Contact Book a call

SaaS · Customer story

An AI outbound sales platform, from contact list to booked meeting.

An outbound sales platform that ingests contacts, writes personalised emails from a knowledge base, runs follow-up sequences, handles replies and bounces, and hands interested prospects to a person.

Get this for your team See the product
Client
IntellReach AI
Industry
SaaS
What we built
AI outbound sales platform
Built with
Supabase, Claude, OpenAI, Instantly
AI outbound sales platform for a intellreach ai: 11 product areas in one platform

The results

In short

IntellReach needed an outbound system that writes like a person and knows when to stop. SBD built a multi-tenant platform that ingests and verifies contacts, writes each email with Claude from a knowledge base, runs follow-up sequences, classifies replies, and routes interested prospects to a human.

The product

Real screens from what we built.

Before and after

What changed for the team.

BeforeAfter
Prospect research and first emails written by handContacts ingested, verified and enriched automatically
Follow-ups tracked in spreadsheetsMulti-step sequences that run on schedule
Replies sorted manually across inboxesReplies and bounces classified, interested prospects routed to a person
No shared knowledge behind the messagingEvery email drafted from a curated knowledge base

How we did it

How the ai outbound sales platform was built.

  1. 1Product design

    A product reference document and kickoff that fixed what the first phase would and would not do.

  2. 2Platform

    Next.js and Supabase with row-level security on every tenant row, so each workspace only sees its own data.

  3. 3Writing and retrieval

    Claude writes each email using retrieval over a knowledge base, with personas and proof points the client controls.

  4. 4Run it simply

    No separate worker or queue service. The queue is a table and three scheduled jobs do all the work.

The workflows

What runs now, and what sets it off.

  1. 01
    Contact ingest

    When a list is uploaded

    Then

    Addresses verified, contacts deduplicated and assigned to a campaign

  2. 02
    Email generation

    When a contact enters a sequence

    Then

    A personalised email written from the knowledge base and queued

  3. 03
    Reply handling

    When a reply or bounce arrives

    Then

    Classified, sequence paused, interested prospects routed to a person

  4. 04
    Readiness checks

    When a campaign is about to start

    Then

    Domains, inboxes and content checked, blockers listed before sending

Built with

The tools behind it.

Every account was created in the client's name.

What made it work

Keeping the machine simple. A table for a queue and three scheduled jobs made it cheap to run and easy to reason about.

What comes next

  • LinkedIn research before the first email
  • Do-not-contact lists shared across campaigns

For saas teams

Act on usage before the customer goes quiet.

The same pattern, built around your tools. Bring the one workflow your team still runs by hand and we will show you what this would look like for you.

Show us your funnel
A software team working at their desks in an office
SaaS Photo: cottonbro studio, Pexels.

Where this applies

The solutions this story proves.

How it starts

Every story on this page started with an audit.

One to two weeks. We map the work your team still does by hand, put an hours-a-week figure on each piece, and give you a costed blueprint in order of payback. The $500 is credited in full to the build.

Start with the audit See how we work
  • Week 1 to 2Audit with the people who do the work.
  • Week 3Blueprint: fixed scope, fixed price, first build chosen by payback.
  • Week 4 onTwo-week sprints, working software every second Friday.
  • After go-liveSixty days of hypercare. Accounts and code in your name.

More customer stories