LinkedIn Outreach on Autopilot

Scrape, qualify, personalize, send — without crossing LinkedIn’s 2025 tripwires.

The flow
Apify logo
Source
Apify
Claude logo
Process
Claude
Make logo
Destination
Make

The stack in the order it runs — data flows from the source through to where it lands.

Why this stack

Apify’s LinkedIn actors are the only ones that haven’t been killed by this year’s session-checks, and the per-run pricing makes "I’ll scrape 500 prospects" a $3 decision instead of a quote-request.

Claude personalizes per-prospect at roughly a fifth of GPT-4 turbo’s cost while keeping the message in voice, not generic-sales-pitch.

Make queues the sends with realistic delay variance (3 to 41 minutes), which is the difference between a real outreach and a flagged account.

The stack (3)

  1. Apify logo

    Marketplace of pre-built web-scraping actors.

    The LinkedIn actors here are the only ones that haven’t been killed by 2025’s session-checks. Predictable per-run pricing.

  2. Claude logo

    Long-context reasoning model from Anthropic — my daily driver for nuanced writing and orchestration.

    Better tone-matching and longer working memory than GPT for the tasks I care about most: guest messages, drafts, code review.

  3. Make logo

    No-code automation builder. Visual scenarios that chain APIs and AI calls.

    Per-operation pricing is cheaper than Zapier at the volumes I run, and the visual editor handles branching cleanly.

How it runs

  1. 1

    Pull prospects with Apify

    Run a Sales Navigator URL through the LinkedIn actor. Persist the JSON output to Airtable.

    uses Apify
  2. 2

    Qualify + personalize with Claude

    Per prospect: pass headline + recent posts + your offer. Claude scores fit 0–1 and writes a connection note in your voice.

    uses Claude
  3. 3

    Drip with Make

    Make picks rows where score > 0.7, sends with random 3–41 min delays, marks "sent" in Airtable. No more than 25 per day per account.

    uses Make

Want me to build this for you instead?

Product Audit and CTO Mode run out of this same thinking. If you’re reading this thinking “I want this, but in my product” — let’s talk.

See services