Stop buying leads: how to build a better pipeline with AI
Stop buying leads: why bought lists are expensive in practice, what GDPR requires, and how to build a better B2B pipeline with an ICP, signals and AI support.
Table of contents
Why should you stop buying leads?
We all know the type. A salesperson rings and says: "Would you like to buy a list of 10,000 decision-makers in your industry?" You say yes and get an Excel file. Then you discover that people changed jobs two years ago, that companies have closed, and that direct numbers go to reception.
Bought B2B lists go out of date quickly. People change title, company, phone number and email address, and companies close or change. The older a list is, the more of it is useless, and you pay in time for every call to a person who no longer exists there. The alternative is not a more expensive list, but a list you build yourself, from who your best customers actually are and when they are open to a conversation.
The key points in 30 seconds:
- The price per contact says nothing. It is the price per meeting that counts.
- A bought list tells you who exists. A list built on signals tells you when to call.
- You need to be able to document where your contact details come from. With a bought list you rarely can.
- Start with a sharp customer profile. The rest of the work becomes easier, and the waste smaller.
Why are bought lists expensive in practice?
There are five reasons why the low price per contact is deceptive.
- Out-of-date data. B2B contact data keeps changing: new titles, new companies, new numbers.
- No context. You get a name and a number, but not why now. The call becomes cold, and the conversation short.
- Everyone else has them too. Lists that are sold are sold to many. Your contacts receive the same calls from your competitors.
- Poor data quality costs time. Every call to a dead number is lost time for a salesperson. It is time that never comes back.
- Law. You do not always know the source, and then it is hard to meet the rules on information and documentation.
The calculation example: what does a meeting cost?
The price of a list is rarely the right comparison. The question is what a booked meeting costs. Here is a calculation example that only shows the method. The figures are invented, so use your own:
What a price per meeting looks like (invented example)
| Step | Bought list | Self-built list, based on signals |
|---|---|---|
| Number of contacts | 1,000 | 300 |
| How many are usable | One in three is out of date or wrong | Nearly all have been checked |
| How many reply | Few, because the call is cold | More, because there is a reason to call now |
| How many agree to a meeting | Very few | More, because the message fits |
| Work before the call | None | More (profile, sources, checks) |
The point is not that one always wins, but that you must calculate the whole way: the list price, wasted calling time, the work before the call, and the number of meetings at the other end. Compare what a meeting actually costs, and let the numbers decide.
What does GDPR say about bought lists?
When you contact a person, you process personal data, including in B2B. That raises three questions.
Where does the information come from? If you have not collected the information from the person yourself, you must as a rule inform them, and say where you got it from (GDPR Article 14). With a list whose source you do not know, this can be difficult.
What is your basis? Many use legitimate interest as the basis for B2B approaches. That requires a documented balancing between your interest and the person's, and that you meet the duty to inform.
What happens when someone says no? The person you contact has the right to object to direct marketing (Article 21), and you must be able to respect and remember that. Also check whether the company is protected in the CVR register, which means the information may not be used for marketing. And be aware that the rules on telephone and mail marketing in the Danish Marketing Practices Act depend on whom you contact and how.
Get a legal assessment of your specific set-up. This is not legal advice.
The alternative: build your own list, based on signals
A list that works is built on three things: a sharp customer profile, sources you are allowed to use, and a check before anyone calls.
Step 1: Define a precise customer profile
Quality lists do not start with a search tool, but with a sharp description of the target group. Not "all carpenters", but for example:
"Master carpenters in Greater Copenhagen with at least ten employees, who are growing and looking for new staff."
The sharper the profile, the less waste in every later step. Also write down whom you will not sell to. That saves just as much time.
Step 2: Find sources you are allowed to use
Good sources can be:
- The CVR register, where you can find companies by industry, size and development (respect the marketing protection)
- Job advertisements, which tell you what a company is in the process of expanding
- News and the companies' own websites, which show changes, new locations and new products
- Your own customers and network, which are the best sources of all
- People who approach you themselves after reading your content
Be careful with tools that automatically scrape information from sites such as LinkedIn. Many services prohibit it in their terms, and it creates both legal and practical problems.
Step 3: Check before anyone calls
Before a call is made, the list should pass a simple check:
- Does the company still exist? (CVR)
- Is the person still employed there?
- Is the contact on a suppression list, or has the person said no before?
- Is it a suitable time to call?
Without the check, you send your salespeople after ghosts.
Step 4: Use signals to choose the timing
The biggest advantage of a self-built list is the signals: signs that the company is open to a conversation right now. A bought list tells you who the customer is. The signals tell you when to call.
Examples of signals:
- The company has advertised new positions
- A new manager has taken up the post
- The company has opened a new location or launched a new product
