The slow follow up, and the second message that never got sent

The deal that did not die, it just sat down

Slow follow up is how a business loses sales nobody ever decided to give up. The first reply is usually fine. The second and the third are the ones that go missing, because sending them depends on a person recalling an unfinished conversation with no prompt in front of them. By the time anyone remembers, the customer has bought from whoever wrote back first.

The enquiry came in, you answered it properly, you sent the price. The customer said they would confirm, or that they were discussing it at home, or that they would come by on the weekend. Then the weekend passed. Nobody said no. Nobody said yes. The conversation is still sitting there in your phone, technically alive, doing nothing.

Multiply that by a few months and you have a shadow pipeline: a large number of people who were interested enough to ask for a price and never heard from you again. It is not that anyone decided to abandon them. Each one was individually forgotten by a busy person on a busy day, which is a much harder problem than laziness because there is nobody to blame and therefore nobody to correct.

Speed on the first reply gets discussed constantly. The gap between the first reply and the second is discussed almost never, and in most businesses it is far wider.

Why the follow up never gets sent

Four reasons account for nearly all of it, and only one of them is about effort.

The first is that follow up lives in memory. There is no artefact that says a customer is due a nudge on Thursday, so Thursday relies on somebody's recall on a day when three other things went wrong. The second is that in most small businesses the person who sold the job is also the person delivering it, so the moment work comes in, following up on the next one becomes the lowest priority thing on the list.

The third is that there is no agreed cadence, so every follow up is a fresh decision. When is too soon. When is too pushy. What do I even say. A decision that has to be made from scratch every time is a decision that gets postponed. The fourth is that nobody has defined what a no sounds like, so nothing ever leaves the list. The list grows until it is too big to face, and then it stops being opened at all.

• No record anywhere that a nudge is due, so it depends on recall
• The seller is also the deliverer, so live work always wins
• No agreed interval, so every follow up is a new decision
• No script, so the message feels awkward to write
• No definition of lost, so the list only ever gets longer

Count the contacts after the quote

Take last month's quotes, however you record them, and go through them one by one. For each, write down how many times you contacted that person after the quote went out. Not how many messages in total, only the ones after the number was given.

Almost every business doing this exercise for the first time finds a column full of ones and zeroes, with the occasional two. Seeing that written down does more to change behaviour than any amount of encouragement, because it stops being a vague sense that follow up could be better and becomes a specific list of people who were quoted once and then left alone.

Do one more pass. Mark each old quote as either closed on purpose, meaning somebody actually decided it was dead, or simply old. The size of the simply old pile is the amount of undecided work your team is carrying around in their heads, and carrying it is what makes the whole list feel too heavy to open.

The free fix: write the ladder down once

A follow up ladder is one page. Three contacts, fixed intervals, fixed wording, and a rule for when the lead closes. It costs nothing and it removes the decision that was causing the delay, because nobody has to work out what to say or when.

The intervals matter less than their being fixed, but something like two days, five days and twelve days after the quote works for most sales here. What matters much more is that each message has a job. A message whose only content is a request for an update gives the customer nothing to respond to, and it is the reason people feel awkward sending it.

The third message is the one most businesses never send and the one that reliably produces replies. It says, politely, that you are going to stop unless they tell you otherwise. People who were genuinely interested answer that message. People who were not can stop feeling rude, and you can close the row and get the weight of it out of your head.

The final piece is mechanical. Book the follow up at the moment of the conversation, while you are still on the call or still in the chat. A calendar entry, a phone alarm, a next action date in your sheet, any of them. A follow up scheduled later, when you have a minute, is a follow up that does not exist.

• Message one, around two days later: add something they did not have, such as a photo, a stock update or a delivery time
• Message two, around five days later: give a reason to decide now, such as a price validity date or the last slot in a schedule
• Message three, around twelve days later: say you will stop following up unless they want you to continue
• Then close the row as lost, on purpose, with one line about the reason
• Book each of these at the moment of the conversation, never afterwards

What automation adds, and where it is the wrong answer

The ladder is the kind of thing systems are genuinely good at. The messages go out on the day they are due, in the wording you approved, whether or not the person who owns the lead is on site, ill or busy. The lead closes itself on schedule if nobody replies. Nothing depends on recall.

