Somebody will tell you this year that your data is the thing holding your outbound back, and that the fix is to stack five providers instead of using one.
For most companies running cold outreach, the honest answer is no, you do not need it, and one question tells you which side of that you are on. It is this: is the number of companies that genuinely match your criteria bigger than the number of people you need to contact this year?
If the answer is comfortably yes, skip it and put the money into verification and volume. If the answer is no, coverage is your binding constraint and stacking providers is the cheapest fix available to you.
What waterfall enrichment actually is
The mechanism is simple and there is nothing wrong with it.
You take a list of people you want to reach. You run it through data provider one, which returns work emails for some of them. Whatever comes back empty goes to provider two, then three, then four, each one filling gaps the last one missed. You pay per successful match rather than per lookup, so in theory you only pay for what you get.
The numbers used to sell it are real enough. One vendor guide reports that single-source lookups often plateau around 40 to 50% of a list while a properly built waterfall clears 80%, and that guide is published by a company selling waterfall enrichment, as most of them are. Hold the decimals loosely. The shape is right. One source does miss something like half a list in most markets.
The mechanism is fine. The ranking is wrong.
The claim underneath every enrichment pitch is that data coverage is near the top of the list of things limiting your results.
For a small sender it is usually near the bottom. You need quantity and quality, both, and waterfall enrichment buys you a small amount of one specific thing: more email addresses for people who are already on a list you already built. It cannot make those people the right people, and it cannot create companies that were never in your market to begin with.
Which means the pitch lands hardest on the companies least able to use it. If you have not yet settled who the buyer is, more coverage on an unsettled list just gets your message to a larger number of people who were not going to answer.
The arithmetic, with two markets
Take a company whose criteria match about 45,000 companies, one named buyer at each. They want to contact 20,000 of those people this year. One provider at 50% coverage returns roughly 22,500 usable addresses. They are finished. Every euro spent on a waterfall this year buys addresses they will not send to until next year, and they pay per record for the privilege.
Now take a company whose criteria match 1,200 companies, one named buyer at each, and who needs to reach about 1,000 of them this year to hit their number. One provider at 50% returns 600 usable addresses. That is almost five months of the plan missing.
Going from 600 to something near 960 is not a marginal gain for them, it is the difference between covering the year and running dry in the summer, and per-record pricing on 360 contacts is trivial against what those contacts are worth.
Same product. Opposite verdicts. The variable is the size of the matched market relative to the annual contact requirement, and almost nobody calculates it before buying.
So do that first. Count how many companies actually match your criteria, multiply by the contacts you need per company, and compare it against your annual target. That count is part of building the list in the first place, and it answers the enrichment question as a side effect.
The small-list argument, read properly
There is one argument on the other side that deserves a careful reading rather than a dismissal.
You will meet it as a statistic: campaigns sent to fifty people report response rates several times higher than campaigns sent to thousands. I went looking for the data behind the version that circulates and could not find it published anywhere I would be willing to link to, so treat the exact figures as folklore.
The pattern itself is familiar enough that anyone who has run both sizes will recognise it. Read literally, it says small lists win, which would argue for tiny hand-built lists with every field enriched to perfection.
I read it as a selection effect. Campaigns of fifty people are almost always run by someone who chose those fifty personally and wrote to each of them. The response rate is measuring the choosing and the writing, not the smallness. Shrinking your market to fifty people does not import the care that made those campaigns work.
The useful version of the finding: precision in who you pick beats size at the same level of effort. And separately, more volume inside a precise pick is more chances to win. Both are true at once, which is why the answer is a tight definition of the buyer and then a lot of sending inside it, rather than a choice between the two.
Where the money is better spent
Verification rather than enrichment, for most senders. An address that exists and accepts mail protects the sending infrastructure you spent a month building; an address that was guessed by a fourth provider and never checked can undo it. Auditing the list you have usually finds more upside than buying coverage on the list you do not.
After that, the thing enrichment is most often bought to avoid: deciding who the buyer is with enough precision that a stranger could apply your criteria and produce the same list. That decision is unpleasant because it means writing off segments. A data purchase feels like progress and postpones it.
And a reminder from the sending side, because it is the part that makes all of this moot. Sending 5,000 carefully personalised messages a week still gets you ignored when the foundations behind the message are broken. Coverage does not touch that. Neither does any amount of software in the background.
When a client's list reaches us, it is almost never short of email addresses. It is short of the right companies.
The rule, in one line
Count your matched market. If it is bigger than your annual contact requirement, spend on verification and on volume. If it is smaller, coverage is your constraint and a waterfall is a reasonable purchase. Run the count before anybody shows you a demo, because the demo is designed to be persuasive and the count is not.