B2B lead database buying guide: coverage, accuracy and cost per contact

Do not compare databases by headline size. Test each one on the same 200 people you already know you want to reach, and measure three things: how many it finds (coverage), how many of the emails and phones are right (accuracy), and what you pay per usable contact (cost). Then ask each vendor where the data comes from, how often it is refreshed and what happens to the credits you do not use.

Every database says it is the biggest and the most accurate. Those claims cannot be checked from a sales page, and they are measured in incompatible ways. The only comparison that matters is how a database performs on your own buyers. This guide gives a way to run that test in an afternoon.

What are you buying?

A B2B lead database is a searchable set of people and companies with fields: name, role, company, location, size, industry, and often an email address and a phone number. You use it to find who fits, and to get the way to reach them.

Sellinger offers a database of over 700 million contacts. You search people and companies, and find verified emails and phone numbers. We state the size because it is a fact about our product. We do not use size to argue it is better for you. The test below does that.

How do you test coverage?

Coverage is the share of the people you want that the database actually holds.

1. Take 200 people you know you want to reach, from your own customers, a past list or LinkedIn. Mix seniority and company size the way your real list will.
2. Look each one up. Count those found with the right role and company.
3. Divide: found divided by 200.

Run the same 200 through each vendor. You will see big differences by region, industry and seniority. A database strong in US software may be thin for logistics in Germany. If the data is weak where you sell, its size is irrelevant.

How do you test accuracy?

Accuracy is the share of the details that are correct.

• Role and company. Check a sample of 30 against LinkedIn. People change jobs; count the stale ones.
• Emails. Run the emails through a verification step, then send to a small batch you can afford to lose and count the bounces. Do not send a cold campaign from a domain you care about to find out.
• Phones. Call or text a small sample if you plan to use them, or compare against a number you know.

Why care about bounces? Google asks bulk senders to keep the spam rate under 0.30% and advises aiming for 0.10%. High bounce rates also damage the sender reputation that gets your mail delivered. Our post on Google and Yahoo bulk sender rules has the details. A database whose emails bounce is not cheap at any price.

How do you work out cost per usable contact?

Use one formula:

Cost per usable contact = total spend divided by contacts you can actually contact

A contact is usable when the person matches your buyer, the role is current and the address or profile works. Here is a worked example, labelled as made up so you can swap in your own numbers.

Check the arithmetic: A is 1,000 × 0.60 × 0.70 = 420, and $100 divided by 420 is about $0.24. B is 1,000 × 0.85 × 0.92 = 782, and $300 divided by 782 is about $0.38. In this example A is cheaper per usable contact, though B is more accurate. The point is that you cannot tell from the price list, and that accuracy only matters when it changes the number you pay per usable contact. Change the inputs and the answer changes. For a small list, a higher accuracy can be worth paying for when each wrong contact costs you something, such as a damaged domain.

On Sellinger, a lookup of an email, a company or a person is $0.01, so the cost per usable contact is the lookup price divided by your own hit rate. A 70% hit rate on a $0.01 lookup is about $0.014 per usable contact. Usage modules start with a $10 credit, which at that price is enough to run the 200 contact test many times over.

Which questions should you ask every vendor?

• Where does the data come from, and is it collected or bought?
• How often is each field refreshed, and when was it last checked for this record?
• What does "verified" mean here? Checked how, and when?
• How are credits priced: per search, per record, per export? Do unused credits expire?
• Can I export, and is there a limit?
• How are opt outs and deletion requests handled?
• Which countries are covered, and under which legal basis?
• What happens to my uploaded lists?

Pay attention to the answer on "verified". A claim without a method is a claim.

What about compliance?

Having an address is not permission to use it. EU and UK rules differ from the US, and some countries require consent. The data protection side, for named people, applies even when the data is public. Read is cold email legal and ask the vendor how it handles suppression and erasure. In Sellinger, before every send on email and LinkedIn the platform checks your suppression list, and an erasure function anonymizes matching leads and keeps the identifiers from being imported again.

When is a database not the answer?

When the best leads come from a reason to write. On our own LinkedIn outbound, signal based lists replied at 54.6% against 37.2% for untagged ones and 21.7% for a Sales Navigator search (case study). A database tells you who exists. A signal tells you who is paying attention. Many teams use a database to find the contact details for people a signal pointed to. The comparison of signal based and database prospecting goes through the trade.

Smaller, named lists can also beat scale. The Green Shield Logistics study wrote to named heads of quality, ESG, procurement and logistics, and 8.2% of named owners replied against 2.2% of generic inboxes (case study). A precise list from a smaller source can outperform a large unfiltered one.

A buying checklist

• [ ] I have a list of 200 people I know I want to reach.
• [ ] I ran the same list through every vendor.
• [ ] I measured coverage, role accuracy and email accuracy.
• [ ] I worked out cost per usable contact for each.
• [ ] I asked where the data comes from and how fresh it is.
• [ ] I know what happens to unused credits.
• [ ] I checked the rules for the countries I will contact.