The honest answer: it depends on who answers
A chat widget — the little message bubble in the corner of a website — is neither magic nor junk. It is a doorway, and a doorway is only worth what is on the other side.
Answered within a minute or two during business hours, chat catches people who were never going to call and were not going to bother with a form. Left unanswered, it does worse than nothing: a visitor who types a question into a dead chat box leaves more annoyed than if the widget had never existed, because now your business has ignored them personally.
So the question is not whether your website should have chat. The question is whether someone on your team can actually answer within a couple of minutes, during hours you are willing to state publicly. Everything else in this article follows from that answer.
When chat genuinely wins leads
Chat earns its place when:
- Your customers have quick pre-sale questions with short answers: do you service my area, do you carry this, roughly what does it cost, can I get in this week.
- A meaningful share of visitors browse from work or late at night, where a phone call is awkward but typing is easy.
- Someone is already at a computer all day — a front desk, a dispatcher, an office manager. For them, one more tab is nearly free.
- Your competitors respond slowly. In local services, the first responder wins a large share of the work, and being the business that answers in under a minute wins jobs on speed alone.
Chat tends to disappoint when every real conversation needs a site visit anyway, or when nobody is at a desk. A solo contractor on a roof is not answering chat, and should not pretend to.
The staffing reality for a small team
Be honest about coverage before installing anything. The rules that make chat sustainable for a small business:
- Show the widget only during hours someone will answer. Every decent product can do this; outside those hours it becomes a leave-a-message form and says so plainly.
- Name one owner per shift. Chat that is everyone's job is no one's job.
- Route it to a phone. Modern widgets ping a cell app, which is how a two-person shop keeps chat alive from a job site.
- Write out answers to the ten questions you always get, so most replies are paste-and-personalize rather than composition.
- State the expectation inside the widget itself — replies in about a minute during business hours.
If reading that list makes it obvious your team cannot cover it, that is a useful answer, not a failure. Pick a tool from the next section instead.
Chatbots, contact forms, and missed-call text-back
The alternatives, honestly:
- A contact form is the baseline, and you keep it either way. Its weakness is speed — by the time you reply in the evening, the visitor already called someone else at two in the afternoon.
- A simple chatbot — an automated responder — is at its best doing intake: collecting a name, the need, and a phone number, and promising a callback. Having it answer real questions is a separate, harder decision. A bot that takes a message well beats a bot that answers badly.
- Missed-call text-back may fit a phone-first business best of all: when you cannot pick up, the caller instantly gets a text asking what they need, and the conversation continues by text. My own business runs phone-first, and this class of tool matches how local customers actually behave — they called you, after all.
You do not need all of these. Pick the one that matches where your leads already come from.
Measure jobs, not conversations
Decide the verdict with numbers, on a 60-day trial:
- Count conversations per week. If almost nobody opens the widget, remove it with a clear conscience.
- Ask every chat lead the same question you should ask every caller — how did you find us — and tag chat-sourced work in whatever you track jobs in, even a spreadsheet column.
- Compare the funnel: chats started, real conversations, quotes sent, jobs won. Chat that produces two extra jobs a month pays for itself many times over for most service businesses; chat that produces pleasant conversations and no jobs does not.
- Watch your median response time. If it drifts past a few minutes, either fix the staffing or shrink the stated hours until the promise is true again.
How to decide this week
- Answer the coverage question in writing: who answers, during which hours, on which device.
- If coverage exists, install a widget on a free or inexpensive tier — $0 to roughly $50 a month covers most small-business needs — set honest hours, and load your ten canned answers.
- If coverage does not exist, set up missed-call text-back or a message-taking bot instead, and keep the contact form prominent.
- Run 60 days, tag every lead source, count jobs.
- Keep it, fix it, or kill it based on the count.
I build websites for a living at HashWhales and I have told plenty of clients not to install chat. The widget is cheap. Pretending to be available when you are not is expensive.