Problem
Finding the right person takes several referrals
“Has anyone dealt with this before?” An employee posts the question in the company chat. Someone suggests asking finance. Finance suggests checking with the account manager. The account manager remembers a colleague who handled something similar, but cannot remember who.
Meanwhile, the work waits. Consider a fictional business where a new billing assistant, Maya, needs to submit an invoice through a customer’s online portal. The portal keeps rejecting the project code. The instructions explain the normal process, but do not cover this error.
Across the business, an operations coordinator named Luis recognizes the problem. He helped set up that customer’s portal when he worked in billing. Maya does not know Luis. His job title gives her no reason to ask him. The knowledge exists. Finding the person who holds it is the missing step.
Examples
Make experience visible beyond the job title
An organization chart tells Maya who works in finance and who manages the account. It does not tell her who understands this customer’s invoice requirements. A staff directory may give her names, email addresses, and departments. She still needs a useful connection between her question and someone’s experience. APQC describes expertise location as a knowledge-management approach that connects people with knowledgeable colleagues.
For a small business, that could start with a shared page answering a practical question: Who can help with which tasks? Luis might be listed under “customer invoice portals,” with a note about the setup he handled. Another colleague might know how to prepare a particular supplier’s order file. Someone else might understand the settings behind a recurring customer report.
Those details give a newer employee a starting point. They also make experience from previous roles visible across departments. Without that starting point, knowing whom to ask depends on whom the employee already knows. Someone who has been there for years may find Luis immediately. Maya may spend the morning collecting referrals.
Solutions
Describe the questions people can help answer
A broad label such as “finance expert” asks the employee to guess what that person knows. “Can help with rejected invoices in customer portals” is more useful to Maya.
APQC’s guidance on organizing knowledge recommends using terminology employees understand and organizing around business concepts rather than the organization chart.
Applied to a directory of people, that means describing recognizable tasks and questions. Ask the colleague to confirm the wording and its limits. A short entry could include:
Start with a short contact entry
- Can help with: the task or question.
- Relevant experience: what the person has worked on.
- Start here: a guide or example, if one exists.
- How to ask: their agreed contact route.
- Alternative contact: another person or group with relevant experience.
- Last confirmed: when the colleague checked the entry.
Give people a workable way to ask
Making Luis easier to find may bring him more questions. His entry needs to explain how that help fits into his work.
For the fictional business, he might agree to receive questions through a shared billing channel. Another knowledgeable colleague can then respond when Luis is unavailable, and the explanation can remain visible to the people allowed to use it.
A good contact route also helps the person asking. Maya should know what information to include and what to do if the question is urgent. Her message could be:
Give the colleague enough context
“The customer portal rejected this project code. I followed the submission guide and checked the code against the job record. Can you help me understand what the portal expects?”
That gives Luis something concrete to work with. Maya can share the relevant details through the approved channel without broadcasting the full invoice to the business. Link routine questions to existing instructions. Leave a clear route to a conversation when those instructions do not fit the situation.
If nearly every question reaches Luis, review why. The guide may be missing a common step. Another colleague may need to learn the task. The contact arrangement may need to change. Being listed should not create an unlimited obligation to answer immediately.
Keep the directory believable
A contact entry can become stale even while the person remains employed. Luis may move to another role, stop using the portal, or lose the time set aside for questions. A new colleague may gain more recent experience. APQC’s comparison of knowledge-management tools identifies accurate, useful profiles as a dependency for expertise-location tools.
Give someone responsibility for maintaining the shared directory. Ask listed colleagues to confirm their topics, contact preferences, and alternatives at an agreed interval, and when their responsibilities change. Make corrections easy. “This is no longer my area” should be a useful update.
Employees should also be able to flag a failed connection: the description looked relevant, but the person could not help. That feedback may reveal wording that is too broad or a topic with no available contact. A short, confirmed entry gives Maya a stronger basis for asking than an impressive profile nobody has checked.
Use automation to support the connection
Once the directory has useful entries, automation could make them easier to maintain and search. A workflow could send a profile-confirmation prompt, collect proposed edits for review, and show approved contacts alongside relevant guides.
An AI search tool could suggest Luis’s entry when Maya describes the project-code error. It should show the reason for the suggestion and link to the confirmed profile. When no suitable entry exists, it should make that gap visible.
Keep employees involved in confirming what they know and how they can help. A high message count or frequent mention of a topic is not enough to establish expertise.
Try the directory with one task
Start with one task where employees regularly ask around. Let someone unfamiliar with the process try to find help. Check whether they reached a relevant person, how many referrals they needed, and whether the contact arrangement worked for the colleague answering.
For Maya, success means she can find someone with relevant experience without first building a network across the business.
How Zoevin helps
Follow one question to the right colleague
At Zoevin, I build scoped automations in a business’s own environment and hand them over on an agreed date.
Bring one question that regularly travels through several colleagues. We can look at how to make the path to useful experience easier to follow.
- Trace how an employee finds someone with relevant experience.
- Identify directory preparation and review that automation could support.
Zoevin has no delivered automation projects yet. Discovery names the seam. If I build, I hand it over on a written date. You get the workflow, the logins, and the write-up. I don't stay on to run it. You buy the software.
Evidence
Sources
Primary reporting first. Open the sources yourself rather than taking this account alone.
primary source
APQC — Understanding Expertise LocationApril 30, 2026 public introduction defining expertise location as connecting people with knowledgeable colleagues. Maya, Luis, and the customer portal are fictional teaching examples, not client work or measured outcomes. No claims are drawn from the linked full article.
primary source
APQC — How should knowledge taxonomies and metadata be designed?September 2026 guidance on familiar user terminology and business concepts rather than organization charts. The task-based contact entry is an editorial application of that guidance, not a prescribed directory template.
primary source
APQC — What Are the Best Tools for Knowledge Management and How Do They Compare?August 2026 comparison identifying accurate, useful profiles as a dependency for expertise location. Contact arrangements, review prompts, and AI suggestions are illustrative proposals requiring human confirmation, not proven savings or autonomous assessments of expertise.