Why Your Chatbot’s "An Agent Will Be With You Shortly" Is a Lie

"An agent will be with you shortly" is not a wait time. It is a sentence written once, by whoever built the chatbot, and shown to every customer regardless of whether the real wait is two minutes or twenty. Customers know this, which is exactly why the phrase reads as dismissive instead of reassuring.

TL;DR

• A canned wait-time message is the same words for every customer, no matter how long the actual queue really is.
• A real wait estimate needs two live numbers: how many people are actually ahead of this customer right now, and how long your team has actually been taking on recent conversations.
• Computing both from real data, instead of writing one sentence and reusing it forever, is what makes a wait time trustworthy instead of a guess dressed up as reassurance.
• QuiQSol's Conversation Flows chatbot hands off with a real queue position and an estimate computed from your own team's last several closed conversations, on WhatsApp and on your website's live chat alike.

Why does "an agent will be with you shortly" feel worse than a real number?

Because it carries no information. The same sentence appears whether one person is ahead in the queue or fifteen, whether your support team responds in ninety seconds or forty minutes. A customer reading it cannot tell whether to wait five minutes or give up and call instead, and that uncertainty is often more frustrating than a longer, real wait time would have been on its own.

What would it take to tell a customer their actual position in queue?

Two real numbers, checked at the moment of handoff rather than written once in advance. First, how many other conversations are currently open or pending, without an assigned agent, ahead of this one. Second, how long your team has actually been taking to close out recent conversations, not an aspirational SLA target but the real, measured average from the last several closed threads. Multiply the position by the average, and you have an estimate grounded in what is actually happening right now, not a sentence copied from a template.

Can a wait estimate be trusted if it is just a guess dressed up as a number?

Only if it is computed, not authored. A support team could just as easily write "about 8 minutes" into the same canned message and it would carry the same lack of credibility as "shortly," just with a number attached. The difference that actually matters is whether that number comes from a live count of the queue and a real average of recent response times, recalculated every time it is shown, or whether it is a static line someone typed in once and never revisited.

Does this work for a chatbot built for one channel, or across all of them?

The same mechanism works across every channel a Conversation Flow can run on: WhatsApp, a website's own live chat widget, and social direct messages. The queue position and wait estimate are computed the same way regardless of where the conversation started, because the underlying question, how many people are ahead and how long has this team actually been taking, does not change based on channel.

How do you build this without hiring a data team?

By reusing data your support tool already has. The queue count comes from conversations currently open or pending with no assigned agent. The wait estimate comes from the real gap between when your last several conversations were opened and when they were actually closed. Neither number requires new tracking or a separate analytics setup: both already exist in any conversation history worth calling a real system of record.

QuiQSol's Chatbot Starter template in Conversation Flows already hands off this way by default: greet the customer, qualify them with a few questions, and route to a human with a real, computed wait time the moment they ask for one. Clone it from the Journey Library, or start a free trial and see your own team's real numbers reflected back the next time a customer asks to speak to someone.