Skip to content
Chatbotscape
Intercom's article on automatically messaging website visitors, Zendesk's article on proactive chat for key visitors and Microsoft's Dynamics 365 proactive chat page were read on 9 November 2026. Plans, limits and menus change, so check each vendor's current documentation.
Proactive chat· Conversation design
Proactive chat is a chat or chatbot message that appears on its own, triggered by something the visitor does or where they are, rather than waiting for the visitor to open the chat widget and write first.
By Chatbotscape Editorial· Methodology· Published 10 November 2026· Updated 10 November 2026

Proactive Chat — When the Chat Speaks First

Quick answer: In reactive chat the visitor opens the widget and asks. In proactive chat a rule decides that this visitor, on this page, at this moment, gets a message first: "Comparing plans? I can help." The visitor can ignore it, and a good setup treats that as a normal outcome, not a failure.

Proactive versus reactive chat

Reactive chatProactive chat
Who starts itThe visitorA rule on your side
Typical triggerClick on the widgetPage, time on page, visit count, referrer, exit movement
Visitor intentAlready expressedGuessed from behavior
Main riskVisitors who never find the widgetInterrupting people who did not want help
Where it fitsEverywhere the widget is shownA few pages where help is likely to be wanted
Where to read moreLive chatThis entry

Most sites run both: the widget stays available, and proactive messages are added on a small number of pages.

Triggers vendors document

Microsoft's Dynamics 365 page lists eight example scenarios: time spent on a page, number of visits (counted with a cookie), leaving the page after a set time, arriving from a particular referrer page, the visitor's country, a date window such as a sale, and a consent or pre-chat survey prompt. Intercom's article gives an example that targets visitors who land on a pricing page from a specific promotion, shows the message after 10 seconds on the page, and lets you combine conditions with AND and OR. Zendesk's article describes three: returning visitors (previous visits greater than zero, excluding people already helped), shopping cart pages after more than 60 seconds, and high-value pages such as pricing.

Common pattern: page plus time, sometimes plus a visit count. Nothing here makes one trigger better than another; the right one depends on your pages and your visitors.

What the vendor pages do and do not say

  • Frequency limits. Intercom's article does not describe frequency caps or per-visitor limits. Zendesk's does not describe cooldowns. Microsoft's page documents no limits on invite frequency. Check your own platform before assuming a message will not repeat.
  • Plans. Zendesk's article says Basic plans are limited to two proactive chat triggers and lists proactive chat under its suites. Intercom's article says you need its Messages product. Neither is a pricing guide.
  • Side effects. Zendesk's article says chats started by a proactive trigger skip the pre-chat form, so you may not get a name or email, and routing that the form would have handled can be bypassed.
  • Annoyance. Zendesk's article does not discuss how to avoid it beyond noting the visitor can choose whether to respond. Intercom's article recommends chat-style messages and shows a time-on-page delay, but gives no frequency guidance.
  • Results. None of the three pages gives a measured conversion figure. This entry does not either. If you read a percentage elsewhere, look for the source and the method before using it.

What happens to the visitor

  • Nothing. They ignore it and carry on. That is fine.
  • A reply. The chatbot or an agent continues the conversation. See human handoff for how it passes to a person.
  • A dismissal. The visitor closes it. If the same message returns on the next page, it becomes an irritation.
  • A form skipped. As noted above, the usual pre-chat questions may not be asked.

Why it matters for metrics

A proactive message changes who starts conversations, so conversation counts rise without any change in demand. Measure proactive and visitor-started chats separately, or an apparent improvement may only be volume you created. Track dismissals and the conversations that follow, not only the number sent. See chatbot abandonment rate and chatbot conversion rate.

FAQ

What is proactive chat? A chat or chatbot message sent automatically when a rule matches, such as time spent on a pricing page, instead of waiting for the visitor to start.

What is the difference between proactive and reactive chat? In reactive chat the visitor starts; in proactive chat your rule starts it. Proactive chat guesses intent from behavior, so it needs more care about where and how often it appears.

What are common proactive chat triggers? Time on a page, a returning visit, a cart or pricing page, a referrer, a country, a date window, or leaving the page. The Dynamics 365, Intercom and Zendesk pages each document some of these.

Does proactive chat annoy visitors? It can, especially if it repeats or appears on pages where people are reading. The vendor pages read here give no figures, so test on a few pages and watch dismissals.

Is a proactive chatbot the same as a proactive live chat? The trigger idea is the same. With a chatbot the first message and the follow-up can be automated; with live chat an agent may need to be available. See live chat.

Can I limit how often a proactive message shows? It depends on the platform. Several vendor pages are silent on frequency limits, so check yours.

Sources