The Complete Guide to Pre-Chat Forms: What to Ask and Why

Before someone starts a live chat conversation with you, you have about five seconds to decide what to ask them. Pre-chat forms sit in that narrow window between clicking the chat widget and typing the first message. The complete guide to pre-chat forms comes down to one principle: ask for exactly what you need, nothing more, and present it so fast that visitors don't abandon before the conversation even starts.
Get this right, and you improve agent efficiency, enable follow-up, and route conversations to the right person. Get it wrong, and you've built a barrier that keeps potential customers away before they've even said hello.
Why pre-chat forms exist (and when they're actually worth the friction)
When a visitor starts a chat without a pre-chat form, your agent knows nothing. No name, no email address, no idea why they're there. The first 90 seconds of the conversation is spent asking basic questions: "What's your name? How can I reach you? What are you looking for?" Meanwhile, the visitor is frustrated because they've already figured out what they want — they just want to talk to someone about it.
Pre-chat forms flip this. By the time the conversation reaches your agent, there's context. They know who they're talking to, how to follow up if the connection drops, and often what the visitor actually needs. The agent can jump straight to solving the problem instead of playing 20 questions.
There's a second, equally important reason: follow-up. If your chat disconnects, if the browser closes, or if the visitor reaches you while your team is offline, an email address means you can actually reach them. Without it, that lead — and whatever they were interested in — is gone forever.
But here's the thing: this friction only works if you keep it minimal. Chat analytics from high-volume chat platforms consistently show that every additional field you add reduces completion rates. Ask too much, and you've built a form instead of a conversation starter.
The golden rule: ask less, not more
This is where most businesses fail. They design a pre-chat form the way they design a contact form, cramming in everything they'd "like to know." Phone number, company name, account ID, detailed description of the problem — the whole works.
Then they wonder why people abandon before completing it.
Here's what research actually shows: form completion drops sharply with each additional field. A two-field form converts way better than a five-field form. Three fields is generally the ceiling before abandonment rates climb noticeably. And there's a legal argument too — the ICO's data minimisation principle under UK GDPR says you should only collect what's necessary. "What would be nice to know" doesn't qualify.
The actual question to ask is not "what information would help?" but "what information is essential for this conversation to work?"
If you're asking that seriously, the answer is usually three things.
The three essential fields (and why everything else is noise)
Name. A first name lets your agent personalise the conversation from message one. "Hi Marcus, how can I help?" lands completely differently than "Hello, what can I do for you?" This small detail makes the interaction feel like a conversation, not a support ticket. Required field.
Email address. This is arguably the most important one. It does three jobs: it's your fallback if the connection breaks, it enables offline message delivery if your team is busy, and it creates a record for follow-up. Required field.
Department or topic. A dropdown or radio button set — "Sales," "Support," "Billing," or whatever makes sense for your business. This one is genuinely optional (unlike name and email), but if you have multiple teams or specialists, it's worth the single extra field because it routes people to the right person on the first try, eliminating frustrating transfers.
That's it. Name, email, optionally topic. Fifteen seconds to complete. Everything an agent needs to start a productive conversation.
You might think this is too minimal. You'd be wrong. Once the conversation starts, your agent can ask follow-up questions if they need them. In a real conversation, context emerges naturally. In a form, every extra field is just friction.
Fields you should actually skip
Phone number. Unless you're calling people back, a phone field adds friction for no reason. People chose chat specifically because they don't want to call. Asking for their number feels like you didn't hear them.
Company name. Useful in pure B2B contexts, but for most small businesses? The agent can ask in the conversation if it matters.
Order number or account ID. This one's tricky. If your chat is exclusively for existing customers with accounts, maybe. But if you support both new prospects and returning customers, this field confuses new visitors. They don't have an order number and won't know what to enter. You lose them before they start.
Detailed message or "describe your issue." This is the most common mistake. A large text area asking visitors to write out their problem is self-defeating. They clicked the chat widget because they want a conversation, not to write a support ticket. The friction is massive. Save this question for the actual chat.
How to design a form people will actually complete
Beyond field selection, presentation matters enormously.
Label clearly. "Your name" beats "Please enter your full legal name." "Email" beats "Email address (we'll use this to send you a transcript)." Brevity increases completion.
Use placeholder text. Show an example: "[email protected]." It clarifies what you want without cluttering the label.
Make the button action-oriented. "Start chat" or "Connect me" works better than "Submit." The button should reinforce that you're entering a conversation, not submitting a form to the void.
Keep it visually light. A compact, clean form signals "this will take 10 seconds" instead of "this is an official process." Avoid unnecessary borders, backgrounds, or decorative elements.
