If you run outbound inside Claude — building prospect lists, drafting sequences, cleaning up CSVs — you've probably hit the wall where you need verified contact data and Claude can't pull it on its own. FullEnrich's MCP integration closes that gap. You stay in the chat, pull real waterfall-enriched emails and phones, and only pay for what actually resolves.
This is a hands-on walkthrough, tested on 2026-06-03. I'll show you what the connection looks like, what a live search actually returns, where the credits go (and don't), and the gotchas worth knowing before you scale it.
What is FullEnrich MCP?
FullEnrich is a waterfall enrichment provider: it finds verified emails and phone numbers by routing each lookup through 20+ data providers in sequence, stopping when it gets a verified hit. That cascade is how it reaches ~80% hit rates where single-provider tools like Lusha or Hunter land closer to 60–70%.
The MCP server (https://mcp.fullenrich.com/mcp) exposes that enrichment engine directly inside Claude. Once connected, you can:
- Search for people by title, company, location, or industry in plain English
- Preview the first 10 results of any search for free — no credits
- Enrich individuals or bulk CSVs (up to 100 contacts per job) for verified emails and phones
- Check your credit balance without leaving the chat
- Poll the status of bulk enrichment jobs
- Run reverse lookups (email → person, domain → company data)
Think of it as a researcher with a FullEnrich seat sitting inside your Claude session.
How to connect FullEnrich to Claude (step by step)
This works in Claude.ai (Pro/Team/Enterprise) and in Claude Code.
In Claude.ai:
- Open Settings → Connectors.
- Click Add custom connector.
- Paste the server URL:
https://mcp.fullenrich.com/mcp - A browser window opens — sign in with your FullEnrich account (or create one; you get 50 free credits, no card required).
- Authorize the connection. Claude confirms the connector is live.
The whole thing takes under two minutes.
In Claude Code, the cleanest way is the CLI — it registers the remote server with the correct HTTP transport:
claude mcp add --transport http fullenrich https://mcp.fullenrich.com/mcp
Or add it to your .mcp.json manually (note the "type": "http" — a bare URL entry won't connect to a remote server):
{
"mcpServers": {
"fullenrich": {
"type": "http",
"url": "https://mcp.fullenrich.com/mcp"
}
}
}
Restart Claude Code, run /mcp to complete the browser sign-in, and the FullEnrich tools appear in your connector list.
A real worked example: finding VP Sales prospects in France
Here's exactly what happened on 2026-06-03 when I ran a live search through the MCP.
The prompt:
Search for VP of Sales contacts in France
What FullEnrich returned:
- Total matches: 1,314 contacts matching "VP of Sales" in France
- Free preview: the first 10 results, shown immediately, for zero credits
- Fields per contact (search stage): full name, location, current employment (company + title), LinkedIn profile, education, languages, skills
- Credit balance after the search: unchanged at 880 credits
That last point is the one to internalize. The search-and-preview step is completely free. You get a real-time market map — "how many VP Sales are in France, and who are the first 10?" — before spending a single credit.
What you do next is enrich the specific contacts you want verified data for:
Enrich the first 5 contacts with verified emails
At that point FullEnrich runs its waterfall — checking provider after provider until it gets a verified hit. Each successful email costs 1 credit; each phone costs 10. If it can't find a verified result for a contact, you pay nothing for that one. That's the pay-per-success model: credits only burn on confirmed hits.
The advanced play: mapping a whole buying committee in one prompt
The search above is the basic use case. This is the one that changed how I work.
Account research used to eat half my day. Pull the company, work out what they actually sell, figure out who owns the decision, check LinkedIn for the org structure, guess at reporting lines, decide who to approach first, then write something that doesn't sound like every other cold email that week.
I ran that entire job as a single prompt. It finished in about three minutes.
The prompt (swap in your own company):
Act as a group of the best market analysts. Answer this: what is the
ICP of [my company / website]? Then find one prospect that is a great fit.
Using the FullEnrich MCP:
- Map the org and the relevant people to contact, and draw it as a diagram
- Tell me who to contact to sell my product and why, and score each
person's relevance
- Write a first cold LinkedIn message and email
- Find the contact details for the leads
Output everything in the chat as a matrix and a diagram.
What came back:
It reverse-engineered my ICP from my own site, picked a real company that fit, and split the buying committee into three roles with a relevance score on each:
| Role | Score | Why |
|---|---|---|
| Contact first | 10/10 | Owns the specific budget line being sold into |
| Sign-off | 8/10 | Controls the budget, but the wrong first message |
| Champion | 8/10 | High influence, founder-led decisions, wrong entry point |
Then it drew the relationships as a diagram, wrote a first cold LinkedIn message and email, and pulled verified contact details for each person.
