How Much of Your Expert’s Time Is Going to Repeat Questions?
____________________
Capture reusable knowledge. Protect expert capacity. Drive business value
By Isaac Bonney
AI-Ready Business, Knowledge Systems
Your best experts should be needed.
But they should be needed for the work that actually requires their experience and judgment, not to answer the same questions over and over.
Consider a simple example.
Assume an illustrative fully burdened senior-engineer cost of $150 an hour. Five repeat questions a week, at 20 minutes each, consume about 87 hours a year.
That is roughly $13,000 in senior labor spent repeatedly accessing knowledge the company has already paid to develop.
And that only counts the expert.
It does not include the employee waiting for an answer, lost billable capacity, context switching, project delays, or rework.
The opportunity is not to make experts unnecessary.
It is to stop spending expert time on work that does not require expert judgment.
What We Tested
That is what CAPTURE Experiment 01 was designed to explore.
We used one structured expert interview and asked AI to turn the responses into candidate organizational knowledge.
AI produced 20 candidate knowledge items.
Eighteen survived initial pressure testing, but because I was both the SME and reviewer, that result is directional rather than an independent performance measure.
Some of the polished output changed the expert’s meaning, added ideas that had not been provided, or removed important nuance.
This was a first test, so the findings should be treated as directional.
The Bottom Line
Capture Is Only the First Gate
Experiment 01 got us through extraction and initial challenge.
What comes next is the validation process we are now testing.
The next step is formal SME validation.
Higher-risk knowledge may also need qualified peer review.
Then comes another important question:
Can someone else actually use the knowledge correctly without going back to the original expert?
That is where knowledge capture begins turning into knowledge transfer.
This Is Also an AI-Readiness Problem
If AI is going to answer questions using company knowledge, the company first needs a way to decide which knowledge is accurate, current, approved, and safe to reuse.
Otherwise, AI can scale the wrong thing just as easily as the right thing.
That is why CAPTURE is not simply about documenting what experienced employees know.
It is part of building a knowledge base that both people and AI can rely on.
The Business Still Has to Benefit
Even accurate, validated knowledge has to improve something.
Does it:
- reduce repeat questions to senior staff?
- lower dependence on one person?
- improve consistency?
- reduce rework?
- speed onboarding?
- free experienced employees for higher-value work?
If none of those things change, the company may have created better documentation without creating much business value.
Don’t Create a New Bottleneck
There is also a cost to the CAPTURE process itself.
A company that has not done structured knowledge capture before has to learn what to capture, how to validate it, who should review it, and how to keep it current.
The first few cycles may take more effort.
Leadership should not have to review everything.
Its job is to decide which knowledge carries enough risk, cost, or business importance to require stronger validation.
Routine, lower-risk knowledge should move through a lighter process.
High-risk, technical, regulated, or high-impact knowledge should get stronger review.
Otherwise, the company may reduce one bottleneck only to create another.
The Standard CAPTURE Has to Meet
The goal is not to build a large knowledge-management bureaucracy.
It is to identify the knowledge that creates the most cost, risk, or dependency, then make that expertise easier for the organization to reuse.
The standard is straightforward:
CAPTURE creates business value when the benefit of reused expertise exceeds the cost of capturing, validating, using, and maintaining it.
Where is your business still using expensive expert time to answer questions that should already be easier to reuse?
Take the EverGX AI-Ready Scorecard:
https://evergx.com/ai-ready-scorecard/
Use it to see where knowledge, systems, and AI readiness may be creating friction inside your business.
Leave a Reply