Finding and qualifying new contacts. The part that matters is the part that discards.
A list of two hundred names taken from somewhere, where thirty have closed, forty are too small to buy and two have been your customers for six years. This piece builds lists from public sources and throws away everything that doesn't hold up, before someone wastes a day on it.
First discard, then search. The exclusion criteria are decided with you and applied first, because a long list of wrong names costs more than a short list.
No row comes from bought lists. Only public sources that can be consulted, with the source and date attached to every contact.
The piece stops before sending. It finds, discards and verifies. Writing to people who have never contacted you has its own rules, and a system doesn't decide them.
This is the page for a single piece of the system. The other pieces, and the criterion for choosing which to start from, are on the services page.
What changes in practice
The salesperson's day stops being a day of research. The work of opening fifty websites, looking for the right name, working out whether the company is the size you need and finding a reliable contact is the work that shifts to the machine.
The second change is that the list is short. Twenty rows that pass every criterion are worth more than two hundred raw names, because nobody calls the two hundred and in the end you go back to calling the usual acquaintances.
A long list isn't an asset. It is the postponement of a selection job someone will have to do anyway, usually at the worst moment.
Bought lists cost you the domain you write from
The damage is technical before it is legal. Google's rules for bulk senders require spam complaints to be kept below 0.10% and never to reach 0.30%, and a list collected by others crosses that threshold within a few hundred sends.
Those who send more than 5,000 messages a day must also authenticate the domain with SPF, DKIM and DMARC and offer one-click unsubscribe, requirements in force since 1 February 2024. When a domain's reputation is ruined, the quotes to real customers end up in the spam folder too.
Then there is the part concerning the rules. The Italian exception that allows writing to a customer without asking for new consent applies only to those who have already bought from you: the Garante's guidelines limit it to those who use the email details provided by the data subject in the context of the sale of a product or service.
A new contact, by definition, isn't in that position. How to proceed from there is an assessment for you and your adviser, and this piece stops before it so as not to make it on your behalf.
Deduplication comes before everything
It is the check almost nobody puts first, and the mistake that follows is the most expensive. Every row is compared with active customers, open deals, contacts approached in the past and those who have objected to being contacted again.
Introducing yourself to a long-standing customer as a stranger says in one line that inside the business nobody knows who the customers are. And it is a mistake that multiplies when two people sell in the business, because each works on their own list.
That is why deduplication needs a reliable history, which is what the adaptive CRM keeps. Work on the contacts that are already yours, on the other hand, is a different piece, done by customer qualification and reactivation.
What the system does, step by step
Every row carries the source it came from, the date and the reason it passed or was discarded.
| Step | What happens | What you get |
|---|---|---|
| Criteriaexcluding ones first | The reasons a company doesn't interest you are written down with you: size, market, distance, type of business. |
The list that arrives is already made of companies you could really serve. |
| Collectionpublic sources | Company websites, Chamber of Commerce lists, trade fair exhibitors, trade associations, public professional profiles. |
Every contact has a declared origin, and nobody has to trust a file that came from who knows where. |
| Deduplicationagainst what you have | Customers, open deals, contacts already approached and objections are compared before anything else. |
Nobody introduces themselves to a long-standing customer as a stranger. |
| Verificationon the live page | The existence of the contact details and of the company is checked on the page as a browser sees it, not on a quick scan. |
Fewer dead rows, and no wrong statement about whoever you are about to contact. |
| Deliveryshort list | The list arrives ordered by priority, with the reason each row is there and what is known about that company. |
The salesperson opens the list and gets going, instead of starting the research over from scratch. |
An automated scan says false things, and it has to be verified
This is the part learned by getting it wrong, and we write it down because nobody talks about it. A tool that opens a page and looks for a contact form or contact details finds false absences all the time: forms that appear only after an animation, pages behind a cookie banner, content that arrives after the first load.
The result is a list that declares a door closed when it is open, and whoever uses it writes wrong sentences to a company that immediately notices the mistake. So verification is done on the live page, as a browser sees it, and ambiguous cases are looked at by hand.
The opposite applies too, and it is why the list stays short: when in doubt a row is discarded. A lost contact costs one contact; a wrong row costs the credibility of the first sentence you write.
What goes out on its own and what waits for a person
Collection, deduplication, verification and building the list go out on their own. They are reading jobs on public information, and none of them reaches a person outside the business.
No outward contact goes out on its own. The first sentence to a company that doesn't know you is the most delicate thing on the whole route, and a person decides it and sends it.
