Live chat at 200 visits a day: when it pays off
Two hundred visits a day can justify live chat — or just add another queue. Weigh what the conversations are worth against the cost of staffing them.
Gabriel Andrade
In this article
A site with 200 visits a day gets about 6,000 visits in a 30-day month. That sounds like enough traffic to add live chat. But traffic alone does not tell you whether chat will earn its cost. The useful questions are how many visitors will start a conversation, what those conversations are worth, and whether you can answer them when you promise to.
A chat button can help a visitor decide, resolve a blocker, or find the right product. It can also create a new queue that someone must monitor. If visitors see “We’re here to help” and wait 20 minutes for a reply, the button has made a promise your operation cannot keep.
Start with the conversations, not the visitor count
First, check what “200 visits a day” means in your analytics. Visits may include repeat sessions, visits from people who already know your company, and traffic with no buying intent. A product page visit is not the same as a visit to your careers page or a blog post.
Then estimate the number of conversations you might receive. Since you do not have your own chat data yet, use scenarios rather than treating an industry average as a forecast. For 6,000 monthly visits:
- If 0.5% start a chat, that is 30 conversations a month.
- If 1% start a chat, that is 60 conversations a month.
- If 2% start a chat, that is 120 conversations a month.
Those figures are planning assumptions, not conversion benchmarks. Your rate could be lower or higher depending on the offer, traffic source, page, chat placement, and whether you show the chat proactively. A prompt that opens after a visitor has spent time on a pricing page will attract a different set of questions from a chat bubble displayed on every page.
For two weeks, record visits by page, chat starts, the visitor’s question, the time the conversation began, and what happened next. Count a conversation once, even if the visitor sends several messages. Separate sales questions from existing-customer support. This will tell you whether chat is serving a buying decision or simply moving existing support work into another place.
Put a value and a cost on the same page
A simple model can help you decide whether a staffed live-chat test is worth running:
Monthly value = incremental outcomes from chat × contribution per outcome
Monthly cost = staff time + software + setup and maintenance
Use contribution margin, not total revenue. If a sale brings in $500 but costs $350 to deliver, the contribution is $150. For a service business, you might instead use the margin from a new client or a qualified appointment that reliably progresses to a sale.
Here is an example using the middle scenario of 60 conversations a month. Suppose 30 are genuine buying questions. If chat helps you win three additional sales that you would otherwise have lost, and each contributes $300, the estimated monthly value is $900. If the people and tools needed to staff chat cost $1,100 a month, the case does not yet work on direct sales value alone.
The word “incremental” matters. A customer who would have bought anyway is not a sale created by chat. Compare similar visitors or use a short test to estimate whether chat changed the outcome. Ask new customers how they found you, review the path they took, and avoid crediting chat for every order that followed a conversation.
Calculate labor using the coverage you actually intend to promise. Sixty conversations that arrive across a month may sound small, but you still need someone available during the hours the chat is open. A few clustered questions during a busy hour can be harder to handle than a larger volume spread across the day. Include the time to check for new messages, research answers, and pass unresolved issues to another person.
Choose the service model that fits the work
“Live chat” can mean several different things. You can staff a real-time conversation, let visitors leave a message, or use a chat window to offer a quick route to email, phone, or another channel. Each choice creates a different promise and workload.
| Approach | Visitor experience | Operating cost and workload | Best fit |
|---|---|---|---|
| Staffed live chat | A person replies during stated hours | Requires scheduled coverage and quick response | Buying decisions where a timely answer can change the outcome |
| Message form in a chat window | Visitor leaves a question and receives a later reply | Less need for continuous monitoring; follow-up must be reliable | Low or uneven volume, or questions that need research |
| Chatbot with human handoff | Immediate answers for a limited set of routine questions | Requires accurate content and a clear escalation route | Repetitive questions with stable answers and occasional complex cases |
| No chat; existing channels only | Visitors use the channels already offered | No new queue, but a visitor may have to choose or wait | Low-intent traffic or a support operation already near capacity |
Do not call an asynchronous message service “live” if no one is watching it. State when a reply is likely, and make sure the stated window matches your operating hours. You can also offer email, phone, or messaging options where they suit your customers. WhatsApp is one option in some markets, not a universal default; visitors may prefer email, SMS, or a website conversation.
For a small operation, a shared inbox can make the message-window option more workable: for example, FlipDesk brings website chat alongside WhatsApp, Instagram, and Facebook messages so a team can handle those conversations in one place. That is useful only if the team can still meet the response times it promises.
Check whether chat solves a real buying obstacle
A chat window earns its place when it removes friction at a moment that matters. Look for questions that repeatedly stall a decision: whether a plan includes a particular feature, which size or configuration fits, how delivery works, or what happens after purchase. If the answer is short and the visitor is close to acting, a timely response can make a meaningful difference.
By contrast, a chat button may just add a queue if visitors mostly ask questions that your product page should answer. If ten people in a month ask about delivery time, improve the delivery information first. If the same request needs a specialist to investigate every time, chat may collect the question but cannot resolve it in real time. Fixing the underlying information or escalation path may help more customers than adding a new channel.
Review where chat starts, not just how many times it opens. Compare the pricing page, product pages, checkout, and support pages. If most conversations come from existing customers asking about order status, you have learned something about support demand—but that is not evidence that live chat is improving sales. If chat attracts high-intent visitors who need a quick answer before they book or buy, the sales case is stronger.
Test it without promising more than you can deliver
Set a short, bounded test with one audience and a defined schedule. For example, offer staffed chat on the pricing page on weekdays from 10 a.m. to 2 p.m. for four weeks. Tell visitors those hours clearly. Outside them, show how to leave a message or reach you through another channel.
Before the test starts, write down what would count as success. Use a small set of measures that connect workload to value:
- Qualified conversations: chats that involve a real buying question or an issue the team can resolve.
- Response and resolution: how quickly a person replies during advertised hours and how often the visitor gets a useful answer.
- Outcomes: appointments, completed purchases, or resolved issues that can reasonably be linked to the conversation.
- Workload: total handling time, peak-time backlog, and whether other channels become slower.
A high number of chats is not automatically a good result. If a chat channel generates 60 conversations but leaves email unanswered or absorbs time without changing customer outcomes, it may be a worse service design. Track the effect on the rest of your operation, not just the new channel’s activity. For a practical starting point on measures, see the customer service indicators guide.
Decide what happens after the test
At the end of the test, compare actual results with the assumptions you made. If chat produces valuable conversations at a manageable workload, expand coverage gradually. Add hours or pages one at a time and check that response quality holds. Do not switch on proactive prompts everywhere just because the first few conversations went well.
If demand is too low or too scattered to justify a live shift, keep the option to message you asynchronously—or remove the chat window and improve the information visitors need. If demand is real but the team cannot keep up, narrow the hours, focus the channel on high-intent pages, or make the handoff clearer. The answer may be to change the promise, not to add more people or another tool.
With 200 visits a day, live chat is neither obviously too early nor automatically worthwhile. It earns its cost when it helps enough visitors take a valuable next step, and when you can deliver the response you advertise without making every other queue worse. Measure those two things before you make the chat bubble permanent.
