Methodology
How we test customer support software
Our rankings are hands-on and evidence-led. Here is who runs the reviews, the criteria we score every tool against, and how we make money - so you can judge the work for yourself.
Last reviewed August 2026
Who does the testing
Every review on Zazachat test 5 is run by our editorial reviewer, Marcus Fitzgerald, who evaluates live chat and AI support tools hands-on against the published criteria below. We buy or request trial access to each product, configure it as a real support team would, and run representative conversations through it before we score or rank anything.
This is a placeholder overview. The full, worked methodology - including the exact test scenarios and evidence we capture for each tool - is maintained here and expanded by our editorial team.
What we score - our criteria
We score each product against a fixed set of criteria so that two tools in the same category are judged on the same terms. These are the dimensions behind every ranking:
- Setup & onboarding. How quickly a team can install the widget or connect channels, import data, and get a first conversation flowing without engineering help.
- Agent experience. The day-to-day inbox: routing, canned responses, keyboard flow, collision detection, and how many clicks a routine reply actually takes.
- Automation & AI. Bot/AI-agent resolution quality, handoff to a human, workflow builders, and how much of a real ticket queue the automation can deflect.
- Omnichannel & integrations. Coverage across chat, email, social, and voice, plus the depth of CRM, e-commerce, and help-desk integrations that matter to support teams.
- Reporting & analytics. The metrics a support lead needs - response time, CSAT, resolution rate - and whether they are usable out of the box or need a data team.
- Reliability & security. Uptime track record, data residency and compliance options, role-based access, and the controls regulated teams have to have.
- Pricing & value. Transparent per-seat and usage costs, what is gated behind higher tiers, and whether the price matches the value for the team size it targets.
- Support & documentation. The vendor's own support responsiveness, the quality of their help center, and how easy it is to get unstuck during a rollout.
The SupportScore metric
Each tool's performance across the criteria above rolls up into a single SupportScore - our named, category-weighted rating. It exists so you can compare tools at a glance without wading through every sub-score. The weighting reflects what matters most for the job a category is hired to do; a live-chat widget and a knowledge base are not weighted identically.
Individual SupportScores publish alongside each review as the content is completed - we never show a score before a tool has actually been tested.
How the ranking order is decided
Ranking order follows this methodology and the evidence from testing - not payment. When a product is a paid or sponsored placement we label it clearly and set it apart in its own slot, separate from the organically ranked list. A sponsored vendor is never placed at the top of the editorial ranking on the basis of payment.
How we make money
Zazachat test 5 is free to read and reader-supported. Some of the links to vendors are affiliate links, which means we may earn a commission when you buy or sign up through them - at no extra cost to you. We also accept sponsored and featured-partner placements, which are always labeled as such on the page.
Affiliate commissions and sponsorships fund the testing, but they do not buy a ranking. The order of our lists is editorial and follows the criteria on this page. For the full picture, see our editorial policy.
Frequently asked questions
Coming soon. Answers to common questions about our testing process and scoring will appear here as this section is completed.