Enrichment is one of those words that gets used to mean almost anything. Someone offers to enrich your file, you say yes because it sounds like an improvement, and what comes back is a spreadsheet with more columns in it. Whether any of those columns are true is a separate question nobody asked.
What it actually is
Enrichment means taking the records you already have and filling in what is missing, or correcting what is wrong, by matching them against another source.
That is it. It is not modelling, it is not segmentation, and it is not a substitute for having decided who you want to reach. It makes an existing list more complete and more accurate. Nothing more, and that is plenty.
The three things it genuinely fixes
Gaps. You have a business name and a phone number but no industry, no size, no address. Matching against a reference source fills those in, which is what makes selection possible. You cannot target by company size on a file that does not record company size.
Decay. Business data goes off faster than people expect. Companies move, change names, get acquired, go under. People leave. A file that was accurate two years ago is now wrong in a meaningful percentage of its rows, and you have no way of knowing which ones by looking at it.
Duplicates and mess. The same business appearing four times with four spellings. Phone numbers in six different formats. Addresses that do not resolve. None of this is glamorous and all of it costs you money on every campaign you run.
What it does not fix
A badly chosen audience. If the list is the wrong people, a more accurate version of the wrong people is still the wrong people. Enrichment improves the record, not the decision behind it.
Permission. This one catches people out. Appending an email address to a record does not give you the right to email it. Consent is attached to the person and the relationship, not to the data field. A beautifully enriched file can still be one you are not allowed to contact.
The gap between a file and a campaign. An enriched list still has to be selected from, and then something has to actually be sent.
Verification is the part worth paying for
The useful distinction is between appending and verifying.
Appending adds data. Verifying confirms the data you already hold is still true, and tells you when it is not. Appending makes the file look better. Verification makes it work better.
The question to ask a supplier is not how many fields they can add. It is what proportion of your existing records they can confirm as current, and what they do with the ones they cannot.
Your own data, improved
Most of this applies to the file you already own, and that is where the best return usually sits. Your customer list is the most valuable data in your business and often the worst maintained, because nobody is responsible for maintaining it.
Clients can upload their own data into Hive and have it matched, cleaned and filled in against our reference sources. We work with an independent business data provider for the verification layer, so the check is against an external source rather than our own assumptions.
What comes back is the same customers, described properly. From there you can model against them, find more households that look like them, and run a campaign without exporting anything to a third party.
How to tell whether you need it
Take a hundred records from your file at random. Ring ten of them. Count how many are still correct.
It is a crude test and it will tell you more than any vendor conversation. If eight out of ten are fine, your file is healthy and enrichment is a marginal gain. If it is five, you have been running campaigns at half efficiency and nobody noticed, because an undeliverable record does not complain.
Not sure what state your data is in? Send us a sample and we will tell you what matches, what is stale and what is missing, before you commit to anything. Get in touch.
