46.7%of private messages arrive after hours
Case study
4 hours a day back, and no potential client without a reply
Comments and DMs on Facebook and Instagram, answered for them. 524 people left their name, phone and email, and about 3 in 10 became clients, the association's own figure.
1 · Where we started
Under every ad there's a queue of people asking
ADifesa is a banking and financial protection association based in Lissone, Italy. It works for people who were charged costs they never owed on a salary-backed loan, a revolving credit card or a loan sold with an insurance policy, and it tries to get that money back.
Who they are
You pay only on what's recovered
The assessment and the claim are free; the fee comes only out of what is recovered. The office is in Lissone, in Brianza, and cases are handled across Italy.
What already worked
The completed forms
The Meta campaigns bring in completed forms, and that part was going well.
The problem
Everything around the forms
Comments pile up under every ad, and direct requests land in the message inbox. These are people describing something private in public: how much they owe a bank.
Whoever writes there isn't just browsing
They have just read the ad and want one thing: to know whether their case qualifies. If nobody answers for a day, they take that question somewhere else.
2 · From comment to record
What happens after a comment
Someone comments under an ad. The automation replies in public, opens the private conversation and collects the details the adviser needs.
- 1
The comment
The person writes under an ad or a post, on Facebook or Instagram.
- 2
The public reply
After a deliberate three-minute wait, the automation replies. If the comment contains a phone number, it hides it.
- 3
The private message
In parallel it opens the private conversation and asks for phone, email and the times to call back.
- 4
The record
Name, phone, email and product: what an adviser needs to pick up the phone.
3 · The rules
The job is working out whether the case qualifies
Anyone can reply "message us privately". The value is knowing whether there is a case, and the association's rules are tight.
- 10
active eligibility rules
- 5
product types
- 11
banks and lenders named
Each rule has its own conditions
It isn't a keyword list: each rule has its own window of years and its own conditions.
- Revolving cards
- opened between 2000 and 2017, with a wider window for two issuers
- Compass loan
- from 2011 to 2020
- Deutsche Bank or Findomestic loan
- from 2017 to 2022, above a certain amount
- Business credit line
- companies only, trading for five years and above twenty thousand euros
- Investment losses
- yes
- Cryptocurrency
- no
Before that, working out what's being discussed
Nobody writes "salary-backed loan renewed in 2016" in a comment. They write the way they talk, and from those words the system has to reach the right product, because the product decides whether a case exists.
"They withhold a fifth of my salary"
Salary-backed loan"I have a card that never ends"
Revolving card"They sold me insurance along with the loan"
Loan with insurance policy
By the numbers
May 1 to September 16, 2026
Production data.
- 2,329
conversations handled across comments and private messages
- 1,033
qualified contacts
- 524
complete records: name, phone, email, product
- 3 in 10
of those records become a client, with at least one case
the association's figure
From comment to case. The last step is the association's figure, not ours.
The first number we look at isn't the total. It's the 524 complete records, which is what an adviser needs to pick up the phone already knowing what the call is about.
Conversations per month2,329 in total
523May
470Jun
July · Instagram760Jul
355Aug
221Sep
From July, Instagram direct messages come in too. Volume follows the association's campaigns, which were scaled back from August. September through the 16th.
- Different people
- 1,794
- Handed to a person
- 111 (4.8%)
- Private messages received
- 2,951
- Private replies written by the automation
- 1,676
- Public replies under comments
- 1,262
Where the 2,329 conversations came from
- 1,065From a comment under an ad
- 232From a comment under a post
- 1,032From a direct message
4 · After hours
Half of these people write when the office is closed
Of the 2,951 private messages received, 1,377 came in outside working hours: 46.7%.
1,377 private messages after hours
- 850On weekdaysbefore nine or after six
- 527At the weekendSaturday and Sunday
267 between ten at night and seven in the morning
It's the easiest number in the project to underrate. It says almost half of these requests arrive when there is nobody there to read them. They come on the evenings and Sundays when someone sits down with their statements, and that is exactly when they decide to write.
5 · The person
111 conversations out of 2,329 were handed to a person
Fewer than one in twenty. The automation closed everything else on its own.
4.8%handed to a person
Control stays with the client
It switches off one conversation at a time
This is the piece a demo never shows. From the control panel, the association switches the automation off on a single conversation and carries on by hand, without touching the others. From that moment the chat is theirs and the bot stays out of it.
No dancing around the fee
It answers in public
The association does get paid, as a percentage of what it recovers. In the comments, "what does this cost me" comes up every day, often with an edge to it. The reply says the assessment and the claim are free and that you pay only if the refund arrives, and it doesn't try to move the question into private. With an audience that has heard too many promises already, an evasive answer is worth less than nothing.
6 · Mistakes
What we got wrong, and fixed
If we only told you what worked, it would be an ad.
"They withhold a fifth of my salary" was landing in spam
The problemThe filter discarded anyone talking about a fraction of their salary, to keep out generic venting. But a fifth is literally a fifth: the association's main product was being mistaken for noise.
