From-Scratch Build · IT Asset Management
A self-hosted system for tracking every piece of equipment in a lab — what it is, where it lives, and who has it right now. Built from scratch to learn how IT asset management actually models the messy real world of laptops, licenses, cables and people.
What it is
An asset inventory system answers two questions that get surprisingly hard at scale: what do we own? and where is it right now? In a busy lab, equipment is constantly borrowed, returned, lost and replaced. Without a system, the answer to "who has the spare laptop?" is a group chat and a shrug.
I built this as a documented, self-hosted inventory. The heart of it is a clean classification: every physical or digital thing is sorted into a category, given an asset tag, and checked out to a person when borrowed. Settings define the structure — categories, models and locations — so the data stays consistent as it grows.
The core idea I wanted to learn: good inventory is a taxonomy problem before it's a software problem. Decide cleanly whether a thing is an asset, a consumable, an accessory or a component, and every later query — depreciation, restocking, who-has-what — falls out naturally.
The model
Everything in the inventory is one of these. Getting the category right is the whole discipline.
Durable items of value the organisation owns and uses — laptops, desktops, servers, phones. Each gets a unique tag and a lifecycle.
Permissions to use software or a service, with seat counts and expiry dates — the digital equivalent of an asset.
Supporting hardware like mice, keyboards and monitors — not the main device, but checked out to people all the same.
Items depleted with use — toner, ink, paper. Tracked by quantity so you know when to reorder.
Parts inside a larger asset — CPUs, RAM, drives. Essential, but not standalone devices.
The users equipment is assigned to. This is what turns "we own a laptop" into "Edu has this laptop."
Structure
Before adding a single item, you define the scaffolding. Three setting types enforce consistency so the inventory doesn't decay into free text:
Logical groups like "laptops" or "wireless keyboards." Every asset and accessory must belong to one, and categories carry inherited rules — such as whether to email a user on checkout.
Templates for creating assets, capturing shared attributes — manufacturer, category, custom fields, depreciation. Standardisation, so two identical laptops are described identically.
Where things physically live. A shelf can be addressed like coordinates — shelf 2, row 4, column 5 — so an item's spot in the lab is unambiguous.
Each asset gets a QR code; scanning it on a phone opens that asset's detail page instantly. The bridge between the physical shelf and the digital record.
Workflows
Once the structure exists, day-to-day use is a small set of repeated actions:
In my build I focused on the documentation and data model — the rules that make a shared inventory trustworthy rather than just a spreadsheet everyone ignores.
Reflection