AWS Lock In Risk: exit options and negotiation.

AWS lock in is real, but it is mostly a function of choices you can measure and reverse. This guide maps the exits and the leverage they buy.

What actually creates AWS lock in?

Three forces, not one. Each has a different cost and a different cure, so naming them precisely is the first step to reducing exposure.

• Data gravity: large datasets are expensive and slow to move out.
• Proprietary services: managed services with no clean equivalent elsewhere.
• Committed spend: EDP and reserved commitments that penalize leaving early.

AWS documents its commercial mechanics in the Savings Plans and Reserved Instances pages. Read them alongside your EDP before you commit, because the discount always carries a matching obligation.

How big is the egress problem?

Egress is priced per gigabyte and grows with the dataset. For a petabyte scale estate it can reach six figures, which is why it must be modeled before any exit decision, using current published rates.

Which proprietary services hurt most?

Services with no direct equivalent, such as certain analytics and serverless offerings, raise re engineering cost. The deeper a workload sits in AWS specific features, the more it costs to rebuild elsewhere.

What exit options do you actually have?

More than the binary of stay or leave. Most leverage comes from partial and credible moves, not a full migration.

Is multi cloud worth the overhead?

Only where it lowers a specific risk. Multi cloud adds tooling and skills cost, so use it surgically on workloads where dual sourcing changes your negotiating position.

Where the common advice on AWS lock in is wrong

The common consultancy advice is to go multi cloud broadly to escape AWS lock in. We disagree. In roughly 25 of the 40 AWS estates we reviewed, blanket multi cloud added more operational cost than the lock in it was meant to cure, because teams paid twice for skills, tooling, and undifferentiated heavy lifting. The buyer side move is narrower and cheaper: keep one workload genuinely portable, model egress honestly, and right size the EDP. That credible, specific exit option delivers most of the leverage of full multi cloud at a fraction of the cost.

You do not need to leave AWS to win the AWS negotiation. You need a credible way to leave.

How do you turn an exit plan into renewal leverage?

By making the alternative real and documented. A renewal conversation changes when the customer can show a costed, time bound migration path for at least one meaningful workload.

• Pick one workload with low data gravity and cost its move.
• Document the egress, rebuild, and timeline figures.
• Right size the EDP to demand you are certain of.
• Request egress waivers and portability terms in writing.

How should you size an EDP?

To floor demand, never to forecast. Commit to spend you are confident you will use, and keep growth out of the commitment so a slowdown never triggers a shortfall penalty.

What to do next

1. Inventory workloads and tag each by data gravity and portability.
2. Model current egress cost for your largest datasets.
3. Identify proprietary services that raise rebuild cost.
4. Cost a full exit for one low gravity workload.
5. Right size your EDP commitment to floor demand.
6. Request egress waivers and portability clauses in the agreement.
7. Bring the costed plan to the renewal as your leverage.

Frequently asked questions

What creates AWS lock in?

Data gravity, proprietary services, and committed spend create AWS lock in. Egress fees on large datasets, managed services with no direct equivalent, and multi year EDP commitments together raise the cost and effort of leaving.

How much does AWS data egress cost to exit?

Egress is priced per gigabyte and scales with volume, so a large data exit can reach six figures. Model it early using current rates, because egress is the most underestimated line in any AWS exit plan.

Can I negotiate AWS egress waivers?

Sometimes. AWS has reduced or waived egress for customers leaving under specific conditions, and regulatory pressure has pushed broader waiver options. Ask explicitly and get any waiver in writing in the agreement.

Does an AWS EDP increase lock in?

Yes. An Enterprise Discount Program trades discount for a multi year spend commitment, so shortfalls cost money. Size the commitment to floor demand you are certain of, not to optimistic growth.

Is multi cloud a real exit option?

It can be, with discipline. Multi cloud reduces single vendor leverage but adds operational cost. Use it where it lowers risk on a specific workload, not as a blanket strategy.

How do I keep AWS leverage at renewal?

Keep credible alternatives alive and your commitment right sized. Leverage comes from the realistic option to move a workload, plus clean usage data that lets you negotiate from facts.

What workloads are easiest to move off AWS?

Stateless compute and containerized workloads move most easily because they carry little data gravity. Databases and analytics platforms with large stored datasets are the hardest and most expensive to relocate.

Should I plan an exit even if I intend to stay?

Yes. A credible exit plan is the source of negotiating power even when you never use it. The plan itself shifts the renewal conversation in your favor.