I get asked about chatbots more than almost any other feature request, usually right after a business owner sees one on a competitor's site. My honest answer, before we get into the details of services worth adding to a website: most small businesses don't need one. A phone number and a well-built contact form already answer the same questions a chat widget would, and an unmonitored chat bubble sitting in the corner of your site can do more harm than good.
Does a Small Business Website Need a Chatbot?
No, not usually. A chatbot or live chat widget solves a real problem for businesses with staff available to answer it and enough traffic to justify the tool, but for most local service businesses, a visible phone number and a short contact form cover the same ground for free. The exception is appointment-based businesses with someone at a desk during business hours who can actually respond, where chat becomes one more way to catch a visitor who wouldn't have picked up the phone.
What's the Difference Between Live Chat and an AI Chatbot?
Live chat puts a real employee on the other end of the conversation, typing back in real time. That works well when someone is actually watching it, and badly the moment nobody is, since a chat window that sits unanswered for twenty minutes looks worse to a visitor than having no chat option at all. An AI chatbot replaces that person with automated software, either following scripted decision trees or, more recently, a language model trained to hold a more natural conversation.
The two get lumped together because they look identical on the page, the same small bubble in the bottom corner. But the cost, the setup time, and the risk are different. Live chat costs you staff time. A chatbot costs you setup and a monthly fee, plus the risk that it answers a question wrong or makes a promise your business can't keep.
What Does Adding Chat to a Small Business Website Cost?
| Option | Typical monthly cost | What it needs from you |
|---|---|---|
| No chat, just phone and a form | $0 | Nothing extra, already built into most sites |
| Basic live chat widget (Tidio, Crisp) | $0-$30 | Someone monitoring it during business hours |
| Facebook Messenger or Google Business chat | $0 | Someone checking a second inbox regularly |
| Scripted chatbot (rule-based FAQ bot) | $20-$60 | Setup time to write the decision tree correctly |
| AI chatbot trained on your business | $50-$300+ | Ongoing review of what it's actually telling people |
The free and cheap options aren't necessarily the safe ones. A free live chat widget that nobody answers is worse than no widget, and a $20 scripted bot that can't handle a question outside its script sends a visitor away frustrated instead of picking up the phone. The AI chatbots at the top of that range can hold a genuinely useful conversation, but they need someone checking transcripts regularly to catch a wrong answer before it costs you a customer.
Which Small Businesses Actually Benefit From Website Chat?
Appointment-based businesses with someone at a desk during the day get the most out of it. A med spa or a dentist's office with front-desk staff already fielding phone calls can often answer a chat message in the same window of time, catching a visitor who was too shy to call but would have booked given a lower-pressure option. I go into this exact split in best website features for a med spa, where the features that convert are the ones matched to how that specific customer actually prefers to reach out.
Emergency and urgent trades are the opposite case. Someone with a dead air conditioner or a burst pipe wants a person on the phone right now, not a chat window that might get answered in an hour. I cover why click-to-call still wins for those businesses in what should an HVAC company website include and in how to get more plumbing leads from your website, where a chatbot adds a decision point that slows someone down instead of speeding them up.
What Are the Downsides of Adding Chat to Your Website?
An unanswered chat window looks worse than no chat window. If a visitor sends a message and waits twenty minutes for a reply, or never gets one, that's a worse experience than a site with just a phone number and a form. Chat sets an expectation of speed that a small business without dedicated staff often can't meet.
A chatbot can say something wrong. Scripted bots can misfire outside their narrow script, and AI chatbots can occasionally state something inaccurate about pricing, availability, or your services with total confidence. The FTC has warned businesses directly about overselling what AI tools can actually do, and that risk applies just as much to a chatbot quietly misrepresenting your business to a customer.
It's one more thing that can slow your site down. Most chat widgets load a script from a third-party server, and a poorly optimized one adds real weight to a page. I've audited sites where a chat widget alone accounted for a noticeable chunk of load time, working against everything else built for sub-1-second load times.
It adds a data collection point you have to disclose. Most chat tools log the conversation and sometimes an email or phone number a visitor types in. That's a change worth reflecting in your privacy policy if you have one, or a reason to get one in place if you don't.
What Should You Do Instead of Adding a Chatbot?
Make sure your service pages already answer what someone would type into a chat window. Most visitors reaching for a chat bubble are trying to confirm something specific: do you serve their area, do you handle their particular job, roughly what does it cost. A clearly written service page with a visible phone number and a short quote form answers all three without needing a single line of chat software. I've seen this fix more lead problems than adding any new tool, since it removes the reason someone needed to ask in the first place.
If you still want a lower-pressure option beyond the phone, a short contact form asking for the job type and a callback time gets you most of the benefit chat promises, without staffing a chat window or paying a monthly fee for a bot that might answer wrong.
What Should You Do Next?
- Check whether your service pages already answer the common questions. If people are asking the same three things over and over, that's a page-content problem, not a missing-chat-widget problem.
- If you're appointment-based with front-desk staff, consider a monitored live chat tool, not an unmonitored bot pretending to be a person.
- Skip the chatbot if you're an emergency trade. Keep the phone number front and center instead, and see how that plays out in a live example inside my portfolio.
- Get a free homepage mockup from my contact page if you want a second opinion on whether your current site already covers what a chatbot would try to fix.
A chatbot can be worth adding once you have the staff and the volume to make it pay for itself. Most small businesses I work with aren't there yet, and a fast, clear website with a real phone number gets them further than a chat bubble ever would.