Editorial Policy
Last updated
SmallBizTech exists to help small business owners make better technology decisions. This policy explains how we create content and what you can expect from us.
Research
Every guide starts with the questions small business owners actually ask. We research using:
- Hands-on use of the software, where we have it (each page says so)
- Independent reviews and research from named, linked sources
- Official vendor documentation and pricing pages
- Public help centres and changelogs
We link to our sources, and pricing and feature details include the date we checked them.
Testing
Most pages on SmallBizTech are currently based on vendor documentation and named independent reviews, not our own testing, and they say so at the top. When we do test software, we sign up for an account (free trial, free plan or paid plan) and work through realistic tasks for the business type the review is aimed at — for example, taking a job from enquiry to paid invoice.
Some products we haven't been able to use ourselves yet. Those pages are researched from the vendor's official website and documentation, and they say so at the top — they never describe an experience we haven't had. When we do get hands-on access, we rewrite the page and update the date.
In reviews we label information clearly:
- Tested: something we did ourselves
- Vendor info: claims made by the software company
- Public docs: information from official documentation
We do not publish star ratings until we have a documented rating methodology.
How we test automations
Automation workflow articles (recipes) follow a stricter standard than software reviews, because they describe something we built and ran ourselves.
What counts as "tested" on this site:
- A workflow built and run by the named author, with the platform version and hosting type stated (for example: n8n 1.x, n8n Cloud Starter)
- A screenshot of the finished workflow canvas
- A test run table with at least three runs: date, input, expected result, actual result, and time taken
- At least one deliberate failure case and what the workflow did
- Costs taken from the vendor's pricing page, with the date checked
How we word it:
We use the label "From our testing" only when all of the above apply. If part of a workflow hasn't been tested — for example, a variation for a different trade — we say so clearly and leave that section out or mark it "not tested".
We never claim results from a business that didn't participate ("this saved a salon 8 hours a week"), invent testimonials, or report hours saved that we didn't measure.
If we later run a workflow for a real client, we report their real results with their permission and their real numbers.
The author is accountable. AI tools help research and draft. Every article is published under a named author who reviewed and approved it. The testing evidence is the author's own work.
Comparisons
We only publish comparisons between products we've researched in depth, where there's a real difference to explain. Comparisons use the same test scenario for each product.
AI use
We use AI tools to help research, draft and check articles against the sources they cite. AI is never used to invent testing experiences, screenshots, prices, features or quotes. Every article is published under a named author, who is accountable for it, and every page lists the sources its facts come from, with the date each was checked.
Updates
We review our most important guides regularly and update them when pricing, features or our recommendations change. The "Updated" date on an article only changes when we make a meaningful update.
Corrections
If you spot an error, please contact us. We correct factual errors promptly and note significant corrections at the end of the article.
Independence
- Vendors can't pay for a review, a higher ranking or a favourable verdict.
- We may earn affiliate commissions from some links. This never affects which products we recommend. See our affiliate disclosure.
- Advertising is kept separate from editorial content. See our advertising disclosure.