Setting Up an Internal IT Help Desk

When IT questions pile up on one person and their real work stalls, the fastest way out is usually not hiring more IT staff. It is deciding on a single entry point for requests and cutting repeat questions with short guides. In a small company you can start by creating one shared channel or form and naming a primary and a backup person. On cost, handling it in-house with a part-time person needs almost no external spending, while outsourcing a help desk generally ranges from a few tens of thousands of yen per month to well over a hundred thousand yen with on-site visits or extended hours (a rough guide that varies by contract). This article covers the types of requests, options for a single entry point, ways to reduce inquiries, a comparison of in-house, outsourced, and hybrid models, a five-step setup, common failure patterns, and a checklist.

An internal IT help desk is the entry point where employees bring IT problems, together with the process for handling them. It does not require a dedicated department or a large system. What matters is that people know where to go, that requests are recorded, and that someone is responsible for replying. Start by sorting the requests your company actually receives.

The table shows that the more frequent a request is, the more room there is for self-service. Repeat requests such as passwords, printers, and software how-tos can be reduced with guides. Security-related reports, on the other hand, are rare but urgent, so you need to decide in advance who receives them first and who takes over. A good starting point for design is separating what can be reduced from what needs a quick decision.

In many small companies the help desk is never designed. People ask whoever is easiest to ask: the person who knows PCs, a general affairs staff member, or an outside contractor the president knows. This causes three problems. First, requests arrive in private chats or in person, so nothing is recorded and the same question is asked again and again. Second, when that person is absent nobody can answer and the requester waits. Third, the person who is asked most often keeps getting interrupted and eventually burns out.

The pattern is especially strong where only one person handles IT. How a solo IT person turns into a jack-of-all-trades, and how to sort the workload, is covered in organizing the workload of solo and part-time IT staff. Here we focus on the step before that: creating a single entry point for requests. Even without adding people, tidying the entry point greatly reduces missed requests and interruptions.

There are four main ways to unify the entry point. None is universally best. It is more realistic to choose according to your size, request volume, and the tools employees already use.

If unsure, using a dedicated channel for intake and a form for the record is an easy way to start small. For choosing a business chat tool and its costs, see business chat tool comparison and costs. Whichever method you pick, decide three things: business hours and a target time for first reply (for example, same business day), how to reach someone for urgent matters (such as by phone), and a backup person when the primary is away. Without these, even a unified entry point ends with phone calls to the same person.

A ticketing tool manages each request as a ticket and records the owner, status, history, and time spent. Common options for IT support include Jira Service Management, Zendesk, and Freshservice. Plans, pricing, and free tiers change, so we do not list amounts here. Check each vendor's official site for current pricing and features before deciding.

A common mistake is choosing by the number of features and then failing to operate it. Detailed categories, priorities, and approval flows become empty if nobody has time to enter and update them. Three statuses (received, in progress, done) are enough at first, and adding fields after operation stabilizes lasts longer. If you also want to link device and license ledgers, getting started with IT asset management helps you see the whole picture.

After setting up the entry point, the biggest gain usually comes from reducing the inquiries themselves. The key is not to document everything. Target only the most frequent requests and turn them into short guides with screenshots. The following cycle is light and easy to sustain.

- Record the type and number of requests for about a month (one spreadsheet column is enough)
- Pick the top five types (passwords, printers, Wi-Fi, software how-tos, and so on)
- For each, write one page covering symptom, what to check, steps, and what to do if it still fails
- Put them where anyone can see (shared drive, internal wiki, pinned chat message) and link them from the help desk notice
- Use the rule that a question asked three times becomes a guide
- Handle requests that the guides did not solve, and fix the gaps in the guides

A guide that is obvious to its author is often unclear to the reader. Having a beginner follow the steps once will reveal gaps and confusing wording. For how to write manuals in general and digitize paper or person-dependent manuals, see digitizing business manuals. For routine requests such as forgotten passwords, providing a way for people to reset them themselves removes the inquiry in the first place.

Who runs the desk falls into three models: in-house, outsourced (help desk outsourcing or external IT support), and a hybrid of both. The types of outsourcing services and what to check before signing are covered in how to use IT support outsourcing and advisory services, so here we compare from the viewpoint of running the desk.

The costs above are only general guides and vary greatly by contract: coverage hours, scope, headcount, term, and whether emergency support is included. The same monthly fee can differ in real value depending on whether it is remote only or includes visits, and whether nights and holidays are covered. When comparing, ask several providers under identical conditions about the target time for first reply, what is included in scope, the rates for out-of-scope work, and cancellation terms, not just the price.

As a rough guide: in-house suits a few to a dozen requests a month with a well-informed handler; hybrid suits a single handler whose absence is risky or a need for wider hours; outsourcing suits companies that cannot or prefer not to keep an IT person. Even with outsourcing you still need an internal contact. Notifying hires and leavers, approving permissions, and deciding on costly work can only be done by the company.

Setting up a help desk does not need to be a big project. A realistic approach is to start small in about two to four weeks with the following five steps.

1. Understand the current state: write down the past month's requests by type as far as you remember, and who handled them
2. Decide the entry point and rules: choose one entry point (channel or form) and set hours, target reply time, urgent contact, and backup person
3. Decide how to record: keep date, requester, content, owner, and status in a spreadsheet or ticketing tool, with minimal fields
4. Turn top requests into guides: make short guides with screenshots in order of frequency and link them from the notice
5. Announce and review: introduce it at onboarding or meetings, and after a month check volume and handling time to revisit guides, owners, and whether to outsource

Unifying the entry point in step 2 and getting people to actually use it are different things. When announcing, prepare a one-line message such as 'if you are stuck, start here', and for the first few weeks patiently redirect requests that come to individuals back to the entry point.

- The entry point exists, but private chats and verbal requests keep being accepted, so it is barely used
- There is one handler and no backup for absences or leave
- A feature-rich ticketing tool is adopted at the start, and too many fields lead to empty records
- Requests are not recorded, so volume and trends are unknown and guide priorities cannot be set
- Many guides are written at once, the team gets exhausted, and they go stale
- Everything is handed to an outsourcer without sharing internal context or permissions, which increases back-and-forth
- Notices of hires and leavers do not reach the desk, delaying account creation and suspension
- No contact is defined for urgent security matters, delaying the first response

Most failures come not from picking tools or outsourcers but from missing rules and nothing that encourages use. The last two, the route for hire and leaver notices and the urgent contact, directly affect damage when something goes wrong, so they are worth deciding first.

- A single entry point is decided and announced to all employees
- Business hours and a target time for first reply are defined
- A primary handler and a backup for absences are named
- Contact method and number for urgent matters (security incidents, lost PCs) are defined
- Requests are recorded with date, requester, content, owner, and status
- Guides exist for the most frequent requests
- The guides' location is decided and linked from the entry point notice
- A route for hire and leaver notices to reach the desk is decided
- What employees can do themselves, such as password reset, is sorted out
- There is a monthly review of volume and handling time
- If outsourcing, scope, out-of-scope rates, response time, and cancellation terms are confirmed in writing
- Even with outsourcing, an internal contact and an approver are named

An internal IT help desk can start without a grand department or tool: choose one entry point, set intake rules and a backup person, and record requests. Then turning the most frequent requests into guides steadily reduces interruptions. Choose in-house, outsourced, or hybrid to fit your situation, treat costs as general guides, and compare quotes from several providers under the same conditions. A good first step is to write down the past month's requests this week.