The person chatting usually cannot say yes
In anything above a small purchase, the person who opens a chat is rarely the person who signs. They are researching for a manager, building a shortlist for a committee, or checking one specific fact that came up in a meeting they were in and you were not. Treating that conversation as a failed sale is the common mistake. It is a different job, and it has its own way of being done well.
The distinction matters because the two jobs call for opposite behaviour. Talking to a buyer, you work towards a decision. Talking to a researcher, you work towards a document: something accurate, quotable and easy to pass on. Push a researcher towards a decision they have no power to make and you get a polite exit.
Find out early, without interrogating
You do not need to ask who signs the cheque, and asking directly in the first minute reads badly. The information arrives on its own if you listen for it. People announce their role constantly without being asked, in the pronouns they use and the questions they choose.
- We are looking at and they want to know usually mean a group decision is underway
- Questions about security, procurement or contract terms rarely come from the person who will use the product
- A request for something in writing is a request for something to forward
- Very specific single questions often mean somebody was sent to find one answer
When you do ask, ask about the process rather than the person. Who else will want to look at this before you decide? is a normal, useful question that nobody minds answering, and it tells you everything the blunter version would have.
Arm them for the meeting you will not attend
Once you know you are talking to an advocate rather than a buyer, the goal changes. They are going to represent you in a conversation you cannot join, probably days later, probably against an alternative you never get to answer. Everything you give them should survive that trip without you.
That means being specific about the things committees actually argue over, which are rarely the features. Price and what changes it, what the setup involves and who has to do it, what happens if it does not work out, and what the alternative would have cost. An advocate who can answer those four questions confidently is worth more to you than one who liked the demo.
Write answers that are built to be forwarded
Chat is a fast medium and its natural register is casual and fragmentary, which is exactly wrong for a message that will be pasted into an email to somebody senior. When you reach the part of the conversation that will travel, change gear deliberately.
- Answer in complete sentences that make sense without the question above them
- Put numbers, dates and conditions in explicitly rather than referring back to earlier messages
- Avoid internal shorthand and product jargon nobody outside the chat will parse
- Link the page that says the same thing, so the claim has a source that is not just you
Sending the transcript at the end of the conversation is the cheapest thing you can do here. It turns your careful answers into a document your advocate already has, rather than something they have to reconstruct from memory.
Price questions from a non-buyer are different
When a researcher asks about price they are usually not negotiating. They are collecting a number to put in a comparison, and vagueness at that moment is genuinely costly, because a blank cell in somebody’s spreadsheet is easy to drop. Give the clearest number you honestly can, say what moves it, and say what is included so the comparison is not made against a competitor’s stripped-down figure.
If your pricing genuinely depends on details they do not have yet, give a worked example rather than a refusal. A team of five on the standard plan comes to this is something they can carry into the meeting. It depends on your requirements is not, and it is the sentence that gets you left off the list.
Keep the thread alive without chasing
Internal decisions take longer than the person in your chat expects, and going quiet is normal rather than a rejection. The useful move is to leave a door open that does not require them to perform enthusiasm: offer to answer questions from whoever else looks at it, by name if they will give you one, and make it easy to come back with a follow-up rather than a commitment.
Ask what happens next and when, then respect the answer. An advocate who says the review is in three weeks has told you exactly when to be useful again, and following up before then makes their job harder rather than easier.
How MyLiveChat fits
Two things carry most of the weight. Emailing the transcript when the chat ends puts your written answers directly in the advocate’s inbox in a form they can forward, and every conversation stays searchable in transcripts, so when a colleague of theirs turns up next week the second agent can read the first conversation instead of starting from nothing.
The agent console matters more than it sounds here too: transfer carries the full transcript with it, so passing a research conversation to a colleague who knows the procurement answers does not cost the visitor a retelling.
What to measure
These conversations look like losses in any funnel that only counts immediate outcomes.
- How many chats end with no decision but a clear next step, tracked separately from dead ends
- How often a second person from the same company appears in chat within a month
- Time from first research chat to a decision, which is usually far longer than anyone assumes
- Whether transcripts sent at chat end correlate with those companies coming back