best MCP servers for sales prospecting
A builder's comparison of MCP servers purpose-built for sales prospecting — Lusha, Explorium, Prospeo, and the broader field — evaluated on whether they…
✦ The account-based marketing API for AI agents ✦
Personalisation, at scale.
Marketing, when it was a craft. Now the data’s rich enough to do it again. At scale this time.
Search API
Find the right accounts and people
One Search verb, two modes. By keyword, find people, companies, jobs, and posts. By company and role, source the exact decision-makers at a target account. It streams over SSE, so your agent gets results page by page.
- By keyword — find people, companies, jobs, and posts across LinkedIn and the web
- By company & role — source the decision-makers at a target account
- Async-streaming — a job handle immediately, then results stream page-by-page over SSE
- Cited candidates — every match carries its source attribution and a confidence score
- One key, one schema — the same envelope your agent uses to Enrich and Create
POST /v2/search
// Request
{
"type": "people",
"keywords": "VP Engineering",
"filters": {
"company": "Datawise Analytics",
"location": "Austin, TX"
}
}
// 202 Accepted — a streaming job handle
{
"data": {
"id": "search_550e8400",
"status": "queued",
"stream_url": "/v2/search-jobs/search_550e8400/stream"
}
}
Outreach Angle
88%“Marcus recently posted about scaling RevOps for hypergrowth. Lead with how ABM.dev eliminates manual data entry—his team is drowning in spreadsheet imports from HubSpot.”
Person Summary
95%Revenue operations leader with 8+ years scaling GTM teams at high-growth SaaS companies. Previously built RevOps function at two unicorns. Known for data-driven approach and strong opinions on CRM hygiene.
Background Highlights
91%- • Built RevOps team from 0→12 at previous company
- • Reduced sales cycle 23% through data enrichment
- • Speaker at RevOps Summit 2025
Matched Persona
92%Technical Decision Maker
Title (VP RevOps) + department (Operations) + seniority (reports to CRO) align with technical buyer profile. High likelihood of evaluating tools independently.
Enrichment API
Ten sources. One envelope.
One API call. A structured envelope back — snapshot of what’s true, narrative of what it means, events as a timeline. Every value carries its citations, the source that won, the sources that lost, and the dates that matter.
- Ten data sources resolved per enrichment — LinkedIn, Companies House, Perplexity, Tavily, Hunter, and more
- Structured snapshot: identity, offices, size & growth, tech maturity, decision structure, competitive risk
- Every value cites its sources. When sources disagreed, the loser is preserved — you decide whether to override
- Events table: funding rounds, leadership changes, acquisitions — trigger off them, don't diff snapshots
- Research angles: save your own questions and have them answered on every matching enrichment
- Destinations: compose source, events, and angles into the exact shape your CRM property wants — one call, eight templates ready to go
One key. One envelope. One workflow.
Three verbs behind one API key: Search the right people and companies, Enrich them into cited, confidence-scored fields, and Create outbound on LinkedIn. One schema, one audit trail.
Search→Enrich→Create
Connects to
03 · For agents
What's in the box.
Enrichment, citations, streaming, CRM sync, batching, webhooks. The whole toolkit for the launch.
Multi-source enrichment
One call, ten providers behind it. LinkedIn, Hunter, Perplexity, Tavily and seven others. aggregated, deduped, reconciled. No per-source bills.
Eighty-nine canonical fields
Forty-three on the person, forty-six on the company. Same schema every call, same shape every response. No mapping headaches.
Citations on every value
Each field tells you where it came from and how confident we are. No fabricated facts. No silent fallbacks.
Real-time SSE streaming
Live progress updates as the enrichment runs. Built for autonomous agent loops, not human dashboard-watching.
CRM auto-syncComing Soon
Enriched records flow into your CRM. Map any of the eighty-nine fields to CRM properties. No CSV exports, no manual entry.
Configurable prompts
Tune how the AI fields synthesise data. Define personas, set outreach tone, name the signals that matter for your ICP.
Webhook notifications
Get a callback when enrichments complete. Trigger downstream agent workflows, update dashboards, or kick off sequences.
Start in 3 minutes
Three minutes, three steps.
Get a key. Make a call. Read the response. We'll see you at step four.
1
2
Make your first request
Give your agent the API key and the code example on the right. It handles the rest.
3
curl -X POST https://api.abm.dev/v1/enrichments \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"type": "person",
"input": { "email": "[email protected]" }
}'Simple pricing
Pay per enrichment, not per seat.
Credits cover every source we touch — LinkedIn, Hunter, Perplexity, Tavily, and the rest. From $0.30 per enrichment at the entry rate, less in volume. No subscription. No per-field charge. Credits never expire.
30credits
$2.99
$0.0997 / credit
100credits
$9.99
$0.0999 / credit
Best value
500credits
$39.99
$0.0800 / credit
2,000credits
$129.99
$0.0650 / credit
Custom
Talk to us
Bigger volume, dedicated rate, annual invoicing. Or self-serve any pack from your dashboard.
From $0.30 per enrichment.
Top up from $2.99. No subscription, no per-field charges. Credits never expire.
All data sources includedNo per-field chargesCredits never expire
From the blog
Notes from the data layer.
b2b data enrichment API comparison
Most enrichment API comparisons rank on coverage and price — the wrong metrics for AI-agent GTM. This post builds an evaluation framework for builders…
AI SDR hallucination
AI SDRs hallucinate in four distinct ways: wrong company data, fabricated contact details, invented intent signals, and stale firmographics — each with a…
✦ No subscription. Pay per enrichment. ✦
Run a real campaign. Then decide.
Top up from a few pounds. Same fields, same speed, same shape every call. Credits never expire. Run one enrichment to feel the shape, then run ten thousand.