- The customer has visited your pages or downloaded something
Examples of openings that use a signal:
- "Congratulations on the new job as sales director. I am calling because new managers often look at..."
- "I can see you are recruiting three new staff. Have you considered how to get them up to speed quickly?"
It is not cold canvassing; it is warmer canvassing. And it works because the conversation starts with something that interests the customer.
Step 5: Learn from each round
Write down which openings and signals led to conversations and meetings, and which did not. Use it to sharpen the profile and choose better sources next time.
Why do salespeople spend too little time on selling?
Salespeople typically spend a smaller part of their time on conversations. The rest goes on research, the CRM, administration and calls to the wrong people. Salesforce's State of Sales shows this year after year. A poor list makes it worse, because the time goes on calls that could never have led anywhere.
A better list is therefore not about saving money on leads, but about giving salespeople time for the conversations that matter.
Get customers to come to you
The best lead is the one that contacts you itself. Content that answers the questions your customers ask makes you visible when they search. It takes time, but it builds something that keeps giving. It is a more patient and more lasting route than a list you buy again every year. It is also a route where the information comes from the person themselves.
What can Agent360 do?
Agent360 does not sell lists and is not a lead engine. We build what makes your own pipeline better, on top of the systems you already use, as part of Your AI department:
- We look at how your sales and your lists work today, and where the time disappears.
- We finish one thing properly before we move on. Examples from the use cases page: a sales cockpit, where salespeople call from a call list with checks before each call (including suppression lists and calling hours), the conversation is written out in Danish, and AI suggests the outcome and drafts a mail for the salesperson to approve. And the content machine: a system that plans, writes and publishes articles and afterwards measures where they stand in Google, so more people find you when they search.
- You decide what the system may do on its own. Everything else waits for approval.
- The price depends on the task, and you receive a quote before you commit.
For AI-based calls we mention our partner JesperAI. It is their product, and we do not promise anything on their behalf.
FAQ: Buying leads or building your own lists
Q: Is it always bad to buy a list? No, but it must be compared on price per meeting, not price per contact, and the source must be documentable. A small, well-chosen list from a supplier who can tell you where the data comes from can be a sensible supplement.
Q: How quickly can you build a usable list yourself? It depends on the target group and the sources. A small list of a few hundred well-chosen companies can be built in a few days if the customer profile is sharp. Most of the time goes on defining whom you want, not on finding names.
Q: Is it lawful to call people based on public information? It depends on whom you contact, how, and what you have documented. Information must be processed lawfully, you must state the source, and you must respect objections and marketing protection. Get a legal assessment.
Q: Does it work in all industries? It works best where companies can be found and known from public sources, and where there are enough of them. In very small markets with few decision-makers, a personal network is often better than a list.
Q: What if a self-built list also gives few meetings? Then the profile is often too broad. Sharpen it, and read what the no's you get are saying. Improve the signals before you produce more volume.
Q: Do we still need salespeople? Yes. Good lists and tools remove the repetitive work, but the conversation, the relationship and the negotiation are still human work.
Q: Can Agent360 get us leads? No, that is not what we do. We help build the systems around your sales, so the salespeople spend their time on the right conversations, and so more people find you themselves.
Getting started
- Write your customer profile down on one page, including the customers you will not sell to.
- Calculate the price per meeting for your current list.
- Choose two or three sources, and build a small test list of a couple of hundred.
- Check before calling, and use a signal in the opening.
- Write down what worked, and repeat.
Would you like an honest estimate of what your pipeline could look like?
Book a free 20-minute consultation. Tell us how you find customers today, and get a view of what is worth building.
Related guides:
Contents
Should your sales reps spend their time talking?
We have built it for our own sales: the call is transcribed, the AI suggests the outcome, and the email is ready. We look at how your sales run today.
Book a reviewShould your sales reps spend their time talking?
We have built it for our own sales: the call is transcribed, the AI suggests the outcome, and the email is ready. We look at how your sales run today.
How we built it · Sales & follow-up
Sales cockpit with AI
A call list for salespeople. The call is transcribed in Danish, AI suggests the outcome (e.g. “Call again”) and drafts the email based on what the customer said.
- The salesperson calls
- AI writes transcript and email
- The salesperson sends
Related articles

Sales Ops Automation in Python: Automating What No-Code Cannot
When is no-code not enough for sales ops? Three Python automation examples (bulk sync, scoring, reports), pitfalls, requirements and who should own them.

Headless Sales AI: What Happens When You Separate Intelligence from the Interface
What is headless sales AI? Separate the AI logic from the user interface so it works in CRM, Teams and email. Benefits, risks and when it is worth doing.

Webhooks for sales teams: how to build a real-time data pipeline
Webhooks for sales teams: what they are, webhooks versus polling, the most useful triggers, the five typical mistakes, and how to start without overbuilding.