How We Review AI Tools
Every score on aisimplr is calculated from five criteria using a weighted formula. No gut feelings, no rounded-up scores to flatter an affiliate partner. Here is exactly how it works.
Five criteria, each weighted by importance.
Each criterion is scored 1–5. The final score is a weighted average out of 5.0.
Does it produce work you can actually use?
We rate accuracy, polish, and whether the output would embarrass you if you shared it with a client or colleague. The most important factor — if the output is bad, nothing else matters.
How long until a beginner gets a good result?
We time the onboarding, note where people get stuck, and score accordingly. A brilliant tool that requires a PhD to operate scores low here.
Does the price match what you actually get?
We score the free tier generosity, the price-per-feature of paid plans, and whether competitors offer the same at a lower price. A cheap tool that does little still scores low.
Does it have the right capabilities for the job?
We score breadth of features relevant to the tool's category, not raw feature count. Bloat does not help the score. Missing features that matter do hurt it.
Does it work consistently over weeks, not days?
We test over four to six weeks — not just on launch day. Uptime, generation consistency, and whether results degrade on free tiers all factor in. A tool that works brilliantly 70% of the time scores lower than one that works well every time.
The exact formula we use.
All criteria are scored 1–5. The formula produces a final score between 1.0 and 5.0, rounded to one decimal place. A score of 4.0+ means we recommend the tool. A score of 4.5+ means it is our top pick in its category.
Four to six weeks, real work, real output.
Sign up fresh
We start from a new account — not a press account, not a beta access. The experience we review is the experience you get on day one.
Use it for actual work
We use the tool for real tasks over four to six weeks. Not synthetic tests. If it's a writing tool, we write real articles with it. If it's a video tool, we make clips we actually want to keep.
Watch a beginner use it
We give the tool to someone who has never used it before and watch where they get stuck. This is how we score ease of use accurately — not by how easy it felt to us after six weeks.
Compare against alternatives
We run the same prompts through at least two competing tools before finalizing scores. No tool is reviewed in isolation.
Score, then write
The score is calculated before the article is written — not after. This prevents reverse-engineering a score to match a conclusion that was already written.
What we will never do.
Inflate scores for affiliates
Affiliate partners get no score advantage. If a tool we earn money from scores 3.8, we publish 3.8. The affiliate relationship is disclosed at the top of every article that has one.
Skip the "skip it if" section
Every review tells you who should not use the tool. A review without a real "no" is not an honest review — it's a product page with extra steps.
Review tools we haven't tested
If we haven't used it for real work, we don't publish a score. We flag articles based on limited testing and update them once we've used the tool properly.
Leave outdated scores live
AI tools update fast. We revisit scores when a major update ships. The "updated" date in the article header is the date the score was last verified, not just the date we fixed a typo.
Questions about a specific score? Email us at [email protected]. We respond to every challenge and will revise a score publicly if the evidence supports it.