What we are trying to do
Most tool directories restate marketing copy and call it a review, which produces pages that are useless at the moment you actually have to choose something. The aim here is narrower and more useful: tell you what a tool is good at, what it costs at real volume, where it breaks, and who should not buy it.
That last part matters. A review that cannot name anyone who should avoid the tool is not a review.
How tools get selected
We review tools that a working marketer would plausibly shortlist. That means a released product with public pricing or a documented sales process, enough public information to assess responsibly, and a clear job it does for marketing teams. Submissions through the submission form are read and assessed against this same methodology, exactly as anything we find ourselves is.
We do not review a tool we cannot describe accurately. If the vendor publishes no pricing, no documentation and no terms, that absence is a finding, not a reason to guess.
The five scoring dimensions
Every tool is scored out of 10 on the same five dimensions, and the overall score is their average. Using one rubric across categories means a 7.5 in email means roughly what a 7.5 in SEO means.
| Output quality | How much of what the tool produces survives a competent review. |
|---|---|
| Workflow fit | Whether it slots into how marketing teams already work, or demands a new process. |
| Ease of use | How quickly a marketer who is not a specialist becomes productive. |
| Value | What it costs against the hours or spend it credibly saves at realistic volume. |
| Support and docs | Documentation quality, transparency of terms and responsiveness when something breaks. |
Two deliberate choices are worth naming. Value is scored against what the tool credibly saves at realistic volume, not against the cheapest competitor, because the cheapest tool that nobody uses saves nothing. Support and docs carries real weight because a tool with unclear terms and thin documentation costs you time every month, and that cost is invisible in a feature comparison.
What the verdict labels mean
Recommended means we would suggest it to a marketer whose situation matches the "buy it if" list without further qualification.
Good with caveats means the tool is genuinely good and there is a specific condition attached, usually price, team size or a limit that catches people out. Read the caveat before buying.
Situational means it solves a narrow problem well and will disappoint anyone who buys it for a general one.
Skip means we found no situation in which we would recommend it over the alternatives.
Hands-on tested versus documentation review
Every review states which of these it is, at the top of the page and in its structured data.
Hands-on tested means we used the tool on real marketing work, not a guided demo, and the scores reflect that use.
Documentation review means the assessment is built from the vendor's public documentation, pricing pages and terms rather than from using the product. These reviews are still original analysis and every factual claim links to its source, but the scores are provisional and we say so on the page. We publish them because an honest documentation review is more useful than nothing, and considerably more useful than a rewritten press release. We do not publish them pretending to be something else.
Sourcing and evidence
Every factual claim about pricing, limits, licensing or terms appears in the sources list at the end of the review, with a link to where we found it. If we could not verify a figure at its primary source, we say that rather than publishing a number we are unsure of. You will find at least one review on this site where we decline to quote a price for exactly that reason.
Vendor pricing changes without notice. Every review carries the date it was last reviewed. Treat anything older than a few months as a starting point and confirm the current numbers with the vendor.
Because different claims go stale at different speeds, each review dates pricing, licensing and capabilities separately rather than presenting one update date for the whole page. Editing a paragraph does not make a price freshly verified, and the fact freshness panel on each review reflects that distinction.
We also run an automated check across every review that finds sentences containing prices, limits, plan names and licence statements, fetches the source cited for each and compares the two. It reports what still matches, what no longer appears at the source, and what it cannot read. It never edits anything: a checker that silently corrects claims it does not understand would create the problem it exists to prevent. Everything it flags is reviewed by a person, and that check is what caught the GetCito pricing error corrected on that review.
What every review must contain
Describing a tool is not the same as helping someone choose one. Beyond the score and the sourced facts, the build refuses to publish a review that does not answer the questions a buyer actually has, so these sections are a requirement rather than a nice addition:
- A thirty second summary covering what the tool is, who it suits, who should avoid it, its single biggest advantage and disadvantage, and whether anything is free. Each line is drawn from findings already sourced further down the page, so the summary can never claim more than the review supports.
- Fit ratings by user type, because a tool that is excellent for an agency is often wrong for a solo marketer. These ratings are our editorial judgement against the review's own findings, not aggregated user scores, and the build rejects a set where every persona receives the same number.
- What it actually costs to run, separating the subscription from the metered usage, the add-ons and the internal time. Sticker price is rarely the real number.
- Questions to ask before choosing it, each with both a yes and a no branch, so the review ends in a decision rather than a summary.
Workflow diagrams are required only where a tool genuinely automates something, and where they appear they never imply we ran the workflow ourselves unless the review says testing was completed.
What a commercial relationship can and cannot change
Where an affiliate link or other commercial arrangement exists, it is labelled on the page it appears on and described in our affiliate disclosure. What such an arrangement cannot change is the assessment: the same five dimensions are applied either way, and no relationship softens a limitation or removes a criticism. We also do not send drafts to vendors for approval before publishing. The broader standards are set out in the editorial policy.
We also do not publish vendor descriptions. Every card and every summary on this site is written by us, which is a slower way to build a directory and the only way to make one worth reading.
Re-reviews and corrections
Reviews are revisited when a vendor makes a material change to pricing, packaging or capability, and periodically otherwise. When a review changes, the last reviewed date changes with it.
If something here is wrong, tell us through the contact page and include the source. Factual errors are corrected promptly and the correction is noted on the page rather than made quietly. Disagreeing with a score is not an error, and we will say so politely, but we will read the argument.