There are rules that shape this on WhatsApp specifically, and they need deciding before anything is built rather than on the morning it launches. Business API messaging runs on opt in, on message templates approved in advance, and on a messaging window that closes after the customer's last reply.

Once that window has closed, only an approved template will go out at all. In practice that means taking permission at the moment the person first writes in, and designing the first two rungs to land while the window is still open.

This is the wrong project in three situations. If your enquiries were never qualified in the first place, a ladder simply produces polite refusals more efficiently, and the fix sits upstream in who you are attracting.

If someone has already said no, running them up the ladder is how a business collects blocks and complaints, so the closing rule has to be honoured by the system and not just by people. And if your close rate on first contact is healthy and your real shortage is enquiries, follow up discipline is not your constraint and you should spend the effort on demand instead.

A follow up you decided to send later is not scheduled. It is remembered, and remembering is the part that keeps failing.

The ladder once it is running, and who stands at the end of it

Write the ladder first and run it with your own team for a month. If three contacts at fixed intervals in fixed wording actually go out, keep it in-house and spend the money on something else, because the ladder is the value and the sending is the cheap part. The reason to automate is not speed. It is that sending stops depending on whether the person who owns the lead is on site, ill or in a meeting, which is a staffing shape problem rather than a discipline problem.

No price appears on this site. What moves the number on a follow up build is how many channels the ladder runs on, whether your quote record can be read by anything at all, how many approved templates the sequence needs, and who maintains it afterwards. The audit takes the first week and the ladder is live inside a fortnight, because a fixed sequence in approved wording is one of the shortest builds there is.

Two rules keep a ladder from becoming a nuisance. It stops the moment the person replies, and that reply goes to a human rather than into another automated step. Anything about a price, a discount or a complaint waits for human approval. The ladder itself, the wording, the intervals and the record of who received what, sits in accounts under your own logins, so you own the system and can change a sentence yourself at nine on a Monday without asking anybody.

The published version of this is Patel Group, where growth had outrun the systems, leads were being managed informally and follow ups were missed, and an AI driven CRM was built around those exact bottlenecks. Massive Capital is the harder edged one: an AI calling layer over a Meta acquisition engine, with conversion up 40 per cent after the calling went in and outreach cost down 60 per cent. Haad owns growth and client solutions at Wobble, which works from Karachi, bills month to month and carries 25 engagements across six countries.

Common questions

How many times should I follow up before giving up?

Three contacts after the quote is a reasonable default for most businesses here, ending with a message that says you will stop unless they say otherwise. What matters more than the count is that the number is fixed and written down, because an undefined ladder either stops at one message or continues until the customer is annoyed.

How long should I wait between follow ups?

Something like two days, five days and twelve days after the quote suits most sales cycles here, tightened for anything urgent and loosened for large purchases people discuss at home. Fix the intervals rather than judging each case, because judging each case is what caused the delay in the first place.

What do I say so it does not sound like nagging?

Give every message a job other than asking for an update. Send a photo they have not seen, confirm that the item is still in stock, name the delivery date they would get if they confirmed today, or state how long the price holds. A message that carries new information is a service. A message that only asks whether they have decided is a request for a favour.

Should follow up be automated or done by a person?

Automate the timing and the standard wording, and keep a person for anything that involves a discount, a complaint or a change to the offer. The most reliable design is a scheduled message that stops the moment the customer replies and hands the conversation to a named human being.

Can I message someone weeks after they went quiet on WhatsApp?

Only within the platform rules, which allow an approved template outside the window that follows their last reply. Keep it specific to the exact thing they asked about rather than sending a general offer, and respect the opt in you took at the start. A vague promotional message to an old enquiry is the fastest way to be blocked.