who is worth evaluating. This checklist tells you how to evaluate them.
Frequently asked questions
Evaluating your options before you sign? The best Web3 PR agencies for 2026 gives you the shortlist. The fractional vs. agency comparison helps you decide which model fits your stage. The full playbook library covers pricing, pitch guides, and narrative strategy.