Hardware Lifecycle Management: How to Manage IT Assets
Everything a vendor says about lifecycle management is wrapping around those three figures.
I have sat through enough conversations about hardware lifecycle management to notice that the vocabulary expands much faster than the substance.
There are five stages, four tool categories, three overlapping acronyms and a maturity ladder. All of it is useful. None of it is the point. Strip the language away and a company’s competence reduces to three measurements: how quickly a configured device reaches a new hire, what proportion of devices come back when people leave, and what one device costs per year of service.
If you cannot state those three, no amount of framework helps. If you can, the framework tells you which stage is dragging the worst one down.
The observation: the stages are fine, the seams are not
Planning, procurement, deployment, maintenance and operation, then retirement and disposition. Each has mature tooling and well-documented practice. I cannot think of a single one that is genuinely hard.
The seam between two stages, on the other hand, typically has no owner at all.
EZO, which builds the AssetSonar ITAM platform, reaches the same diagnosis in its guide to IT lifecycle mistakes, where a failed offboarding is described as a cross-functional handoff problem rather than an offboarding problem. I find that framing more useful than any stage-by-stage checklist.
Look at where the information actually lives in a 200-person distributed company. Purchases sit in Amazon Business or a reseller account. Configuration sits in Jamf or Intune. Faults arrive as Jira tickets. Departures are events in the HR system. Disposal lives with an outside ITAD supplier. That is five places, and each gap between them is bridged by somebody retyping serials, addresses and dates.
Three of those crossings reliably cost the most money. A device bought and never enrolled is invisible to security. A device offboarded and never retrieved combines asset loss with data exposure. A device past its refresh date but still in service fails audits and slows down whoever is using it.
The explanation: the terminology hides who does the work
Three terms overlap here and the confusion is expensive, because it obscures which supplier is going to physically touch your hardware.
IT asset management is the widest: hardware, software licences and cloud subscriptions in one register. Hardware asset management narrows that to the physical estate: which serial belongs to whom, and when its warranty runs out. Lifecycle management is what sits above both, because the order still has to be forecast, a price still has to be agreed, the machine still has to reach your MDM, a courier still has to be booked, a battery still has to be swapped, and somebody still has to fetch the laptop when its user hands in notice.
Which gives a clean test. Asset management tells you what you own. Lifecycle management is everything you then do about it, and only one of those two can be bought as software.
The discipline has moved up the agenda for three reasons worth separating. Distributed hiring means issuing equipment in countries with no office and no vendor account, each one adding currencies and customs rules. Compliance work has arrived at the endpoint: teams tell us three questions surface the moment SOC 2 or ISO 27001 preparation begins. Where is each machine. Who is able to log into it. What method destroys the data when it is retired. And finance now expects a defensible total cost of ownership per device instead of a lump procurement line, which quietly rewards vendors who publish rates.
The evidence: what each failure sounds like in a buyer’s own words
I collect these because they are more precise than any survey.
On sourcing, an IT manager evaluating us: “We use Amazon for UAE, a local reseller in India, and a different one in the UK.” Between five and ten such relationships is normal. Every quote comes in its own layout and its own currency, and each border ends the warranty that preceded it.
On growth, a buyer explaining why the old process gave out: “We went from 70 to 200 people in one year, what worked before doesn’t scale.”
On security, a CISO describing the reimbursement model: “Employees buy their own devices and we reimburse, no consistent security controls.”
On offboarding, in almost identical words every time: “Our offboarding process is ‘please mail it back’ and hope for the best.”
Get Tom Stawarski’s stories in your inbox
Join Medium for free to get updates from this writer.
And on the state of the register, from a team that had stopped pretending: “We don’t even know how many devices are unaccounted for.”
The last one is worth dwelling on, because it is honest and common. Capterra’s 2022 offboarding survey puts a number beside it: “71% say at least one employee didn’t return company-owned equipment, like a laptop or smartphone”, with the loss valued at $1,963 in equipment per non-returning employee. That figure measures incidence, employers burned at least once, so treat it as a statement about how ordinary the failure is. And the hardware is the floor, since whatever data sat on an unwiped machine leaves with it, against a 2025 global average breach cost of $4.4 million in IBM’s reporting.
