The short version: for most small businesses, the choice isn’t really between live chat platforms, it’s between doing it yourself with a tool like Tidio or Olark, paying an outsourced answering service to staff it for you, or bolting an AI chatbot onto your website and hoping nobody notices it’s a robot. Pick based on how many chats you get and who’s free to answer them, not on which tool has the flashiest homepage.
What “live chat services” covers
People ask me this question and mean three completely different things, so let’s untangle it first.
- Software you run yourself: a widget on your site, you or your team answer it. Think LiveChat, Tidio, Olark, Crisp, Freshchat.
- Outsourced human chat: a company employs the agents, you pay per chat or per month, they answer as “you” using a script and knowledge base you give them.
- AI-first chat: a chatbot handles the first reply, sometimes the whole conversation, sometimes it hands off to a human when it gets stuck.
Most small businesses I work with end up mixing two of these once they’ve been through a launch or a busy season. I’ll get to why in a minute, but the mistake I see constantly is buying the software first and figuring out who’s answering it later. That’s backwards. Staffing is the actual product. The chat widget is just the doorbell.
The tools worth your money
If you’re going to run chat yourself, here’s what I’ve used or watched clients use, with real 2026 pricing so you’re not guessing.
- LiveChat starts around 20 dollars per agent a month on the basic plan, rising to roughly 59 dollars for the Team plan. Solid for a small service business that just wants a clean widget and decent reporting.
- Tidio has a usable free tier for one seat, then jumps to about 29 dollars a month for small teams. Good starter for a solo founder or a tiny ecommerce shop.
- Olark sits around 29 to 49 dollars per agent a month. Simple, a bit old-fashioned, but reliable, and it plays nicely with basic CRMs.
- Crisp has a free plan and paid tiers from roughly 25 dollars a month, and it’s the one I point B2B founders towards when they want chat plus a shared inbox without paying Intercom money.
- Intercom starts near 39 dollars per seat but realistically most small businesses land closer to 80 to 150 dollars a month once they add the features they need. It’s the most capable and the easiest to overspend on.
- Freshchat ranges from about 15 to 69 dollars per agent depending on the tier, and it’s the one that plays best if you’re already using Freshdesk for support tickets.
- Gorgias is built specifically for ecommerce and prices by ticket volume rather than seats, starting around 10 dollars a month for tiny stores and climbing steeply for high-volume Shopify brands.
If your team spends any real time selling through chat rather than just answering support questions, the software you pick matters less than whether it routes leads into your CRM. That’s a separate decision, and I’ve written a longer breakdown on how to pick B2B live chat software for a sales team if that’s your situation, because the checklist is different from a support-only setup.
What happened when I tested this on my own site
A few years back I put a chat widget on my site during a launch week, mostly because someone told me it would “capture more leads.” I ran it myself, on my phone, in between everything else I was doing that week. Within four days I had 31 chats. Six were real enquiries. Three of those became actual paying clients. The other 25 were people asking questions I’d already answered in my FAQ, or bots crawling the widget, or one memorable person asking if I could help with their crypto trading course, which I politely declined.
The number that stuck with me wasn’t the leads, it was the response time. I was averaging just under two minutes because I had the tab open constantly, and every single one of the three clients who converted told me later that the speed was what made them trust me. Not the answer itself. The speed.
That’s the whole argument for live chat in one sentence: it’s not a better support channel than email, it’s a faster one, and speed is what people are paying for when they choose chat over a contact form.
Outsourcing it instead of running it yourself
The week after that launch I couldn’t keep answering chats at 11pm forever, so I looked seriously at outsourced services. This is where a lot of the advice online gets thin, because most posts compare software features and skip the actual staffing question entirely.
Outsourced live chat services generally price one of two ways: per chat, usually somewhere between 1 and 5 dollars per conversation, or a flat monthly retainer that typically starts around 200 to 400 dollars a month for a small business with modest volume and climbs from there based on hours covered. Companies like Ruby, Apex Chat, and various white-label agencies that resell tools like LiveChat or Olark all sit roughly in that range, though the exact number depends heavily on whether you want 24/7 cover or just extended business hours.
Before you sign anything, read through what to check before hiring a live chat outsourcing company, because the questions that matter aren’t the ones on their sales page. Ask specifically how they handle a question they don’t know the answer to, and ask to see an anonymised transcript of a real conversation, not a demo script. If they won’t show you one, that tells you something.
