How Third-Party Maintenance Contracts Actually Work: SLAs, Parts, and Escalation
Most infrastructure leaders know the headline pitch for third-party maintenance: the same break/fix coverage the OEM sells, at 40–70% less. Fewer have seen inside a contract before signing one. This is the walkthrough we wish every prospect had before their first call — including the parts that should make you ask harder questions.
What a TPM contract covers
At its core, a third-party maintenance agreement covers three things for a defined list of assets (by serial number):
- Technical support — someone to call when a device faults, degrades, or behaves strangely. With a good provider, that call lands with an L2/L3 engineer, not a script-following triage queue.
- Parts replacement — failed components (drives, controllers, power supplies, line cards, fans) replaced under a defined SLA.
- Onsite labor — a field engineer dispatched to swap the part when remote hands aren’t appropriate.
What it deliberately doesn’t cover: software updates and new firmware from the OEM. Your gear runs the code it has. For EOSL hardware this is moot — the OEM stopped shipping updates anyway — and it’s the reason TPM fits stable estates rather than platforms still evolving under you.
SLAs, decoded
TPM service levels mirror the OEM tiers, and the vocabulary is the same:
- 7×24×4 — around-the-clock coverage, four-hour onsite response with part in hand. For genuinely critical systems.
- NBD (next business day) — the workhorse tier for most of an estate.
- 8×5×NBD or parts-only — for dev, DR, and things that can wait.
Two details matter more than the tier names. First, response versus restore: a four-hour response SLA means an engineer and part arrive in four hours, not that you’re fixed in four. Ask how restore times are measured and reported. Second, SLA mixing: unlike most OEM renewals, a good TPM provider will happily put your production arrays on 7×24×4 and your DR site on NBD in the same contract. Right-sizing SLAs is where a chunk of the savings comes from — most OEM contracts have entire racks over-covered by default.
The part everyone should scrutinize: sparing
Parts strategy is what separates real providers from brokers. When a controller fails at 2 a.m., the question is not whether the part exists on the secondary market — it’s whether your provider has one within SLA distance of your site. Ask any prospective provider:
- Do you forward-stock spares against my specific configurations, or source on demand?
- Where is the stocking location relative to my sites, and what’s in it?
- How do you handle firmware/compatibility matching on replacement parts?
- What’s your parts quality process — testing, burn-in, failure-rate tracking?
A provider who answers with specifics is doing engineering. A provider who answers with “extensive global network” is doing logistics arbitrage, and you’ll find out the difference during an outage.
Escalation: who’s actually behind the phone
OEM support pyramids put thousands of L1 agents in front of a small L3 core. TPM firms invert this — their engineers are typically OEM-alumni specialists on the platforms they cover. When you evaluate a provider, ask to meet the engineers who’d own your account’s escalations, and ask how they handle a fault they can’t resolve: root-cause analysis capability, engineering lab access, and multi-vendor interoperability experience are the tells.
What the commercial terms look like
Expect per-asset annual pricing driven by device type, age, SLA tier, and location; co-termed contracts that align your maintenance renewals to one date (a quiet operational win); and flexibility to add and remove assets mid-term as you decommission or acquire. That last one matters — OEM contracts punish estate shrinkage, TPM contracts generally don’t.
Where to start
The standard entry point is a coverage assessment: you provide a serial-numbered inventory, the provider returns a quote per asset and tier. It costs nothing, it’s genuinely useful even if you renew with the OEM (leverage is leverage), and it tells you quickly whether the provider knows your platforms. You can check lifecycle status for your models in our EOSL database, and when you’re ready for real numbers, request a quote — we turn inventories around in 24 hours.
Get support for what you run
How we can help
Keep it running after end of support
Hardware or software, the end-of-support date doesn’t have to force a refresh. We keep enterprise infrastructure maintained, secure and under SLA long after the vendor moves on — typically at 40-70% below OEM pricing.
End-of-life support →Migration services
When you do decide to move, we plan and execute the migration. Your current environment stays under vendor support while your contract is active — and if the renewal lapses mid-move, our third-party support covers most issues until the last workload is off it.
Migration & hybrid cloud services →24×7 remote administration
Short on hands to run it day to day? Our NOC engineers monitor, patch and administer your environment around the clock — incident response included, at a fraction of the cost of an in-house night shift.
24/7 operations & remote administration →