How we review and score AI tools for small businesses.
Seven checks behind every review
What each score measures
Each review rates the tool on the categories that matter for its type of product, on a scale of 0 to 10. The core categories are:
| Category | What we look at |
|---|---|
| Ease of use | How quickly a non-technical owner gets real value, and how steep the learning curve is. |
| Features | Whether the tool covers the jobs a small business needs, without paying for an upgrade. |
| AI capabilities | How good the AI output is in practice, how often it needs fixing, and whether it is more than a marketing label. |
| Customer support | How easy it is to get help, on the plans a small business is likely to buy, and what users report. |
| Integrations | Whether it connects to the tools small businesses already use, such as email, calendars, CRMs and payments. |
| Value for money | What you get at each price, hidden costs such as credits or onboarding fees, and how steep the upgrades are. |
| Scalability and automation | How well it holds up as a team grows, and how much repetitive work it can take off your plate. |
How to read a score: 9 to 10 is exceptional and rare. 7 to 8.5 is a strong choice with caveats you should know. 5 to 6.5 is usable but has real drawbacks. Below 5 is not something we would recommend paying for.
How the overall score works: the overall score is an editorial judgment informed by the category ratings, not a simple average. We weight the things that matter most to small businesses, especially value for money, ease of use and support, so a tool with brilliant features but punishing pricing can score lower than its feature rating suggests.
Every score we have published
All 21 reviews in one place. Scores come straight from each review, and each tool links to its full write-up with the reasoning and sources behind it.
| Tool | Type | Overall | Ease of use | Features | AI | Support | Value |
|---|---|---|---|---|---|---|---|
| Zapier | Automation | 8.5 | 9.5 | 9 | 8 | 6 | 6.5 |
| Canva AI | Design | 8.5 | 9.5 | 9 | 8 | 7 | 8.5 |
| ChatGPT | AI assistant | 8.0 | 8 | 9.5 | 9 | 4 | 7 |
| HubSpot CRM | CRM | 8.0 | 9 | 8.5 | 7.5 | 7.5 | 7 |
| Hostinger | Web hosting | 8.0 | 8.5 | 8 | 6 | 7 | 8.5 |
| Notion AI | Workspace | 7.5 | 6 | 8.5 | 7.5 | 7 | 6.5 |
| Gamma | Presentations | 7.5 | 9 | 8 | – | 5.5 | 7 |
| HeyGen | AI video | 7.5 | 8.5 | 8 | 9 | 5 | 6 |
| Runway | AI video | 7.5 | 6 | 9 | 9 | 3.5 | 5.5 |
| Apollo.io | Sales prospecting | 7.5 | 8.5 | 8 | 6.5 | 5 | 6.5 |
| OpenRush | SEO data | 7.5 | 8 | 7.5 | – | 8 | 8 |
| Synthesia | AI video | 7.0 | 8.5 | 8 | 7.5 | 5 | 6 |
| Perplexity | AI search | 7.0 | 9 | 7.5 | 6.5 | 3.5 | 6.5 |
| Jasper | AI writing | 7.0 | 7 | 8.5 | 7 | 7 | 5.5 |
| Midjourney | AI images | 7.0 | 6.5 | 7.5 | 8.5 | 4 | 6 |
| Pika | AI video | 7.0 | 8 | 8 | 6.5 | 5 | 6 |
| Lindy AI | AI assistant | 7.0 | 8 | 8.5 | 8 | 5.5 | 6 |
| Lovable | App builder | 6.5 | 8.5 | 8 | 7 | 5.5 | 5.5 |
| Clay | Lead data | 6.5 | 5 | 8.5 | 7.5 | 3.5 | 5.5 |
| Base44 | App builder | 5.5 | 8 | 7.5 | 5.5 | 3.5 | 5 |
| Fireflies.ai | Meeting notes | Not scored | – | – | – | – | – |
A dash means that category was not scored for that tool, usually because it does not apply. Fireflies.ai was reviewed before we introduced scores and has not yet been re-scored. No score on this page was paid for. See our editorial policy.
How we keep scores current
- AI tools change monthly, so reviews show when they were last checked. We re-check pricing and major features when a significant update ships, and aim to review every score at least once a quarter.
- When a score changes, we update the review, this table and our verdict boards together, and note what changed.
- We cover major updates to tools we have reviewed in AI News, linked from the review.
What we will never do
- Sell a score, a ranking or a place on a “best tools” list.
- Base a review on a press release or a vendor’s own claims without saying so.
- Invent test results, testimonials or user quotes.
- Hide a commercial relationship. Our full rules are in the editorial policy.