Skip to content
All customer stories

Defence and commercial construction contractor, Government and commercial construction

AI lead intake: every enquiry scored and routed

A defence and commercial construction contractor sorted LinkedIn replies, web form enquiries and emails by hand. The SalnetHub team built AI lead qualification, then a RAG knowledge base and AI agent in n8n that scores and routes every enquiry.

6
enquiry types sorted and routed automatically
100-pt
qualification score from the company's own rules
2
projects, the client came back for a second build
AI lead qualification: LinkedIn replies, a web intake form and email feed an AI agent that checks an AI knowledge base and routes six types of enquiry

The challenge

Replies from LinkedIn campaigns, the web intake form and emails all landed in one place. Someone had to read each one, decide whether it was a real buyer, a returning client or noise, and copy it into JobTread. Good leads waited, and every person sorted them a little differently.

What we built

The SalnetHub team first built AI lead qualification in Zapier for LinkedIn replies. The client came back for a second build: a RAG system in n8n: an AI knowledge base chatbot built from Google Drive, OpenAI embeddings and Pinecone, and an AI agent that scores and routes every enquiry, with leads sent to JobTread through Zapier.

“Bwire does a great job for our company and we have used him multiple times. Great hire! I suggest you do the same!”
CEO, Defence and commercial construction contractor (Upwork, 5.0 stars)
1

LinkedIn replies, qualified by AI

Project one: when a prospect replies to a LinkedIn campaign, a Zapier workflow has AI read the reply. Real buyers go straight into JobTread for outreach; the rest go to a nurture list.

  • We have a modular facility requirement next quarter. Can you quote?
  • Not right now, maybe next year.
  • Who handles your MEP and fit-out bids? Send capabilities.
AI

Reach out now

2 replies → JobTread

Nurture

1 reply → follow-up list

Example replies, written for this page.

2

A knowledge base the AI checks first

Project two: a RAG system in n8n. The company's rules live in Google Drive, are embedded with OpenAI and stored in Pinecone, so every decision follows the contractor's own scoring, not a guess.

Knowledge base

  1. Google Drive

    Company rules, services, scoring

  2. OpenAI embeddings

    Each file turned into vectors

  3. Pinecone

    Searched by the agent on every message

Add or change a file in Drive and the agent knows it on the next sync.

100-point qualification score

86
  • 80–100Advance to estimate
  • 60–79Founder review
  • 40–59Nurture
  • 0–39Disqualify

Points for budget, decision maker, timeline, fit, margin and capacity.

3

Every enquiry routed in seconds

Each form or email is read by an AI agent, matched against the knowledge base and sent where it belongs, with no one sorting the inbox by hand.

Web intake form

Email

AI agent in n8n

Checks the knowledge base, decides the type, returns clean JSON.

  • New lead→ JobTread pipeline
  • Job enquiry→ Sales task created
  • Meeting request→ Calendly event, CEO notified
  • Returning client→ Invoice generated
  • Service question→ Google Sheet for follow-up
  • Promotional→ Discarded
  • Leads and enquiries reach JobTread through Zapier.

Inside the build

The RAG system in n8n: one pipeline keeps the knowledge base in Pinecone up to date from Google Drive, the other uses it to score and route every new enquiry.
The RAG system in n8n: one pipeline keeps the knowledge base in Pinecone up to date from Google Drive, the other uses it to score and route every new enquiry.

Want a result like this?

A 30-minute call is enough. We tell you plainly what we would automate first, and what it would cost.

What happens on the call

  1. 1We look at how the work happens today
  2. 2We work out what the problem costs you
  3. 3You get a 14-day pilot plan. No obligation

Free, 30 minutes, on Google Meet.