Test on mobile. More than half of web traffic comes from mobile now. Your form needs to work on a small screen with a touch keyboard. Fields should be large enough to tap without frustration, and the form shouldn't require scrolling.
Chat widgets that handle pre-chat forms should let you customise all of this — field labels, placeholder text, button text, and styling — without code. If your platform doesn't, you're using a platform that wasn't designed for actual businesses.
Making the data count
Collecting information means nothing if you don't use it.
Personalise immediately. If you have a name, use it in the first message. "Hi Tom, what brings you in today?" versus a generic greeting sets a completely different tone.
Route based on topic selection. If someone selects "Sales," connect them to someone who can discuss your offerings. "Support" goes to someone with access to account history. This avoids the frustrating ping-pong of being transferred three times.
Store the email for follow-up. Make sure the email address gets logged into your CRM or contact management system. This creates a record you can actually act on later.
Analyse the patterns. Over time, what topics are people selecting most? Are certain departments getting disproportionate volume? This tells you where your website messaging might be missing the mark. Chat analytics should show you this automatically.
When to skip the form entirely
Pre-chat forms aren't always the right call.
Skip them if speed is your priority. Every barrier reduces chat starts. Some e-commerce businesses find that removing the form entirely increases volume noticeably — an effort-reduction principle that Harvard Business Review covered. Agents can ask for details during the chat. Volume might matter more than context.
Use the form if follow-up capability is critical. If your business model depends on contacting visitors after the chat — you're selling something, closing deals over multiple conversations — then collecting an email upfront is worth the friction cost.
Use the form if routing prevents chaos. With multiple departments, a topic selector keeps things sane. Without it, every chat lands with whoever's available, then bounces around.
Skip it for returning visitors. If your system recognises repeat visitors through cookies or login status, don't ask for information you already have. Show the form only to first-time visitors.
Two metrics that actually matter
Form completion rate. What percentage of people who see the form actually finish it and start a chat? Below 50%, you're asking too much.
Agent efficiency. Are your people spending less time on admin questions at the start of conversations? If the form works, time-to-first-substantive-response should drop because context is already there.
These two metrics will sometimes point in opposite directions. A form that achieves 100% completion but collects useless data isn't helping anyone. A form that collects detailed information but only 30% of visitors complete it loses too many potential conversations. Find the balance.
For measuring success in your chat setup, these metrics paired with completion rate give you the full picture.
Frequently Asked Questions
Q: Should I make all fields required, or allow optional ones? A: Make name and email required. Topic/department can be optional if you're willing to route less efficiently. Most businesses find one optional field is fine; any more than that and people skip them all.
Q: What if a visitor enters a fake email? A: Some will. You can't prevent it without verification (which adds friction). Accept that a small percentage of visitors will throw in a junk email and decide if that's acceptable. Many businesses find that 90% legitimate + 10% trash is fine because you're still capturing most of the genuine leads.
Q: Can I use conditional logic — different fields based on earlier answers? A: Yes, and it's smart. If someone selects "Sales" from a topic dropdown, you might show them different follow-up fields than someone who selects "Support." This reduces visible form complexity while collecting relevant data. But only do this if your chat platform supports it well; poorly executed conditional logic feels clunky.
Q: What if most of my visitors are returning customers? A: You should only show the form to new visitors. Use cookies or login status to identify returning people and let them skip straight to chat. They shouldn't have to re-enter their name every time.
Q: Can I A/B test different pre-chat form configurations? A: Absolutely, and you should. Test a form with two fields vs. three fields. Test required vs. optional topic selection. Test different label wording. Platforms like Relentify let you run these tests without code, tracking completion rate and chat volume to see which version actually works better for your business.
Q: What happens if the chat disconnects after they've filled the form? A: The email (and name, if you collected it) stays in your system. You can reach out via email, either to reconnect them or to follow up on their original question. This is the entire point of capturing email in the first place.
Q: Should I apologise in the form itself for asking questions? A: No. Don't say "Sorry, we need a quick bit of info before we connect you." You're not inconveniencing them; you're preparing your team to have a better conversation. Confidence is fine.
Q: What if my form is in multiple languages? A: Keep the same minimalist approach in every language. Don't add extra fields for non-English-speaking visitors. A three-field form translates cleanly.
The bottom line
Pre-chat forms are small but meaningful. Three fields. Clear labels. A quick submit button. Done.
The difference between a pre-chat form that works and one that doesn't isn't complexity — it's restraint. Collect a name and email. Route based on topic if you have multiple teams. Then let the conversation start.
If you're setting up live chat for the first time, here's how to add chat to your website in under 5 minutes. Once it's live, spend more time automating routing than designing forms. And if you're building followup workflows, capture that email and use it — that's where the real value lives.