Everyone's talking about AI chatbots right now. Every second LinkedIn post tells you that if you don't have an AI chatbot on your website, you're basically leaving money on the table. And look, there's some truth to that. But there's also a lot of hype that needs to be separated from what actually works for small businesses.
Can I share something with you? I've tested chatbots on client websites, on our own website, and across different industries. The results have been mixed. Not bad. Not magical. Mixed. And I think you deserve an honest take rather than another "AI will change everything" article.
Do the conversion numbers actually hold up?
The commonly cited stat is that AI chatbots can increase website conversion rates by 15 to 25 percent. And in our experience, that's roughly accurate, but with a massive caveat. It depends entirely on the type of business, the type of visitor, and how well the chatbot is configured.
For service businesses where the visitor has a specific question, like "Do you service my area?" or "How much does a basic electrical inspection cost?", a chatbot that can answer those questions instantly performs really well. It removes friction from the enquiry process. The visitor gets their answer, feels confident, and submits a form or books a call.
For businesses with more complex offerings where the visitor is browsing and researching, a chatbot that pops up after three seconds asking "How can I help?" is more likely to annoy than convert. We've all been that person who closes the chat window 0.5 seconds after it appears.
The conversion lift is real, but it's not automatic. It requires thoughtful implementation.
When do chatbots actually help?
Chatbots earn their keep in a few specific scenarios. First, after-hours enquiries. If your business operates 9 to 5 but your website gets traffic at all hours (and it does), a chatbot can capture leads at 10pm on a Sunday when nobody's around to answer the phone. It collects the visitor's name, email, and what they need, then feeds that into your CRM for follow-up first thing Monday.
Second, FAQ handling. If you find yourself answering the same five questions over and over (pricing ranges, service areas, turnaround times), a chatbot trained on those answers saves you time and gives the visitor instant information. It's like having a receptionist who never takes a lunch break.
Third, lead qualification. A well-configured chatbot can ask a couple of qualifying questions before passing the lead to you. "What suburb are you in?" and "What type of service do you need?" filters out the tyre-kickers and ensures you're only spending time on leads that are actually a fit.
When do chatbots frustrate users?
Here's the thing. A bad chatbot is worse than no chatbot at all. And there are a lot of bad chatbots out there.
Losing leads to slow follow-up? Let's map out an automation system that works while you sleep.
Book a free callThe worst offenders are the ones that pretend to be human. If your chatbot introduces itself as "Sarah from the team" and the visitor quickly realises they're talking to a bot, you've lost trust. Immediately. People don't mind talking to AI as long as you're upfront about it. They very much mind being tricked.
Then there are the chatbots that can't handle anything outside their script. The visitor asks a reasonable question, the bot spits out an irrelevant answer, and the whole interaction feels like talking to a broken phone menu. This is the "press 1 for sales, press 2 for support" experience translated to a chat window, and it's just as frustrating.
Finally, there are the overly aggressive chatbots. The ones that pop up immediately, follow you as you scroll, reappear after you close them, and generally behave like that salesperson who follows you around the shop. Visitors leave. Bounce rates go up. The chatbot that was supposed to increase conversions actively drives people away.
What should you actually look for in a chatbot solution?
If you're considering adding a chatbot to your website, here's what I'd look for based on what we've seen actually work.
First, make sure it integrates with your CRM. A chatbot that captures lead information but dumps it into a separate system defeats the purpose. The lead data should flow directly into your pipeline so your automated follow-up sequences can take over.
Second, look for natural language processing that's actually good. The bar has risen significantly with large language models. A chatbot built on modern AI should be able to handle conversational queries, not just keyword matching. If a visitor types "do you guys work on weekends" and the bot can't figure out they're asking about your availability, it needs more work.
Third, insist on guardrails. Your chatbot should only answer questions about your business, using information you've provided. It should not make up pricing, promise turnaround times you can't deliver, or go off-script. The best chatbot solutions let you define exactly what the bot knows, what it can say, and what it should escalate to a human.
Fourth, timing and behaviour. The chatbot should not appear the instant someone lands on your site. Give the visitor 30 to 60 seconds to orient themselves. Better yet, trigger it based on behaviour. If someone's been on your pricing page for 45 seconds, they probably have a question. That's the right moment.
So should you get one or not?
My honest take? If you're a service business getting more than 500 website visitors a month, a well-implemented chatbot is worth testing. Set it up, run it for 30 days, and look at the numbers. Did your form submissions increase? Did you capture leads you wouldn't have otherwise? If yes, keep it. If not, pull it.
If you're getting fewer than 500 visitors a month, your priority should be driving more traffic through SEO and ads before worrying about chatbot optimisation. A chatbot on a low-traffic site is like putting a receptionist in an empty waiting room.
And whatever you do, don't buy into the hype that a chatbot will magically transform your business overnight. It's a tool. Like any tool, it works when used properly and creates a mess when used carelessly. Get the fundamentals right first (a fast website, clear calls to action, a CRM that captures and follows up with every lead) and then layer a chatbot on top if it makes sense for your situation.
The businesses that win aren't the ones with the fanciest tech. They're the ones that respond fast, follow up consistently, and make it easy for customers to say yes. If a chatbot helps you do that, great. If it just adds complexity without results, skip it.
