Method
How we grade
The criteria
Agencies and tools are compared on different questions, because they do different jobs. The questions are the same whatever the channel, from Google and Meta to Amazon, TikTok, email and SEO, so options can be compared side by side.
Tools and apps
- Setup effort
- How long does it take to set up, and can a store team run it day to day without a specialist?
- Data accuracy
- Do the numbers it shows line up with the store, the ad platforms and the other tools it touches?
- Integrations
- Does it connect cleanly to the store platform and the channels a store actually sells through?
- Pricing clarity
- Is the pricing public and easy to understand, and is it clear what changes as the store grows?
- Support
- When something breaks, is help easy to reach and does it solve the problem?
Agencies
- Specialism
- Do they focus on e-commerce, and on the channels, platforms and verticals they say they do?
- Proof of results
- Can they show work and results that can be checked, rather than logos and adjectives?
- Transparency
- Do clients see what is being changed, why, and what it did, in reports they can read?
- Account ownership
- Does the store keep full ownership of its ad accounts, data, pixels and creative?
- Communication
- Is it clear who does the work, how often you hear from them and how fast they reply?
- Pricing clarity
- Are fees, minimum terms and notice periods stated up front?
A guide can add questions that matter for its topic, for example how a feed tool handles Merchant Center errors. When it does, the guide lists them up front.
The grade scale
Letter grades run from A to F and are given to setups and practices: campaign structures, tracking setups, feed rules and the like. A grade is a judgement, not an average. One serious failure, such as tracking that reports sales that did not happen, can pull a grade down on its own. The reason for every grade is written next to it.
- A
Excellent
Does the job well, with no serious gaps. The setup or practice we would want on our own accounts.
- B
Good
Solid and worth keeping. A few limits that matter for some stores but not most.
- C
Average
Works, with trade-offs you should know about. Often a fit for a narrow case only.
- D
Weak
Real problems: wasted spend, unreliable data or missing basics. Replace it when you can.
- F
Fail
Actively harms results or data. Fix it before anything else.
Comparisons, not verdicts on brands
Guides that cover tools and agencies are comparisons. They explain what each option is built for, where they differ and what to check before paying. A product, tool or agency only gets a letter grade on the evidence of first-hand use, and the page says so when it does.
- Facts are dated.Pricing tiers, features and integrations are taken from each vendor's own site, and every guide and listed item shows when it was last checked.
- Unclear facts are left out. When something cannot be confirmed, the page leaves it out instead of guessing.
- Examples are labelled. Worked examples use round, illustrative numbers and say so. They are not client results.
Keeping guides current
Tools change their features and prices, and agencies change their teams. When a fact changes, the guide is updated and its last checked date moves with it.
Listings
- Submitting is free. Every submission is reviewed for fit before it is listed.
- Every listed item shows the date its facts were last checked, and that date moves when they are re-checked.
- Comparisons are written by the editorial team from vendor documentation and first-hand account work.
The full policy is on the editorial policy page. If you think something is wrong or out of date, tell us.