Waterfall Enrichment: Combine Data Providers Well

No single B2B data provider has complete, accurate information on every company and contact. One may be strong on company firmographics in one region, another on direct phone numbers, another on technology data. Waterfall enrichment combines several providers in a defined order: ask the first source for a field, and if it has no answer or a low-confidence answer, move to the next. Done well, it raises coverage and accuracy while controlling cost. Done poorly, it multiplies spend, mixes conflicting data and creates compliance questions. This guide explains how waterfall enrichment works, how to order providers, verify results, control cost and keep records of where each value came from.

Why one data provider is rarely enough

Providers build their data in different ways and cover different markets well. Common gaps include:

• Missing records for smaller companies or certain regions
• Outdated job titles and contact details after people change roles
• Fields some providers do not collect, such as technology use or direct dials
• Inconsistent industry and size classifications

Relying on one source means accepting its gaps. Querying every source for every record wastes money. A waterfall sits between these extremes.

How waterfall enrichment works

A waterfall defines, for each field, an ordered list of sources:

1. Check whether the field already has a trusted value in your CRM
2. Query the first-choice provider for that field
3. If the result is empty or below a confidence threshold, query the next provider
4. Stop when a value meets your quality rules or the list ends
5. Verify the value where possible, such as checking email deliverability
6. Store the value with its source, date and confidence

The order can differ by field and by segment, because the best provider for European company data may not be best for North American phone numbers.

Decide which fields to enrich

Enrich only fields that support real decisions:

• Company fields for segmentation and scoring, such as industry, size and location
• Contact fields needed for outreach, such as role, seniority and business email
• Signals that help prioritise, such as technology use or hiring activity

Your ideal customer profile should drive this list. If a field does not affect targeting, routing or messaging, it probably is not worth paying for.

Order providers by evidence

Choose provider order using tests, not marketing claims:

• Take a sample of records where you already know the correct answers
• Run the sample through each provider
• Measure match rate, accuracy and freshness for each field
• Compare cost per successful match

The provider with the highest accuracy for a field usually goes first, unless a slightly less accurate provider is much cheaper and accuracy differences are small. Repeat tests periodically, since provider quality changes.

Worked example: a waterfall enrichment design

The table below shows a hypothetical waterfall for three fields. Providers are named generically, and match rates and costs are illustrative results from a sample test, not real figures for any provider.

FieldOrderSourceSample match rateSample accuracyCost per lookupRule to stopCompany size1Existing CRM value under 12 months old40%TrustedNoneStop if presentCompany size2Provider A70% of remaining90%LowStop if confidence highCompany size3Provider B50% of remaining85%LowStop if foundBusiness email1Provider B55%88%MediumStop only if verification passesBusiness email2Provider C40% of remaining85%MediumStop only if verification passesDirect phone1Provider C30%80%HighOnly for priority accounts

Notice that the expensive phone lookup is restricted to priority accounts, and emails are accepted only after verification.

Verify before you trust

Enriched values can be wrong. Add verification where possible:

• Check business emails for deliverability before use
• Compare company domains with the company name and website
• Flag job titles that conflict between sources
• Prefer values confirmed by more than one source for important fields

When providers disagree, apply clear rules, such as preferring the most recent value or the source with higher tested accuracy for that field, and send important conflicts for review.

Control cost

Waterfalls can become expensive if not managed:

• Skip enrichment when a trusted, recent value already exists
• Enrich only records that match your target segments
• Limit expensive fields to priority accounts
• Set monthly budgets per provider and alert when close
• Cache results so the same record is not looked up repeatedly

Track cost per enriched record and per field to see where money goes.

Record where every value came from

Store source, date and confidence alongside each enriched value. This lets you:

• Re-check old values when they pass a freshness limit
• Remove or correct data from a provider found to be inaccurate
• Answer questions about where personal data came from
• Measure provider performance over time

Our guide to CRM data enrichment covers how to structure these fields and keep enriched data clean.

Respect data protection rules

Contact enrichment involves personal data. Depending on where you and the people concerned are located, laws such as the GDPR may require a lawful basis, data minimisation and transparency. Under the GDPR, where personal data is not obtained from the person directly, there are obligations to provide information to them (GDPR Article 14). Check how each provider sources its data and what contract terms they offer, and get legal advice for your situation. This article is general information, not legal advice.

Refresh data over time

B2B data decays as people change jobs and companies change. Set freshness limits per field, such as re-checking job titles and emails more often than industry, and re-run the waterfall only for records that are stale and still relevant.

Measure whether enrichment pays off

Enrichment is only worthwhile if it improves results. Compare outcomes for enriched and unenriched records, such as routing accuracy, reply rates or conversion from lead to opportunity, and review cost per useful record each quarter. Drop fields and providers that add cost without improving decisions.

Integrate with your CRM carefully

Run enrichment through a controlled process: write to dedicated fields, never overwrite values entered by sales staff without rules, log every update and respect API limits. Test on a sample before enriching the whole database.

Frequently Asked Questions

What is waterfall enrichment?

It is querying several data providers in a defined order for each field, moving to the next only when the previous one has no answer or a low-confidence answer, then verifying and storing the result with its source.

How do I choose the order of providers?

Test providers on a sample of records with known answers, measure match rate, accuracy and cost per match for each field, and put the best performer first.

Does waterfall enrichment cost more than one provider?

It can, but skipping records with trusted values, focusing on target segments and limiting expensive fields to priority accounts keeps cost under control.

What should be stored with each enriched value?

The source, the date it was obtained and a confidence or verification status, so data can be refreshed, corrected or explained later.

Combine data sources without the mess

We test providers, design waterfalls, add verification and source tracking, and connect enrichment safely to your CRM. See our B2B lead mining and prospect verification service, or talk to us about waterfall enrichment.

About the Author

Manav Sharma is the founder of Insyrge and a CRM solutions architect. He works across Zoho, Salesforce, HubSpot, Odoo and more, designing the CRM setups, integrations and automations that keep sales and operations teams working from the same data.