How do I use NPI data to find target accounts?
The national provider registry is a free, comprehensive map of providers and locations. Here is how to prospect with it.
Use NPI data as the first layer of finding target accounts: filter the free NPPES file to Type 2 organization records in your taxonomies and states, group providers by practice location, then triangulate each candidate against the company website, reviews, social, and your own CRM to confirm it is real, active, and worth working. The National Plan and Provider Enumeration System (NPPES) is the most complete public map of the U.S. healthcare provider market, it is free, and CMS refreshes it weekly and monthly. It is one input among many, not the whole picture.
This guide covers finding and building the account list. For turning firmographics into a ranked score, see How do I build a Fit Score? For adding size and ownership, see How do I use CMS data to qualify accounts?
What is in the NPPES NPI file?
Every provider and provider organization in the United States has a National Provider Identifier (NPI), and the public NPPES download lists them one row per NPI. The fields that matter most for building a prospect list are these.
| Field | What it is | Use it to |
|---|---|---|
| NPI | The unique 10-digit provider ID | The stable key you match and dedupe on |
| Entity Type Code | 1 = individual provider, 2 = organization | Filter to Type 2 when you sell to practices, not individuals |
| Healthcare Provider Taxonomy Code | The specialty classification (repeats up to 15 times, with a Primary Taxonomy Flag) | Filter to the specialties you sell into |
| Provider Business Practice Location Address | The physical practice address | Group providers into a single location and set geography |
| Authorized Official | The named official on the organization record | A starting contact at the organization |
In short: keep Type 2 records in the taxonomies and states you cover, group the providers who share a practice address so a ten-provider group reads as one account rather than ten leads, then match against your CRM to separate existing accounts from net-new ones.
Illustrative example
Illustrative, based on how SignalRx customers use NPPES. A company selling scheduling software to primary care wants independent family-medicine groups in Texas. They filter NPPES to Entity Type 2, the Family Medicine taxonomy (for example 207Q00000X), and TX, then group providers by practice address. A cluster of six family-medicine NPIs at one address is one six-provider group, not six leads. Matched to the CRM, most are net-new, and the six-provider groups rank above the solo practices.
How do I go beyond the registry?
NPPES gives you a candidate list. It does not tell you whether the account is thriving, who leads it, or how big the opportunity is, so each candidate gets validated and enriched from sources that carry the signal the registry cannot.
- The company website confirms the account is active and shows its service lines, locations, and pricing or financing pages, real proof the practice operates and what it sells.
- Google reviews and the Google Business Profile, plus Instagram, are the primary signal where NPPES is thin, which is most aesthetics, wellness, and cash-pay medicine. Review volume and rating and an active Instagram say more about a med-spa than any registry row.
- LinkedIn gives headcount and leadership so you know the size of the group and who decides.
- Job postings flag growth, new locations, and new service lines before they show up anywhere else.
- Your own CRM and revenue data are the match-and-prioritize layer, telling you which candidates you already touch and which look like your best customers.
Caveat for consumer and cash-pay markets: when you sell into aesthetics, dental, wellness, med-spa, or any cash-pay vertical, lead with web, social, and reviews and treat NPPES as secondary. Many of these businesses are barely represented in the registry, so a website, an Instagram following, and a wall of Google reviews are better evidence the account exists and is worth working than an NPI ever will be.
How does SignalRx handle this?
SignalRx treats finding accounts as multi-source triangulation with NPPES as the first layer. It pulls NPPES in, groups providers to locations, resolves the owning group, and then corroborates and enriches each account from the company website, reviews and social, job postings, and CMS for size and ownership, matching everything to your CRM and writing a clean, deduplicated account onto the record. When a CRM account arrives missing firmographics, SignalRx fills them by triangulating these sources so a thin record becomes scoreable instead of being dropped. For net-new prospects the scoring runs on this combined public, web, social, and review picture, and inference decides which source to trust when they disagree, so the resolved account is the most reliable version. Fit Scores and tags are written onto both the account and its contacts and become Revenue Plays and lists native in Salesforce. Your own internal signals, revenue, support, and product usage tracked over time, join the picture once a prospect converts to a customer, and that longitudinal view is what competitors cannot copy.
Key takeaways
- NPPES is a free, complete, weekly-updated map of providers and locations, and the first layer of finding accounts, not the whole picture.
- Filter to Entity Type 2 and your taxonomies, group providers by practice address, then validate each candidate against the website, reviews, social, and your CRM.
- For consumer and cash-pay markets, lead with web, social, and reviews, not NPPES.
See it in action: Discover High-Fit Prospects with SignalRx.
Related: How do I build an ICP from public data? and How do I resolve parent-child ownership?
Want any of these running on your data?
Frequently Asked Questions
.svg)
.avif)