Waterfall enrichment tries multiple data providers in sequence until it finds a result. Instead of relying on a single provider (which may not have data for every contact), the waterfall automatically moves to the next provider when one comes back empty. This gets you better overall coverage.
How it works
Say you need email addresses for a list of leads. With a single provider, you might get results for 60-70% of your list. With waterfall, you chain multiple providers together, and each one fills in gaps the previous ones missed.
Here's a real example of waterfall enrichment finding emails for 4 leads:
This table scrolls sideways on small screens.
| Lead | Provider A | Provider B | Provider C | Result |
|---|---|---|---|---|
| Jane Smith | Found | — | — | From Provider A |
| Mark Johnson | Not found | Found | — | From Provider B |
| Sarah Lee | Not found | Not found | Found | From Provider C |
| Tom Brown | Not found | Not found | Not found | No result |
- Jane was found by the first provider, so the waterfall stopped immediately.
- Mark wasn't found by Provider A, so Provider B was tried and succeeded.
- Sarah needed all three providers before getting a result.
- Tom wasn't found by any provider. That happens, but you maximized your chances.
Without waterfall, only Jane would have been found using Provider A alone.
Configuration options
When you set up a waterfall column, you can control two settings:
Stop on first success —When enabled (the default), the waterfall stops as soon as any provider returns a result. This saves credits by not querying providers you don't need.
Max providers to try —Sets a cap on how many providers are attempted per row. If you have 5 providers configured but set the max to 3, only the first 3 will be tried even if they all come back empty. This helps control credit spend.
Credit efficiency
Waterfall is more credit-efficient than you might expect:
- You only pay for providers that are actually queried. If Provider A finds the result, Providers B and C are never called.
- Most results are found within the first 1-2 providers, so your effective cost per result is often much lower than querying every provider for every row.
- Compare this to running each provider separately across your entire list, where you'd pay for every query regardless of whether you already have the data.
Ordering your providers
The order of providers in your waterfall matters. Here's how to think about it:
- Put your most accurate provider first. You want the best-quality data as your first result whenever possible.
- If two providers have similar accuracy, put the cheaper one first to save credits.
- Providers with broader but less precise coverage work well as fallbacks further down the chain.
- Faster providers near the top mean quicker results for the leads they can handle.
There's no single "correct" order. It depends on your data type, target audience, and budget. Experiment to find what gives you the best balance of quality, coverage, and cost.
Testing your waterfall
Before running waterfall enrichment on your full dataset:
- Run on 10-20 rows first to see how each provider performs.
- Look at which provider found each result and assess quality.
- Check how many credits were used and extrapolate to your full list.
- If a later provider is consistently outperforming an earlier one, move it up.
- Once you're confident in the setup, run it on your full dataset.
When to use waterfall
- Email finding — Different providers have different databases. Waterfall gives you a better chance of finding a valid email.
- Phone number lookup — Coverage varies by region and industry. Multiple providers fill the gaps.
- Company data enrichment — No single provider covers every company. Waterfall gives you broader reach.
When not to use waterfall
- When a single provider consistently gives you 90%+ coverage for your specific use case.
- When you need data from a very niche source that only one provider covers.
- When your credit budget is tight and you'd rather have fewer but guaranteed-quality results.