The part that genuinely impressed me: the obvious move is to contact the founder, because founders are visible and it feels like the shortcut. It scored the founder as a champion, not a first contact, and pointed at the person who actually owns that budget line. That's the call an experienced AE makes and a junior one gets wrong.
What I checked afterwards: the ICP it generated was genuinely accurate, not generic language that could describe any company. The org structure was exact when I verified it against LinkedIn.
That was one run. I'd still verify both every time, because a wrong org map is worse than no org map: you contact the wrong person and never learn why the deal died.
What the whole thing cost: the research, the ICP work, the org mapping and the scoring are all free. Only the final contact details consume credits. Three verified emails is 3 credits. Adding phone numbers for all three takes it to 33.
Which points at the workflow that actually makes sense: research for free, decide who matters, enrich only those people, and pull phone numbers only for the ones you genuinely intend to call. Most people do it backwards, enriching two hundred contacts and then working out who matters, which is how you burn a month of credits in an afternoon.
What it actually costs: the credit math
| Action | Credits |
|---|---|
| Search + preview (up to 10 results) | 0 |
| Check credit balance | 0 |
| Verified email (per hit) | 1 |
| Verified phone (per hit) | 10 |
| No match found | 0 |
Plans:
| Plan | Price | Credits/mo | What that buys | Rollover |
|---|---|---|---|---|
| Start | $29/mo | 500 | 500 emails or 50 phones | 3 months |
| Pro (most popular) | $55/mo | 1,000 | 1,000 emails or 100 phones | 3 months |
| Scale | from $500/mo | Custom | Custom emails & phones (+ SSO) | 3 months |
Every plan includes waterfall enrichment across 20+ sources, bulk CSV, the API plus Zapier/Make/n8n/Clay integrations, triple email verification, reverse email lookup, and unlimited free seats. There's also a 50-credit free trial with no card. Annual billing extends rollover to a year. Pricing verified 2026-06-04 — check the vendor pricing page for current numbers.
For a list of 100 contacts where FullEnrich hits its ~80% average, you're looking at ~80 verified emails = ~80 credits. On Start ($29 for 500 credits), that's about $0.058 per verified email — roughly $4.60 per 100-contact batch. Phones cost 10×: 80 verified phones would be 800 credits, past a full month on Start. If you cold-call, budget for Pro or Scale and test your phone hit rate on a small batch first.
For the full pricing breakdown versus Clay, Apollo, and Lusha, see our full FullEnrich review.
What you can actually do with FullEnrich in Claude
The workflows that hold up once the connector is live:
- Build a targeted prospect list on the fly. "Find 20 heads of growth at B2B SaaS companies in the UK with 50–200 employees." FullEnrich searches, previews 10 free, and you pick which to enrich.
- Enrich a CSV without leaving Claude. Reference a CSV of names + companies (or LinkedIn URLs): "Enrich these contacts with verified emails and phones." It queues up to 100 per job and returns a results table.
- Check your balance before a big run. "What's my FullEnrich credit balance?" — returns the number instantly, so you don't hit zero mid-job.
- Reverse-lookup an email. Got an inbound email but don't know who it belongs to? FullEnrich maps email → person → company.
- Map an org before a call. Domain → key people and roles, so you walk into an account call already briefed.
Honest gotchas
Search is not enrichment. Worth repeating, because it trips people up: previewing 10 contacts is free; getting verified emails costs credits. Two separate steps. A preview contact does not have a verified email until you enrich it.
The LinkedIn Chrome extension is gone — FullEnrich discontinued it in June 2024. If your old workflow was one-click LinkedIn enrichment, you now do it through the MCP or a CSV upload. No loss if you're already working in Claude.
No native Salesforce integration as of mid-2026. HubSpot syncs natively; Salesforce users export → enrich → import (or bridge via Zapier/Make).
Bulk caps at 100 contacts per job. For 1,000+ you run sequential jobs. The MCP can poll job status so you can automate the check-ins, but the per-job cap is real.
No ICP scoring or data normalization. FullEnrich returns raw contact data. Cleaning job titles, normalizing company names, or scoring against your ICP is a separate step — a custom Claude prompt, or a tool like Clay layered on top. Teams with complex ICP logic may still prefer Clay despite the higher price.
Is FullEnrich MCP worth it?
Best for: SDRs, AEs, and founders running outbound inside Claude who want verified contact data without tab-switching. The pay-per-hit credit model works in your favor versus flat-fee tools if your lists are messy.
Not for: teams that need native Salesforce sync, built-in ICP scoring, or one-click LinkedIn enrichment. High-volume teams enriching 10,000+ contacts a month should price out Scale against Clay's flat structure.
If you already live in Claude for outbound, the two-minute setup and 50 free trial credits make it a low-risk thing to test on your next list.
