How we test
Every score comes from the same five-part rubric, and every review follows the same method. Here’s how both work.
- Updated
Every TechKea score is out of 10, to one decimal, and comes from the same five measures weighted the same way for every tool. Commission rates never enter the math. This page sets out the rubric, how the score is calculated and the testing method behind it.
The rubric
Each tool gets a sub-score out of 10 for five measures. The weights reflect what matters most when you run a business on your own or with a small team: whether you can actually use the tool, and whether it does the job you’re paying for.
| Measure | Weight | What it asks |
|---|---|---|
| Ease of use | 25% | Can you set it up and do the main job without help? |
| Features for the target user | 25% | Does the plan you would buy do the job you bought it for? |
| Value for money | 20% | Is the price fair for what you get, now and as you grow? |
| Support and reliability | 15% | Can you get help, and can you rely on it with your work and data? |
| Integrations | 15% | Does it connect to the tools a small business already uses? |
| Total | 100% |
Ease of use, 25%
A high score means you can sign up, set up and finish the tool’s main job, such as sending an invoice or publishing a page, without reading the documentation, and everyday tasks stay quick once you know your way around. Sensible defaults, clear settings and a mobile app that works when you need it all count.
It scores low when setup needs a consultant or a long tutorial, common tasks hide behind menus, or the interface fights you on the tenth use as much as the first.
Features for the target user, 25%
We judge features against the person the tool is for, not against the longest feature list. A high score means the plan a freelancer or small team would actually pay for covers the jobs they need done, with nothing essential held back for an enterprise tier.
It scores low when key features sit on expensive plans, the basics are thin, or a pile of extras gets in the way of the core job.
Value for money, 20%
Value isn’t the same as cheap. A high score means the price is fair for what the plan includes, the free plan or trial is honest about its limits, and costs grow predictably as you add clients, contacts or seats.
It scores low for steep jumps between tiers, per-seat pricing that punishes small growth, paid add-ons needed for basic use, and renewal prices far above the first-year offer.
Support and reliability, 15%
A high score means help is easy to reach on the plan you would buy, answers are useful, the help docs are current, and the service is dependable. We also look at how it looks after your data, such as account security options and a clean way to export everything.
It scores low when support is locked to higher tiers, replies are slow or scripted, or we run into outages, lost work or a hard way out when you want to leave.
Integrations, 15%
A high score means it connects directly to the payment, accounting, calendar, email and storage tools small businesses commonly use, the connections hold up in daily use, and automation tools can reach it when there is no direct connection.
It scores low when integrations are few, cost extra, break often or only sync one way.
How the score is calculated
Each measure is scored out of 10 in half points, so 7, 7.5 and 8 are all possible but 7.3 is not. The overall score is the weighted average of the five, shown to one decimal. When the weighted average lands exactly halfway between two tenths, it rounds up.
The math runs in whole numbers, so the same breakdown always gives the same score, and the site’s build checks every published score against its breakdown. A score that doesn’t match the rubric can’t go live.
Here’s a worked example with made-up sub-scores, not a real tool:
| Measure | Sub-score | Weight | Points |
|---|---|---|---|
| Ease of use | 9.0 | 25% | 2.250 |
| Features for the target user | 7.5 | 25% | 1.875 |
| Value for money | 8.0 | 20% | 1.600 |
| Support and reliability | 7.0 | 15% | 1.050 |
| Integrations | 6.5 | 15% | 0.975 |
| Weighted total | 7.750 | ||
| Overall score, to one decimal | 7.8 | ||
The weighted total of 7.750 sits exactly halfway between 7.7 and 7.8, so it rounds up to 7.8. In a review, the overall score sits at the top of the verdict box with all five sub-scores beneath it, so you can see where a tool earned or lost points.
How we test a tool
The method is to use each tool the way you would: on a real account, doing real small-business work, and recording what happens. Every review follows these steps.
- A real account. We sign up like any customer, on free trials or plans we pay for, usually the plan a freelancer or small team would most likely buy. If a vendor gives us a free account, we say so in the review.
- Real tasks. Each tool is put to work on the jobs it’s sold for, such as sending and chasing an invoice, publishing a page, booking a meeting or sending a campaign.
- At least a week of real use. Every tool gets at least a week of real use on its web app, plus its mobile or desktop app if it has one.
- Evidence you can see. We capture screenshots of what we describe, and screen-recorded walkthroughs on YouTube where a video shows it better. All of it comes from our team’s own testing. We never use mockups, stock images or AI-generated pictures of software.
- Prices checked at the source. Prices, plans and trial terms come from the vendor’s official pricing page, checked when a review is published and again every 90 days. Every pricing table shows “Prices checked: Month YYYY” beneath it, and if we can’t verify something, we say so instead of guessing.
- Scored with the rubric. Each measure gets a half-point sub-score, and the overall score is calculated as described above. The review also names who the tool is best for and who should skip it.
- Dated. Every review shows “Last tested: Month YYYY” in its verdict box, so you know how fresh the findings are.
Keeping reviews current
Software changes constantly, so a review is only as good as its last test. That’s why every review shows when it was last tested and when its prices were last checked. We re-check prices every 90 days and re-test reviews at least once a year or after a major update.
When a product changes in a way that affects the verdict, such as a redesign, a new plan, a price change or a removed feature, we re-test the parts that changed, update the score if needed and add a dated entry to the review’s “What changed” log. The “Updated” date only moves for changes like these, never for a fixed typo.
Commissions and scores
Commission rates never influence a score, a ranking or a recommendation. The rubric has no line for commission, and tools without an affiliate program are tested and scored exactly the same way.
Some links on TechKea are affiliate links. Our affiliate disclosure explains how they work and what they cost you.
Products we own
If TechKea’s owner or Aprici Inc. owns a product we cover, or a company connected to them makes it, we say so prominently at the top of the page. We don’t score it, rank it against competitors or mark it as a winner in a comparison. The ownership section of our editorial policy has the details.
What “Best for” and “Skip if” mean
Every verdict box has two short lines that can matter as much as the score, because the right tool depends on who you are and what you need.
Best for names the reader the tool suits best, such as “solo freelancers who invoice a handful of clients a month”. A tool with a middling score can still be the right pick if that line describes you.
Skip if names the situation where you should look elsewhere, such as “you need payroll or inventory”. Every review has one, because every tool is the wrong choice for someone. When we can, we point you to a better fit, and our comparisons put two or three options side by side.