The uncomfortable bit nobody wants to say out loud
Here’s what most articles on this topic won’t tell you plainly: a badly staffed live chat is worse for your business than no live chat at all. If your widget says “we typically reply in a few minutes” and then a visitor waits eleven minutes and gets a copy-paste answer that doesn’t address their question, you’ve just demonstrated to a warm lead that you’re disorganised. They’d have trusted you more if the chat bubble simply wasn’t there and they’d emailed you instead. I’ve seen this happen with two separate small business clients who added chat before they had anyone dedicated to watching it, and both saw their conversion rate on chat leads drop below their email conversion rate within a month. The tool wasn’t the problem. The gap between the promise on the widget and the reality of who was behind it, that was the problem.
So before you buy anything, answer this honestly: who, specifically, by name, will answer a chat that comes in at 4pm on a Tuesday when you’re in a client meeting? If you don’t have an answer, you’re not ready for live chat yet, you’re ready for a “leave a message” form that looks like chat, which is a perfectly respectable thing to build instead.
Where AI fits, and where it doesn’t
Every live chat platform above now has some flavour of AI bolted on, usually pitched as answering 60 to 80 percent of chats without a human. In practice, for a small business with under a few hundred chats a month, the maths rarely works in your favour once you account for setup time and the awkward conversations where the bot confidently gives a wrong answer. AI chat earns its keep once you’re past a certain volume, roughly the point where a human answering every chat would need to be a full-time hire rather than a shared task.
If you’re weighing whether AI-powered chat is worth building into your setup this year, it’s worth reading how the wider AI landscape is shifting for small business owners, and I cover the practical, non-hype version of that regularly, including in this week’s AI news roundup for small business. If you’d rather have someone assess your specific setup instead of guessing from a blog post, that’s exactly the kind of decision an AI consultant for small business should be walking you through before you spend money on tools you might not need.
How to decide, step by step
- Count your current enquiries. Look at your last 30 days of emails, contact form submissions and DMs. If it’s under 20, chat is a nice-to-have, not urgent.
- Decide who answers it, by name, before you pick software. If the answer is “whoever’s free,” it won’t get answered consistently and visitors will notice.
- Start with the cheapest tool that does the job. Tidio’s free tier or Olark’s entry plan is enough to test whether chat converts for your business before you spend Intercom money.
- Run it for 60 days and track exactly what I tracked: number of chats, number of real enquiries, number that became paying customers, and your average response time.
- If volume grows past what one person can handle, that’s your signal to look at outsourcing rather than hiring. The maths on outsourced chat usually beats a part-time hire until you’re well past 300 to 400 chats a month.
If outsourcing is where you land, it’s worth understanding what outsourcing live chat really involves before you commit, because the practical setup, handover documents, tone of voice training, escalation rules, takes longer than most agencies let on in the sales call.
A quick word for anyone thinking of reselling chat services
I get asked this fairly often by consultants and marketers who see live chat as a nice recurring revenue add-on for their clients. It can work, but the margins are thinner than people expect once you account for support time, and the client relationship gets complicated fast if a chat goes badly and you’re the one who set it up. If that’s on your radar, I’d read whether reselling live chat software is worth it before you pitch it to a single client, because the answer depends heavily on how much hand-holding you’re prepared to do after the sale.
What I’d tell a friend starting from scratch
If someone messaged me tomorrow asking what to set up, here’s the honest, unglamorous answer: start with Olark or Tidio, staff it yourself for 60 days, watch the transcripts every single one, and decide from real data rather than a demo. The businesses I’ve seen get the most out of chat are the ones treating it as another sales channel with a real owner, not a piece of software they installed and forgot about. And the ones who see the biggest lift in actual sales, not just chat volume, tend to be the ones who read up on how live chat services help close more sales before they build their script, because a chat that just answers questions is very different from one designed to move someone towards buying.
Frequently asked questions
What’s the cheapest live chat service for a small business?
Tidio’s free plan for one seat and Crisp’s free tier are the cheapest usable options in 2026, both good enough for a small business testing whether chat converts before spending anything.
Is outsourced live chat worth it for a small business?
It’s worth it once you’re past roughly 100 to 300 chats a month and can no longer staff it consistently yourself, typically costing between 200 and 800 dollars a month depending on hours and volume, but it’s not worth it for a business getting under 20 enquiries a month.
Should a small business use AI chatbots instead of a human?
AI chatbots make sense once volume is high enough that a human answering every chat would need to be a dedicated hire; below that, a human answering promptly usually converts better because visitors can tell within a few messages whether they’re talking to a real person.
What’s the biggest mistake small businesses make with live chat?
Buying the software before deciding who, specifically, will answer it, which leads to slow or inconsistent replies that damage trust more than having no chat widget at all.
Where to check the details
Related reading: How to Use Live Chat for Lead Generation in a Small Business and Which Live Chat Software Features Matter Most for Small Business (And Which Are Just Noise).
If you want the full breakdown, here is everything I know about small business.