How it endedFixed.
The opening message went out twice
The problemSomeone already in a chat, coming back to comment under an ad, received a second "how can we help you?".
How it endedNow if the conversation is already open, the second message doesn't go out.
Reels shared in private restarted the bot
The problemInstagram sends a shared video as a message with no text, and the automation replied to every share.
How it endedNow an attachment with no words gets no reply.
On Instagram every contact was called "User c168f1"
The problemThe name was being read from a field that doesn't exist on Instagram.
How it endedFixed, and the old names recovered: from 33 nameless contacts down to 4.
7 · Clients
How many of those contacts become clients
Our database stops at the complete record. What happens next, once the adviser calls, only the association knows, so we asked them.
3 in 10complete records that become clients, with at least one case opened
It's their number, out of their system rather than ours, and we report it as they gave it to us. It isn't 30% of everyone who comments: it's 30% of the people who reached the end of the conversation and left a name, a phone number, an email and a product.
What these numbers don't say
We don't count the human work after a handover, because from that point the conversation continues inside Meta's own tools and not in our system. And the outcome of the eligibility check isn't recorded as such: the adviser makes that call, with the documents in hand.
How many refunds came through after that, and for how much, is not our figure and doesn't belong here. A case study that promises revenue without holding the revenue figure is an ad in disguise.
In detail
Behind the scenes, if you want to see it
Numbers in the open, Instagram and the cost of a viral post.
Numbers in the openA phone number under a post is a problem, not a lead
It happens a lot: someone types their phone number into the comment, in the open, and wants a call back. Meanwhile anyone passing by can read that number.
- 1
The comment
It contains a phone number or a sensitive detail.
- 2
It hides it
The automation treats it as an incident to close, not a lead to bank.
- 3
It moves to private
The conversation moves into private messages.
This applies under ads and under organic posts, on Facebook and on Instagram.
3 minutes
A deliberate wait
Under comments, the public reply comes after three minutes. That isn't slowness: a reply that appears the same second someone finishes typing tells everyone there is a machine underneath.
InstagramInstagram came later, and it shows
The project started on Facebook. Instagram direct messages took an extra step: five permissions to be approved by Meta, each with a demonstration video.
- May
The first submission
It didn't go through.
- May 28
Five videos re-shot
With a different method, showing the chain of identity between the Meta portal and the system. This time it went through.
Meta asked for a demonstration of a function that doesn't exist in its own API: editing an Instagram comment. You can create, hide or delete one; you cannot edit it. The demonstration was rewritten around what the API actually does.
Qualified contacts from Instagram
6May
29Jun
117Jul
22Aug
6Sep
September through the 16th.
The costThe post that cost eighteen dollars a day
In July one of the association's posts went viral in the worst way: thousands of angry political comments, not one of them a case to take on.
About $18a day, answering nobody
The automation read them one by one and correctly classified them as irrelevant, but to know that it had to call the model first.
A counter, not stricter filters
Every post keeps a tally of the comments seen and how many turned out to be spam. Past a threshold, that single post is muted and its comments stop reaching the model.
- 257
posts tracked, in the window observed
- 37
muted, in the window observed
A muted post still lets through the words that point to a real product, like the name of a lender or "salary-backed loan".
We checked it against real comments, not invented examples, before it went live: no good case lost out of thirty-five tried, and all ten genuine leads recognized.
Frequently asked questions
What people ask us most
How many conversations did the automation handle?
From May 1 to September 16, 2026 it handled 2,329 conversations across comments and private messages on Facebook and Instagram, with 1,794 different people. They produced 1,033 qualified contacts and 524 records complete with name, phone and email.
Does it answer public comments too, or only private messages?
Both. Under a public comment it replies after a deliberate three-minute wait, and opens the private conversation in parallel. In the window observed it wrote 1,262 public replies and 1,676 private messages.
What happens if someone posts their phone number in a comment?
The comment is hidden and the conversation moves into private messages. A phone number sitting in the open under an ad is a risk to the person who wrote it, not a lead to collect.
When does a person step in?
Whenever the client decides: from the control panel the automation is switched off on that single conversation and the team carries on by hand. It happened on 111 conversations out of 2,329, fewer than one in twenty.
How many of those contacts become clients?
According to the association, 3 out of every 10 complete records become clients with at least one case opened. That is their figure, from their system: our database stops at the complete record, meaning name, phone, email and product.
How long does it take to set up?
It depends on how well defined the rules are. The long part isn't the software, it's writing down exactly when a case qualifies and when it doesn't: here that meant ten rules with windows of years, minimum amounts and named lenders. For Instagram direct messages, factor in Meta's approval of the permissions as well.
If you'd like to see how we work before we talk, here's how we build automations.
Do you run campaigns on Meta? There's already a queue of people under your ads asking you something
The question is whether your rules are written well enough for someone else to answer in your place. Thirty minutes, and we'll tell you where we'd start.
Want to see more cases? Systems already at work in other businesses, with their numbers.
All case studies