IT helpdesk for an SME: how to organize

This article was machine-translated from French and is being reviewed. Prices are in Algerian dinars excl. VAT.

An helpdesk informatique is the single point of contact for support requests: a PC that won’t start, an unprinting printer, a password reset. Without a single entry point, requests get lost among calls, messages and hallway chats, and no one knows who handles what. For an SME the challenge is a simple circuit: who receives, who processes, when to eskalate, and how to keep track.

Structure 1: minimal internal helpdesk

You appoint an IT reference, possibly part‑time, who:

• receives requests via a form, a dedicated address or a ticketing tool;
• performs the first diagnosis: restart, cables, consumables, password reset after verifying the requester’s identity;
• escalates to the provider for failures requiring specialized skills (server, network, backup, security).

Advantages: quick response to minor issues, fewer calls to the provider, knowledge of the company.

Disadvantages: knowledge rests on one person. Appoint a deputy and keep documentation up to date, otherwise support stops when the reference is absent.

The main cost is the reference’s time. The maintenance contract covers escalation.

Example request form

`` Who? [name, position] What? [brief problem description] When? [date and time] Impact? [blocks entire company / blocks a service / inconveniences someone] Restart performed? [yes/no] Actions already attempted? [list]``

A short form with mandatory fields avoids exchanges like “did you receive my message?”.

Structure 2: multi‑level support (N1, N2, N3)

As the fleet grows, split requests into levels (L1, L2, L3 in English):

• N1: reception and first diagnosis. It records the request, qualifies it (priority), resolves simple documented cases (password, printer, connection) and escalates the rest.
• N2: technician. It performs an in‑depth diagnosis: workstations, printers, LAN, accounts, software. It can be internal or external.
• N3: expert. It handles complex cases: servers, backups, security, firewalls, business applications. This is often the external provider, sometimes the software vendor or manufacturer in case of a product defect.

Operating rule: each level documents what it did before escalating, so the next does not repeat questions. Disadvantage: more levels require more coordination and a tool to track transfers.

Helpdesk tools

Ticketing software

A ticketing software centralises requests, timestamps them, manages priorities and keeps the history. Real options include:

• Free software to host yourself : GLPI (ticket management, asset inventory, knowledge base and service level tracking, published by the French company Teclib'), Zammad (email support and other channels), osTicket, UVdesk, Request Tracker.
• Online services (paid for the essentials) : Zendesk, Freshdesk, HubSpot, Jira Service Management (Atlassian, with a limited free tier by number of agents: check current terms).

A self‑hosted tool adds a server to update and backup. An online service stores your tickets with the vendor: consider this if requests contain personal data. Selection criteria: ticket creation by email, priorities and deadlines, reports, user simplicity. Wi‑Fi access requests from visitors recur: plan them in the tool (see Secure guest Wi‑Fi).

Remote access tool

Remote maintenance speeds resolution but gives a third party control of a workstation. Common tools are RustDesk (free software, with a server you can host), AnyDesk and TeamViewer (commercial products whose free versions are for personal use only). Requirements:

• the user accepts each session, except for unsupervised access that is planned and controlled;
• named accounts, a strong password and two‑factor authentication when the tool offers it;
• a connection log and removal of access at contract end;
• a confidentiality clause: the technician sees the screen, thus any client, patient or accounting data;
• no login credentials are shared with someone who calls themselves: fake tech support is a known fraud;

For broader network access, see link sites via VPN and why a firewall in SMEs.

Knowledge base

An internal wiki gathers procedures (install a printer, change a password, connect to the VPN), FAQs, and contacts: provider, helpdesk address, emergency procedure. It allows N1 to resolve more cases alone. Labeling and the rack diagram have their place: see organizing a network rack.

Monitoring

A monitoring tool alerts when a disk fills, a service stops or a connection drops. Zabbix, Nagios and Centreon are free monitoring tools; Uptime Kuma, simpler, monitors service availability. Grafana displays dashboards but does not replace a monitoring tool. The goal is to detect failure before the user reports it: that is preventive maintenance in action.

Set deadlines (SLA)

Distinguish the support (the request is logged and processed) from the recovery (service is restored) and the final resolution (the cause is eliminated). Priority comes from impact and urgency. Here’s a template grid, without numeric values, to fill in with your provider:

Delays start from ticket logging. An urgent call must therefore end with a ticket created by N1. The numeric values depend on scope, coverage hours and price: they are set in the contract, and this guide does not provide them.

Documentation: the helpdesk’s secret weapon

Each intervention is recorded in the ticket:

• the observed symptom;
• the actions taken;
• the applied solution;
• duration;
• the cause, if known.

After a few months, this history shows recurring problems (same printer, same almost full disk) and allows root‑cause treatment. It is also the basis for the report provided under a contract.

When to outsource everything?

Without an internal technician, you can hand all support to a managed services provider (MSP) by contract. Advantages: no recruitment, extended hourly coverage possible per the contract. Disadvantages: dependence on the provider, and onsite visits when remote maintenance is insufficient. The contract must cover scope, deadlines, backups and reversibility.

When to scale up to AsloTech?

An internal reference person, complemented by an external provider, may suit a SME that does not want to hire a technician. AsloTech intervenes when the internal N1 or N2 needs an expert: server, backup, network, security. See the IT maintenance page to define your contract, or the guide IT maintenance contract. For equipment covered by network security, see our network security and firewall solution.

Organizing a working helpdesk

A good helpdesk rests on three pillars: a clear process (every request becomes a ticket), documentation (knowledge base and history) and written responsibilities (levels and deadlines in the contract). Start with the maintenance contract, then add a ticketing tool and a knowledge base. For hardware replacements that recur in tickets, see repair or replace a PC and renew the IT fleet. For backups, see the 3-2-1 rule.

Frequently asked questions

Also read