Consulting · 10 consultants

Consulting team: knowledge without searching through chats

Guides and past solutions are scattered across folders and chats, and newcomers interrupt experienced colleagues.

Situation

Ten consultants keep guides, templates and past solutions in several folders and chats. Newcomers interrupt experienced colleagues, and a document they find may be out of date or belong to another client.

What changes

Natural-language search over an agreed knowledge base that respects access rights. Answers include excerpts, links, the document date and a warning about contradictions.

Where people stay in control

When the sources are not enough, the assistant says there is no answer. Before launch the team names document owners and removes the main duplicates.

What changes for the team

The reasoning behind an answer comes with the source document — especially helpful when joining an unfamiliar project.

How the example is calculated

Time spent searching per month
Before
100 h/month
After
46 h/month

Preliminary estimate · month

10 people × 6 searches × 5 minutes × 20 days = 100 h; at 2 minutes — 40 h + 6 h of upkeep = 46 h

Assumptions

10 people; 6 searches a day; 5 minutes now and 2 minutes after; 20 working days; 6 hours of knowledge-base upkeep. Search frequency and time must be measured first.

What to check in a pilot

Answers linked to current documents, correct refusals, no access to restricted material, task success after the search.

Explore a similar situation

The map starts from the same kind of problem. None of this story’s volumes or results are assumed for your business.