Pick what the bot is for, where it must work and who will build it. The finder filters our product database, the same documented facts behind every review, and shows why each builder was kept or dropped. It is a filter, not a recommendation engine: it narrows the list, and the reviews help you choose.
9 builders fit
Funded support teams that want an AI agent, a shared inbox and ticketing in one tool
Support teams with a developer on hand that want agent building and a helpdesk in one tool
Teams already running sales or support in HubSpot that want a bot writing straight into their CRM
Marketing teams in Europe that want lead-capture flows on a website, with prices in euros and data hosted in the EU
Small online shops and support teams that want an AI agent, simple automations and a shared inbox from one vendor, with published self-serve prices
Creators and social-first shops that sell through Instagram, TikTok or Messenger DMs
Developers and technical marketers who want cheap per-chat pricing or a self-hosted bot builder
Technical CX teams and agencies that want a voice and web-chat agent with model choice, staging environments and code
Agencies and small sales teams that want an AI agent answering leads on WhatsApp and Instagram
A filter over documented facts. Matches are ordered by fit, then by rubric score. It is not based on hands-on tests. Budgets compare the vendor's entry price in its own currency.
How the finder decides
- Hard requirements exclude. A channel, EU data hosting, self-hosting or a budget removes every builder that does not meet it, and the reason is shown. EU hosting and self-hosting count only when the vendor documents them fully; a partial answer is treated as not met.
- Budgets compare entry prices in the vendor's own currency, without conversion. A builder with a free plan passes every budget. A builder whose prices are not published is dropped from the paid budgets.
- Use case and skill decide the order. A builder made for your job ranks above one that is not, and a match on who will build it comes next.
- The rubric score breaks ties. When two builders fit equally well, the one with the higher score from how we review comes first.
- Nothing is based on tests. The finder reads documentation facts. We have not run bots on these platforms; the reference-bot test is designed but not yet run.
What it cannot tell you
The finder cannot judge how good a bot's answers will be, how long a build takes, or what your bill will be at your volume. It does not know your helpdesk, your CRM or your legal requirements. Treat the results as a shortlist, then read each review and pricing page, and check the billing unit before comparing prices.
- best/chatbot-builders
- guides/chatbot-vs-ai-agent
- guides/what-counts-as-a-conversation
- guides/chatbot-pricing-models
- how-we-review
Frequently asked questions
How does the chatbot finder rank builders?
The chatbot finder first removes builders that miss a hard requirement, such as a channel, EU hosting, self-hosting or budget. It then orders the rest by fit with your use case and builder skill, and breaks ties with our published rubric score. It is a filter over documented facts, not a recommendation engine.
Is the chatbot finder based on testing the products?
No. The chatbot finder uses the facts in our product database, which come from vendor pricing pages, help centers and trust pages. We have not built or run bots on these platforms; our reference-bot test is designed but not yet run.
Why was a chatbot builder excluded by the finder?
The chatbot finder lists every excluded builder with its reason, such as no documented channel, no free plan or an entry price above your budget. EU hosting and self-hosting only count when the vendor documents them fully. If the vendor has changed its offer, report it and we will re-check the source.
Researched and drafted with AI assistance from the sources listed on this page. We have not built or tested bots on these platforms yet. Method: How we review