On phone calls there is an extra constraint that applies in Italy: the Public Register of Objections (Registro Pubblico delle Opposizioni) now covers all national numbers, landline and mobile, and those who run phone campaigns must consult it periodically and in any case before every campaign. The reasoning about the human constraint is on the page about the principles we build with.
The same thing changes name with the trade
The method is identical; the useful source isn't. It is worth looking at your own case, because almost always the best source is one you already know and have never used systematically.
| Sector | The source that works | Where you see it |
|---|---|---|
| Food and agricultureand export | The exhibitor and visitor lists of trade fairs, foreign distributors' catalogues, import registers by country. |
|
| Hospitalityaccommodation and events | The agencies and professionals who bring groups, local companies that organise meetings, the portals where accommodation is already listed. |
|
| Restaurantsand bars | Companies based nearby for business lunches, offices and building sites that have recently opened, those who organise courses and events in the area. |
The service of a dining room where the phone rings unanswered |
What this piece doesn't do
It doesn't buy or sell lists, and doesn't use databases bought from third parties. If a row has no public source that can be named, that row doesn't go in.
It doesn't send anything. Research and sending are two different jobs, with different responsibilities, and keeping them separate is how you avoid a system deciding on its own to write to someone.
It doesn't promise a response rate. That depends on who writes, what they write and the market, and anyone who promises a percentage before knowing your first sentence is selling an invented number. What is measured here is the quality of the list, not the outcome of the conversation.
Questions and answers
Do you buy contact lists?
No, never, and it isn't a moral stance. A bought list contains addresses collected by someone else for reasons you don't know, so it produces bounces and spam complaints that damage the domain you write from.
Google's rules for bulk senders require spam complaints to be kept below 0.10% and never above 0.30%: a dirty list crosses that threshold on its own.
Where do the contacts come from?
From public sources that can be consulted: company websites, Chamber of Commerce registers and lists, trade fair exhibitors, trade associations, public pages of professional profiles.
They are the same sources you would look at one by one, and every row carries the source it came from and the date, so it can always be traced back.
Do you send the emails?
No, this piece stops before that. It finds, discards and verifies; sending is a separate activity with its own rules, to be looked at case by case with your adviser.
The Italian exception that allows writing without consent applies only to those who have already bought from you: the Garante's guidelines limit it to those who provided their address in the context of a sale, so it doesn't cover new contacts.
How do you avoid contacting people who are already our customers?
With deduplication, which is the first check, not the last. Every row is compared with active customers, open deals, contacts approached in the past and those who have objected: if it appears in one of those lists it doesn't go on the list.
Introducing yourself to a long-standing customer as if you didn't know them is the kind of mistake you pay for in a single phone call.
How do you measure whether it's working?
With the share of rows that survive a manual spot check, which must be high: if three out of twenty rows checked are wrong, the list isn't ready. The second number is the share of contacts discarded and the reason, because that is where you see whether the criteria are working.
The third, later on, is how many real conversations came from that list, counted one by one.
Notes on sources
- The 0.10% and 0.30% thresholds on spam complaints, the limit of 5,000 messages a day, SPF, DKIM and DMARC authentication and one-click unsubscribe are in Google's sender guidelines, in force since 1 February 2024. They apply to mail delivered to Google accounts, which is a large part but not all of the world's email; other providers have their own, similar rules.
- The limit of the exception for email to your own customers is in the Guidelines on promotional activity and the fight against spam of the Garante per la protezione dei dati personali, 4 July 2013, which refer it to Article 130(4) of the Privacy Code. It is a rule, so it has no sample. We don't provide legal advice and don't tell you what you can send: that assessment stays with you and your adviser.
- The coverage of the Public Register of Objections, extended to all national landline and mobile numbers, and the obligation to consult it periodically and before every phone campaign are described by the register itself.
- The problem of false absences in automated scans is our own working observation, not a research finding: it comes from forms loaded after an animation and from pages behind a cookie banner, which look absent to a quick read. We write it down because it determines how the verification is built.
- This page doesn't report results obtained for a client, because this piece hasn't yet been delivered to a client. The tests cited are functional checks carried out in testing.
The other pieces in this group
Customers who are out there, and don't reach youFifteen minutes, with your case in front of us.
How many hours a week does your salesperson spend looking for names instead of talking to people? If nobody has ever counted, that is already the most useful information. In fifteen minutes on the phone we look at it together and tell you where it makes sense to start, even if we don't end up working together.
You get Mattia Esposito, who then builds the system: there's no salesperson in between. If you'd rather measure on your own before talking, the Diagnostico (in Italian) is twenty questions and five minutes.