By KYN AI Advisory Team — AI implementation specialists, Singapore
Answer a "send me more information" reply with a short message that does two things: it gives one useful piece of information, and it asks one question that shows whether the person has a real need. Do not send a brochure and wait. The reply is a soft signal. It can mean genuine interest, polite exit or a request from someone who will never decide. Your reply should sort those three in one exchange.
What does "send me more information" actually mean?
It means one of three things, and the sender rarely says which.
- Real interest, low effort. They want to look at something before committing time.
- A polite no. It is easier to say "send info" than "no".
- Wrong person. They will forward it, or never open it.
You cannot tell from the words. You can tell from what they do next. A question back, a named colleague or a date is a signal. Silence after a brochure is the other signal.
So the goal of your reply is not to deliver the information. It is to learn which of the three you have, at the lowest cost to both sides.
What should the reply contain?
Four parts, in under 90 words.
- A thanks in one clause. No flattery.
- One relevant fact or link. Pick the single item that matches their business, not the full deck.
- One qualifying question. About the need, not about the budget.
- One next step with a time. An offer of 15 minutes, with two slots.
Template:
Thanks, [Name]. The page that matches a [industry] team of your size is here: [link]. It shows [one specific thing].
One question so I send the right material: is the issue [A] or [B]?
If it helps to talk it through, I have [day, time] or [day, time] for 15 minutes.
The A or B question is the useful part. It is easy to answer in one word. It also splits the three meanings: real interest answers, a polite no does not.
Which qualifying questions work?
Ask about the situation, not about their authority. A direct question on budget in the second message reads as pressure.
- "Is this for the Singapore entity or the group?"
- "Are you handling this by hand today, or in a tool?"
- "Is there a date you are working to?"
- "Who else would look at this with you?"
The last question finds the wrong-person case early without accusing anyone. A reply that names a colleague turns into a new contact for review. Do not email that colleague automatically. A person approves first. See reply classification for how a wrong-person reply is logged.
How should an agent handle these replies?
An agent can do most of the work if the rules are clear.
- Classify. The reply is labelled "question" or "interested", with the flag "info request".
- Pick the asset. Rules choose one link from your approved library by industry and size. The agent does not invent a document.
- Draft. The agent fills the template: thanks, link, question, two calendar slots read from the rep's calendar.
- Review. A rep approves or edits. Replies that carry your name wait for a person.
- Set the follow-up. If there is no answer, the agent drafts one follow-up at day 3 and a last one at day 10, each adding something new. Both wait for approval.
- Close the loop. After the last follow-up, the contact is marked "info sent, no response" and the sequence stops.
The agent should not attach a price list unless one is approved for general use. If the reply asks for pricing, the draft says a person will confirm and the item goes to review. For the rule on price, see why sales agents need hard limits on discounts.
What goes wrong with these replies?
Four failure modes appear often.
- The brochure dump. Five attachments, no question. The reader opens none.
- The three-paragraph pitch. The reply repeats the original email at length.
- Fake urgency. "Only two slots left this week" with no real constraint. It reads as pressure and damages trust.
- Endless follow-up. Eight nudges. After the second, you are adding to the spam rate, not the pipeline.
Cap the number of follow-ups in code, not in a reminder. Two is a reasonable limit for an info request. Mark the contact and stop.
How do you measure whether it works?
Track four numbers per month.
| Measure | What it shows | |---|---| | Info requests received | Volume of soft signals | | Answered qualifying question | Share that were real | | Calls booked | Share that became a conversation | | No response after two follow-ups | Share that were polite exits |
If the answered-question rate is low, change the question first, then the asset. If calls booked is low but answers are high, the problem is the next step, not the interest.
Keep a record of the A or B answers. After 50 replies, you will see which need your market names most often, which is useful for your site and your first line.
What does a worked example look like?
Take an illustrative reply from an operations manager at a logistics firm: "Thanks for your note. Please send more information."
The agent labels it "question" with the flag "info request". Rules pick the library page for logistics firms with a small team. The draft reads:
Thanks, Hui Min. The page for logistics teams is here: [link]. It shows how a warehouse team reconciles supplier invoices against receiving.
One question so I send the right material: is the pain more in matching invoices, or in chasing late deliveries?
If it helps to talk it through, I have Tuesday 10:00 or Wednesday 15:00 for 15 minutes.
A rep approves it in under a minute. Three outcomes are possible. She answers "matching invoices", and the rep books a call. She names a colleague, and the colleague becomes a draft contact for review. Or she says nothing, and the day-3 follow-up goes to approval. Each outcome has a defined next step, and none depends on the agent guessing.
What if the reply asks for a price list or a deck?
Some senders ask for a specific file. Answer what they asked, with limits:
| Request | Response | |---|---| | Price list | Send only if an approved public list exists. Otherwise say a person will confirm and route to review | | Company deck | Send the one-page version for their industry, not the full deck | | Case studies | Link to industry-level summaries; never name a client without written consent | | References | A person decides; route to review | | Security details | Link to the security page and offer a call with a named person |
Keeping this table in the system means the agent never improvises an attachment.
Where KYN fits
KYN builds the reply handling above as part of a sales system in your own cloud account. The agent classifies the reply, picks the approved asset, drafts the answer and reads open slots from the calendar. Every action is logged with the actor. Anything outbound waits for a named person. Sorting one reply costs about S$0.01 in a live KYN system for an insurance brokerage, depending on the model and the text length.
The first 14 days are a review of your stack and interviews with your team. A full build ships in under 90 days. Rights in the custom code and configuration pass to you on full payment.
Frequently asked questions
Should I send a PDF or a link?
Send one link to the page that matches their industry. A link tells you whether it was opened if your tracking allows it, and it can be updated. Attach a PDF only if the recipient asked for a file.
How soon should I reply?
Reply the same working day. A short answer within hours does more than a long one next week. Set your own response target, measure it and review the misses.
Is it acceptable to follow up if they do not answer?
Yes, twice, each with something new. After that, stop and mark the contact. A third nudge adds little and raises complaint risk.
Checked 30 October 2026.
Curious whether Marketing agents fits your business? Talk to KYN on WhatsApp — no forms, just a conversation.
