The three options at a glance
| Global SI | Boutique partner | In-house team | |
|---|---|---|---|
| Best for | 100+ apps, multi-year programmes | 1-50 apps, speed + senior engineers | Long-term platform ownership |
| Day rate | High (blended, many juniors) | Mid (mostly senior) | Salaried + ramp-up time |
| Speed to start | Slow (weeks of SOW) | Fast (days) | Slow (hiring + learning) |
| AWS MAP funding | Yes | Yes (if AWS Partner) | No (funding needs a partner) |
| After cutover | Rolls off to next account | Stays for run/optimisation | Owns it permanently |
There is no universally “best” option - the right answer depends on your app count, timeline, in-house capacity, and whether you need to own the platform afterwards. The rest of this guide is how to work out which fits.
This guide is for CTOs, engineering leads, and founders who are comparing ways to get onto AWS and want a straight answer on cost and fit rather than a sales pitch. If you already know you want to run the numbers on the migration itself, see our AWS migration cost calculator. For the end-to-end process, see the complete AWS cloud migration guide.
Comparing partners for your migration?
Book a free 30-min call
The real decision: partner vs in-house team
Before comparing partners against each other, decide whether you need a partner at all. This is the “AWS partner vs in-house migration team” question, and it comes down to total cost of ownership, not day rate.
When in-house wins
- You are migrating continuously (not a one-off) and will keep evolving the platform for years.
- You already have 2+ engineers with real AWS depth (not just certification).
- The workload is your core product and you want the knowledge to stay in the building.
When a partner wins
- The migration is a one-time or infrequent event - hiring permanent senior cloud engineers for a 3-month project is expensive and slow.
- You need it done in weeks, not after a 2-3 month hiring cycle plus ramp-up.
- You want AWS funding: MAP and PoC funding require an AWS Partner to submit the application. An in-house team cannot access it directly.
The TCO comparison people miss
One-off migration: in-house vs partner (illustrative, mid-market)
| Cost element | In-house build | Boutique partner |
|---|---|---|
| Senior cloud engineers | 2 hires × 3-6 months ramp before productive | Productive from week 1 |
| Fully-loaded cost | €120K-200K/yr each (salary + overhead) | Project fee, scoped |
| AWS MAP funding | Not available | Covers 70-80% of partner fee |
| Time to first cutover | 4-8 months (hire → learn → deliver) | 4-12 weeks |
| Knowledge after | Stays in-house (the point) | Transferred at handover |
For a one-time migration, a partner is almost always cheaper after MAP funding than hiring - and months faster. For a permanent platform you keep changing, in-house ownership wins on the multi-year horizon. Many teams do both: a partner runs the migration and trains the in-house team who then own the run phase. See our migration guide for how a handover-focused engagement is structured.
Global SI vs boutique partner
If you have decided on a partner, the next fork is a global systems integrator versus a specialist boutique. Both can be AWS Partners; the difference is structure, not badge.
| Dimension | Global SI | Boutique partner |
|---|---|---|
| Team composition | Blended - a few architects over many junior consultants | Mostly senior engineers doing the work |
| Pricing model | High blended day rate, large fixed scopes | Lower effective rate, tighter scopes |
| Minimum engagement | Often large (six-figure floors) | Flexible, can start small |
| Process overhead | Heavy (governance, change control) | Light |
| Speed to start | Weeks of SOW negotiation | Days |
| Best fit | 100+ apps, multi-year, board-level programme | 1-50 apps, needs pace and senior hands-on delivery |
| Risk | You may get juniors on the keyboard | Less bench depth for very large, parallel programmes |
Neither is “better.” A global SI is the right call for a 200-application, multi-year enterprise programme that needs deep bench, formal governance, and a name the board recognises. A boutique is the right call for a startup-to-mid-market migration that needs senior engineers moving fast without six-figure minimums and layers of process. The common trap with a global SI is paying architect rates for junior delivery; the common trap with a boutique is engaging one without enough bench for a very large parallel programme. Ask directly who will be on the keyboard.
The questions that actually reveal a good partner
Day rate tells you almost nothing. These questions do:
- “Who specifically will do the work, and how senior are they?” Ask for the names and the seniority mix on your project, not the company average.
- “Do you submit AWS MAP / PoC funding for us, and have you done it before?” A real AWS Partner does this routinely. Vagueness here is a red flag - it is worth 70-80% of the fee.
- “What happens after cutover?” A migration that lands and then abandons you in week two is worse than no migration. Ask about the run/optimisation phase and knowledge transfer.
- “Show me a comparable migration you delivered and what went wrong.” Everyone shows wins. A partner who can talk candidly about a hard cutover is one who has actually done it.
- “Is the architecture documented as code (Terraform/IaC) that we own?” If the environment only exists in someone’s head or console, you are locked in.
- “How do you handle our compliance requirements?” For regulated workloads (SOC 2, ISO 27001, FCA, HIPAA), the partner should speak fluently about controls, not treat it as an add-on. See our ISO 27001 on cloud guide.
How partner pricing really works
The headline day rate is the least useful number. What determines your actual cost:
- MAP funding changes everything. AWS covers 70-80% of qualifying partner engagement costs. A €92,000 project can net out to €18,000-28,000 out of pocket. Only an AWS Partner can submit this - it is the single biggest lever on price. See the detail in our migration cost calculator.
- Seniority mix, not headline rate. A boutique charging a higher day rate but staffing two senior engineers often costs less in total than an SI blending five people where three are juniors learning on your budget.
- Scope discipline. Fixed-scope, outcome-based engagements protect you from the open-ended time-and-materials creep that inflates large-SI programmes.
- The run phase. Cheap migration + expensive or absent post-cutover support is a false economy. Price the first 6 months of operation, not just the cutover.
For the underlying migration numbers by company size, the cost calculator breaks down one-time project cost and ongoing AWS spend for startup, mid-market, and enterprise.
Specialist scenarios where fit matters most
Some migrations have a specific shape that narrows the field:
- Heroku / PaaS → AWS. Look for a partner who has done PaaS-to-hyperscaler moves specifically - the gotchas (managed add-ons, dyno-to-container mapping, connection pooling) are particular. See migrating from Heroku to AWS.
- Database-heavy (Oracle → PostgreSQL, legacy SQL). This needs DMS/SCT depth, not generalists. See our Oracle to PostgreSQL guide.
- Regulated / security-first. Compliance evidence should be a first-class output of the migration, not bolted on later.
- Large portfolio (100+ apps). This is where a global SI’s bench and governance genuinely earn their premium - or a boutique that can prove parallel-wave capacity.
Frequently asked questions
Is it cheaper to use an AWS partner or an in-house team for migration?
Should I choose a global systems integrator or a boutique AWS partner?
What should I ask an AWS migration partner before hiring them?
How does AWS MAP funding affect the cost of hiring a migration partner?
How do I choose an AWS partner for a Heroku to AWS migration?
Talk to a boutique AWS partner
We are a boutique AWS partner - senior engineers, MAP funding handled for you, and a run phase after cutover rather than a disappearing act. If that fits your migration, see our cloud migration services or consulting. If you are a large enterprise with a 100+ application programme, we will tell you honestly whether a global SI is the better fit for your scale.
Want a straight answer on partner fit and cost?
Tell us your app count, timeline, and whether you plan to own the platform afterwards, and we will tell you honestly whether a boutique partner, a global SI, or an in-house build fits - including what MAP funding would cover. Book a call or Send us a message.