Our main outbound list was failing, and the numbers were bad enough to stop it. Only about 1% of the leads were qualified companies. The rest was padding: hospitals, universities, government administration offices, and job titles with no say in buying anything we sell.
It is tempting to treat a big list as a head start. Thousands of contacts looks like a pipeline. But every unqualified name costs a send, a sequence slot, and some of your sending reputation, and it tells you nothing about whether the market is working. You end up measuring the list, not the offer.
The cause was simple. The list was scraped and never enriched. There was no profile to check the person’s role, no revenue figure to confirm the company was the right size, and no industry tag to filter out public bodies. Without those fields, nobody could tell a hospital administrator from a manager at a company we could actually help until a reply came back, and by then the time was already spent.
There is a second problem that is easy to miss. Our systems are shared. When one rep uploads a raw scraped file, it lands in the same CRM and sequencing tools as everyone else’s data. Records without enrichment create duplicates, broken fields, and contacts that look live when they are not. The next person who pulls a list inherits that mess and has to clean it up before they can work. That is what made the decision for us. This is not one person’s campaign going wrong. It affects everyone using the system.
So we have changed the rules.
Self-sourcing from scraped lists is paused. Any custom list now needs approval before a single message goes out. The approval is a real check, not a formality. The list has to be enriched with the fields that make a contact usable, filtered against our ideal customer profile, and cleared of the wrong industries and non-decision-maker titles before anyone sees it.
Here is the practical lesson, whatever your stack looks like.
Define the ideal customer in terms of both the buyer and the company before you pull anything. “Companies with 50 to 500 staff” is not enough if it includes hospitals and government bodies. Exclude those sectors explicitly.
Enrich before you load. If a record cannot tell you the person’s role, the company’s size, and the company’s sector, it is not a lead yet. It is a guess.
Judge a list by its hit rate, not its size. Ask what share of the contacts you would be happy to see on a call. If that share is small, cut the list down rather than sending to all of it.
Keep one owner for shared systems. Whoever runs the outbound stack should decide what gets loaded, and that person should be able to say no without it becoming a debate.
Thousands of unqualified leads feel like activity. In practice they are the fastest way to damage your sending reputation, frustrate your reps, and make your data untrustworthy. Fewer, better-enriched lists will feel slower for a week or two, and then they will pay for themselves.