The measurement I trust more than a self-assessment
Maturity ladders have a flaw: you score yourself using the same records whose accuracy is in question.
So do this instead. Export serial numbers from three places, your MDM, finance’s fixed-asset register, and whatever inventory sheet IT maintains. Count how many serials appear in exactly one of the three lists.
That orphan count is the whole health of your lifecycle in a single figure, and it is unarguable in a way a questionnaire is not. A connected or automated process produces something near zero. A reactive one usually produces more than anybody at the table expected.
Do it every quarter, one country at a time. Wherever you began hiring most recently is where those exports part company soonest, and where the next incident sits waiting.
Five questions an IT leader should then be able to answer without opening a ticket: where every device physically is; what share of devices offboarded in the last twelve months came back; how many are past refresh and what replacing them costs; how long a new hire in your newest country waits for a configured machine; and the all-in annual cost of one device including service, retrieval and storage. Fewer than half answerable means connecting your ITAM and MDM before shopping for anything, since automation amplifies whatever process it finds.
The refresh detail almost nobody sets correctly
One blanket replacement rule is the default and it is wrong in both directions.
Texas DIR guidance puts laptops at three to four years and desktops at four to five. California Energy Commission analysis gives monitors about 6.5 years of design life and small-scale servers about five.
Run monitors on the laptop cycle and you replace perfectly serviceable screens on a schedule invented for machines that get carried around in bags. Nobody notices, because the line item reads as a laptop refresh.
At the other end, the routing decision for a returned device follows the same logic. Under two years and healthy, back into the deployment pool. Two to four years, sell while resale value survives, which our buyback puts at 20 to 70% of initial value by condition. Beyond that, or damaged, certified recycling. Skip that step and you are feeding a stream the UN’s Global E-waste Monitor 2024 sizes at 62 million tonnes for 2022, with documented proper collection and recycling covering under a quarter of it, 22.3%.
The disclosure, and the questions I would ask us
We sell the physical layer, so weigh what follows accordingly.
Machines are purchased for whatever the local shop charges, through a reseller network of 600+ reaching 180+ countries, and take three business days on average to arrive, with enrollment attached at purchase so the machine claims itself into Jamf, Intune or JumpCloud on first boot. Repairs run in-country with a loaner bridging the gap, service jobs at $70 to $150 and any out-of-warranty work quoted before it begins. Retirement runs on an HRIS trigger, a courier at the employee’s address, and a certificate filed against the asset. The platform fee stays at $99 monthly at any fleet size, with no per-seat charge and nothing below 100 devices, and the pricing page carries the rate card plus a calculator you can run against your current vendor.
Ask us the same five questions you would ask anyone: where devices are sourced, whether enrollment happens at purchase, how the pricing is structured, what minimum applies, and what recovery rate we will put in writing. Then put the two we answer badly: hardware is never leased or financed by us, in any market at all, and what we track is the fleet in our care rather than everything on a network, which leaves discovery to a register running alongside.
For contrast on minimums, the entry plan at Workwize begins at 150 users, per its published tiers, which decides matters for a smaller team long before fit comes up. And on pricing structure, one buying team told us an incumbent platform had quoted $3,300 for a laptop retailing at $2,000, which is why the all-in figure on one named configuration, in writing, is the only comparison worth building.
The reflection
The thing I would tell a team starting this from scratch is to resist buying anything for a month.
Produce the three numbers. Produce the orphan count. Both are free, both take days rather than quarters, and together they name the stage that is actually costing you money. Almost every expensive mistake I have watched in this category came from buying software for a problem that turned out to be physical, or hiring hands for a problem that turned out to be a broken record.
The frameworks are fine. They are just downstream of knowing which of those two you have.
Read the full story here: https://www.tequipy.com/blog/hardware-lifecycle-management
