← All case studies
Live implementationSales Knowledge Assistant

Sales reps get cited technical answers in seconds instead of chasing other departments.

Industrial products company · Live · name withheld

99%answer accuracyReported
Hundredsof products and 30+ services coveredReported
Secondsto a cited answerTargeted
01
The problem

An industrial products company sells hundreds of products and more than 30 services, backed by terabytes of manuals, datasheets and service documents.

£What it was costing them

Slower replies to customers, slower deals, and expert time spent answering the same questions.

02
The existing workflow

How the work was done, and where it broke.

It starts with

A sales rep types a customer's question, in plain language.

What goes in

Product manuals, datasheets, technical files, service documents and approved business knowledge.

Done by hand today

Emailing other departments, searching shared drives and waiting for an expert to reply.

✕

For each customer question, reps emailed different departments and waited.

✕

Answers were buried in terabytes of manuals and technical files.

✕

Every slow reply slowed the deal behind it.

03
The solution

Grounded, not guessing. If the documents don't hold the answer, it says so.

1Brought the documents together

Indexed manuals, datasheets and service documents in one searchable place.

2Grounded every answer

The assistant answers only from those documents and shows the source.

3Kept experts involved

Reps and specialists correct or approve answers as they go.

4Improved from feedback

Weak answers are flagged and the next answer gets better.

☺
Where a person decides

Reps and subject experts correct or approve answers. That feedback improves the next answer. Sources are always shown so the rep can check.

04
Architecture

From input to outcome.

1

Inputs

  • Product manuals and datasheets
  • Technical and service documents
  • Approved business knowledge
2

Ingestion

  • Document parsing and chunking
  • Embeddings generated per passage
3

Retrieval

  • Vector search index
  • Most relevant passages selected per question
4

AI layer

  • Large language model writes the answer from retrieved passages only
  • Model chosen per deployment
5

Human review

  • Reps and experts rate and correct answers
  • Feedback loop into the knowledge base
6

Outcome

  • Cited answer for the customer
  • Faster lead qualification
  • Usage and feedback logs
05
Integrations

Works with the systems already in place.

⇄

Document stores and shared drives (export or connector)

⇄

Optional: CRM, to log questions against opportunities

06
Business value

What changed for the business.

99% answer accuracyReported

Across technical and support questions, from manuals to service documents.

No more chasingReported

Reps ask once instead of emailing different departments.

Leads qualified soonerReported

The right answer at the right time helps the team score leads.

07
The evidence

Shown on real work.

Hundreds of products, 30+ services and terabytes of manuals. Reps used to email different departments for each answer. Now they ask once and get a cited answer.
About the evidence

The method behind 99% (question set, grading, date) is to be confirmed before external use.

How we label results

Have a process like this?

Tell us about it. The first discovery call is free.

Start a similar projectAll case studies