Is live chat worth it for a small business?

The useful test is not whether everyone else has a chat bubble. It is whether people get stuck on a question that a human could answer in one or two lines, right at the moment they are deciding.

If three or more of these sound familiar, chat will probably earn its keep. Every chat that ends in 'great, ordering now' is a customer who might otherwise have quietly closed the tab.

• You get the same pre-purchase questions by email every week: sizes, delivery times, whether it works with something they already own.
• Your product or service needs a little reassurance before someone pays, such as custom orders, bookings or software trials.
• Visitors leave from your pricing or checkout page and you never find out why.
• Someone on the team already sits at a computer during working hours and can glance at a chat between other tasks.
• Your customers would rather type a quick question than pick up the phone.

Chat that nobody answers is worse than no chat. A visitor who types a question and stares at an empty window for twenty minutes learns something about you, and it is not flattering. These are the cons that matter more than any feature list.

• Nobody can watch it. If everyone is on the shop floor, on the road or in meetings all day, chat becomes a slower email with a worse first impression.
• Your questions need research. If each answer means checking an order system, asking a supplier or reading a contract, an email you can answer properly serves the customer better.
• You get very little traffic. A handful of visitors a day rarely produces enough chats to justify building the habit of watching for them.
• Your help pages are a mess. Chat will turn into the place people ask what a clear FAQ page should already say. Fix the page first; it is cheaper.

The subscription is usually the small part of the bill. The big part is attention: chat promises a quick reply, so someone has to notice it quickly.

Chat also interrupts. A message arriving while you write a quote breaks your concentration, and getting back into it takes a few minutes every time. Decide who is on chat for which hours, so the rest of the team can ignore it without guilt.

There is a hidden saving too. A two-minute chat often replaces an email thread that drags over two days. Questions that would bounce back and forth ('Which model do you have?' 'The blue one.' 'The old blue one or the new one?') get settled while the customer is still on your site, instead of after they have bought elsewhere.

You do not need round-the-clock cover. You need clear hours and an honest promise. Here is the setup that works for a team of one to five.

If you want to test this cheaply, suppp's free plan includes the live chat widget and a shared inbox for one person and 100 conversations a month, which is enough to find out whether visitors use chat at all. The widget goes on your site by pasting one script tag and can be styled in your own colours. Business hours, the pre-chat form and saved replies (canned responses) come with the Starter plan, and new accounts get a 14-day Starter trial with no credit card. Every plan alerts you to unanswered live chats, so a busy afternoon does not turn into a forgotten customer.

• Set your hours and show them. Outside those hours the widget should say you are away and ask for an email address, instead of pretending someone is there.
• Say how fast you reply. 'Usually replies within a few minutes' sets the expectation up front, so a three-minute wait feels normal rather than rude.
• Greet as a person. For example: 'Hi, I'm Anna. Ask me about sizes, delivery or returns, I'm here until 5pm.'
• End your away message with a time. For example: 'We're offline until 9am tomorrow. Leave your email and we'll reply first thing.'
• Save your most common answers. Delivery times, return steps and password resets should be one shortcut away, then tweaked so they sound human.
• Give each shift one owner. 'Everyone keeps an eye on chat' means nobody does.
• Ask only for what you need. A name and email before the chat starts is enough to follow up if the visitor leaves mid-conversation.

Give it four weeks, then read the chats themselves, not just a count of them. The conversations tell you whether chat helped customers decide or just gave them a new place to say hello.

If chat brought in even a few customers you would otherwise have lost, and nobody on the team dreads it, keep it. If it mostly produced 'hello?' at 11pm and the same three questions, limit it to business hours, fix your help pages and try again later.

• Count the chats that ended in a purchase, booking or signup, or clearly removed the last obstacle to one.
• List the questions that came up more than twice. Those belong on your website, next to where people asked them.
• Check how many chats waited too long or got no answer. If that happened often, the problem is cover, not chat.
• Ask whoever ran it how much it interrupted their day. Winning a few sales by wearing out your only support person is a bad trade.