Cloud credits are a discount on year one. They are useful, but they do not tell you what the platform will cost to run, and they should never be the reason a migration happens. Plan the move as if the credits did not exist, then treat them as a bonus.

I have seen credits win arguments that the business case could not. A provider offers a large credit package, the migration suddenly looks free, and the hard questions about run cost and operating model get pushed to later. Later arrives when the credits run out.

What do cloud credits cover?

Most credit programmes cover consumption for a fixed period, often twelve months, sometimes tied to a migration or a minimum commitment. They usually pay for compute, storage and some managed services. They rarely pay for:

  • the people who have to run the new platform
  • licences you still need, or now need twice during the transition
  • the tooling for monitoring, security and backup
  • the inefficiency of workloads that were moved as they are, without being right-sized

So even in year one, the platform is not free. The credits only hide part of the bill.

What happens when the credits end?

The full consumption cost appears, usually at the same time as renewal of everything else. If the environment was built while nobody was watching the meter, it is often oversized. Test environments that never switch off, storage nobody cleans up, instances sized for a peak that never came.

This is the point where finance asks why the cloud is more expensive than the data centre it replaced. The honest answer is that the cost was always there. The credits were covering it.

How to plan with credits

  1. Build the business case without them. Three years of run cost, licences, support and people. If the move only makes sense with credits, it does not make sense.
  2. Watch the meter from day one. Tag resources by owner and environment, set budgets and alerts, and review spend monthly even while it is covered.
  3. Right-size during the credit period. Use the free year to learn what the workloads really need, then adjust before you start paying.
  4. Read the terms. Check what the credits are conditional on: minimum spend, specific services, deadlines. A credit tied to a commitment can become an obligation.
  5. Put a date in the diary. Three months before the credits end, review the expected bill with finance. No surprises.

The short version

Credits are a good way to pay for learning. They are a poor way to make a decision. Take them, use them, but make the decision on the numbers you will live with in year two.

If you are comparing providers or proposals, the same rule applies to every discount. I wrote more about that in why the cheaper proposal usually costs more in year two.