Tool
What will a Clay enrichment waterfall cost per matched row?
A waterfall runs data providers in order, sending each row to the next provider only when the one before found nothing. Order matters because every step is charged on the rows it sees, so cheap, high-hit providers go first and expensive ones last. Enter your providers, costs and hit rates to get the expected cost per matched row.
Estimate your waterfall
Add the steps in the order Clay runs them. The suggestions are the providers in our atlas that Clay connects to; type any other provider or action name.
Enter the row count and at least one step's cost per row and hit rate.
Credit prices come from your own Clay plan and each provider's terms inside Clay; hit rates from your own runs. Nothing here is pre-filled from Clay, and nothing is sent or stored.
How does the estimator work?
Step 1 runs on every row. Step 2 runs only on the rows step 1 missed, and so on down the list. Expected matches at a step are the rows reaching it times its hit rate. Expected cost is the rows charged times the cost per row. Cost per matched row is total cost divided by total matches. The table shows each step's arithmetic so you can check it.
Whether a step is charged on every row it runs on or only on the rows it matches is set by Clay and the provider, not by this page: check the action in Clay and set the toggle to match. The arithmetic also treats each provider's hit rate as independent of the others. In practice the rows every earlier provider missed are harder than average, so later steps usually find less than their standalone rate. Treat the matched total as a ceiling and test the order on a sample before you run the full table.
Where do the numbers come from?
From you. Credit prices per action come from your own Clay plan and each provider's terms inside Clay. They differ by plan and change over time, so this page pre-fills none and makes no claim about what any action costs. Hit rates come from your own past runs; a test on 100 rows gives a usable figure. Clay's docs say data credit costs vary by data type and that waterfalls with multiple providers change how many credits you use, which is the reason to work the order out before the run.
A worked example with made-up numbers
Example numbers, not Clay's prices: 1,000 rows through three providers, with the hit rates and credits per row below. The same three providers in two orders find the same 880 rows, because matches do not depend on order, but cost 4,200 credits one way and 7,400 the other. The expensive step runs on 1,000 rows when it goes first and on 200 when it goes last.
| Step | Rows reaching | Hit rate | Matched | Credits per row | Credits |
|---|---|---|---|---|---|
| 1. Provider A | 1,000 | 60% | 600 | 2 | 2,000 |
| 2. Provider B | 400 | 50% | 200 | 3 | 1,200 |
| 3. Provider C | 200 | 40% | 80 | 5 | 1,000 |
| Total | 1,000 | 88% | 880 | 4.77 per match | 4,200 |
| Step | Rows reaching | Hit rate | Matched | Credits per row | Credits |
|---|---|---|---|---|---|
| 1. Provider C | 1,000 | 40% | 400 | 5 | 5,000 |
| 2. Provider B | 600 | 50% | 300 | 3 | 1,800 |
| 3. Provider A | 300 | 60% | 180 | 2 | 600 |
| Total | 1,000 | 88% | 880 | 8.41 per match | 7,400 |
Sources
- Clay: compare plans, features and costs · the current plans; work out your own $ per credit from the plan you pay for
- Clay Docs: Actions and Data Credits · how actions and data credits are counted; costs vary by data type
- Clay: waterfall enrichment · Clay's own description of running providers in sequence
- Data providers, in Clay and not (this site) · where the suggested provider names come from