Five ways to solve this, honestly compared
BuiltForProd is one of them, and not always the right one. Here is what each alternative is good at, what it costs you, when to pick it over us, and how to find out which one you are before you commit.
The choice you are actually making
Every team that needs a production foundation has the same five options: build it in-house, hire a consultancy to build it, move onto a platform as a service, run your cloud provider’s landing-zone accelerator, or keep what you have. BuiltForProd is a sixth — finished code, customized, deployed into your accounts by the people who wrote it, and handed over — but it only makes sense against the alternatives, so here they are.
We have tried to be fair rather than flattering, which means this page tells you where we lose. Vendor offerings change, so treat the descriptions of categories as general characterizations and check them against whatever you are actually being quoted.
One path runs across all six: find out first. The BuiltForProd Assessment measures what you run today against the Standard and says, in writing, which option you are, before you commit to any of them. It is described below the five.
Build it in-house
Hire platform engineers and build the foundation yourself, to your own conventions.
What it is good at
You get exactly what you want, the knowledge stays in the building, and the team that runs it is the team that built it.
What it costs you
Time and hiring. A foundation of this scope is months of work by people you have to find first, and the quality depends on which engineers you happened to hire. The knowledge also concentrates: one or two people end up holding it.
Pick it over us when
Platform engineering is a deliberate core competency, and you can wait for it.
Hire a consultancy
A project-based engagement with a partner or DevOps firm that builds a bespoke environment for you.
What it is good at
Flexible scope, a team that adapts to your situation, and the ability to cover application work at the same time.
What it costs you
Every engagement starts from a blank repository, so quality and speed depend on the engineers staffed on your project. Access often stays with the vendor afterward, and the knowledge frequently does too. Documentation quality varies widely.
Pick it over us when
You need a bespoke program or application modernization that goes beyond a standard foundation.
Move to a platform as a service
Run the application on a hosted runtime and let someone else own the infrastructure entirely.
What it is good at
Fastest path to a running application, very little to operate, and an excellent choice while there is nothing at stake.
What it costs you
You do not own the environment, so the compliance, network and data residency answers are theirs and not yours. Costs scale with usage and the exit is a migration. When an enterprise buyer asks for your account structure, there is nothing to show.
Pick it over us when
The application is early and simple, and production does not yet carry consequences.
Run a landing-zone accelerator
Use your cloud provider’s own landing-zone service or solution to set up the account structure and guardrails.
What it is good at
Native, supported by the provider, and quick to get a baseline account structure in place at no license cost.
What it costs you
The foundation is a managed service rather than code you own, and the parts that matter most — the network design, the identity model, framework mappings and the delivery pipeline — are largely left to you. Coupling to the service is hard to undo later.
Pick it over us when
You want a provider-managed landing zone and accept its model and its boundaries.
Stay as you are
Keep the environment you have and fix things as they come up.
What it is good at
No spend, no disruption, and it is genuinely the right answer more often than vendors admit. If nothing is at stake yet, nothing needs to change yet.
What it costs you
The bill arrives as events rather than invoices: a security questionnaire you cannot answer, an audit that takes two engineers for two weeks, an outage a customer tells you about, the infrastructure lead resigning. Each one costs more than the fix would have.
Pick it over us when
Production does not yet carry consequences, and you are honest with yourself about when that changes.
Find out first
You do not have to guess which option you are.
Every row on this page assumes you know how far your current environment is from production. Most teams do not, and the ones that guess usually guess kindly. The BuiltForProd Assessment replaces the guess with a measurement: a fixed-fee, evidence-based assessment of what you run today against the eight properties of the Standard.
You get findings, priorities, a future-state architecture and a phased migration plan, in about two weeks. The fee is credited up to 50% toward a Baseline within 90 days. And if the honest conclusion is that you should stay as you are, fix one gap yourselves, or hire, the report says so. A customer who did not need the thing they bought is not a reference.
What it settles
- Whether staying as you are is still safe, and for how long.
- Whether the gap is one thing you can fix yourselves, or a foundation.
- What a Baseline would replace, keep and migrate, and in which order.
- Which controls your next audit will ask for that do not exist today.
- Who could run the result, and what would have to be true for that.
Side by side
| Dimension | BuiltForProd | In-house build | Consultancy | PaaS | Landing-zone accelerator | Stay as-is |
|---|---|---|---|---|---|---|
| Time to a production foundation | Weeks of hands-on engineering, engagement-specific | Months, after hiring | Months, project-dependent | Days for the app; no foundation | Days for the base; design still yours | Never, by definition |
| Who owns the code | You, full source in your repositories | You | Usually you; quality varies | Nobody — there is no code to own | A managed service, plus some code | Whatever exists today |
| Is there a published standard | Yes, eight properties mapped to controls | Whatever the team decides | The vendor’s conventions | The platform’s conventions | Provider guardrails | No |
| Compliance evidence | Produced by the infrastructure, mapped to 11 frameworks | If someone builds it | If it was scoped | Their certifications, not yours | Guardrails; mapping largely yours | Screenshots, every cycle |
| Day-2 operations | Runbooks, troubleshooting, drift detection, upgrades | Internal effort | A new statement of work | Handled, and opaque | Provider releases; the rest is yours | Improvised |
| Who operates it | You, or Managed through your own access | You | Often the vendor, with retained access | The platform | You | Whoever is free |
| Vendor access afterward | None — no principal, no credential | Not applicable | Frequently retained | Full, by design | Provider service roles | Check who still has admin |
Who we are not for
- Prototypes, hackathons and products with nothing at stake yet
- Teams looking for the cheapest fix to a single symptom
- Buyers who want hourly engineers with no opinions of their own
- Teams who want to build every layer themselves, deliberately
- Problems that live in the application code rather than the platform
- Anyone buying a single tool rather than a foundation
- Organizations unwilling to adopt ownership, runbooks and change management
We’ll tell you if we’re not the right fit.
That is not a slogan we keep for the website. Qualifying out early is cheaper for both of us than a delivery that was never going to land, and a customer who did not need the thing they bought is not a reference.
If you are not sure which side of the line you are on, the two-minute self-check will usually settle it, a 30-minute call will settle it honestly, and an Assessment will settle it in writing.
Questions about the alternatives
Why not just hire two platform engineers?
Sometimes you should. Hiring gives you permanent capacity and deep context, and if platform engineering is a core competency, build it. The trade is time and concentration risk: you have to find the people first, and the resulting system is usually understood by the two people who built it.
How are you different from a DevOps consultancy?
A consultancy sells hours and starts from a blank repository. We start from finished, tested code and sell a defined outcome, at a fixed fee, with no standing access afterward. The result does not depend on which engineer happened to be available the week you signed.
We already use our cloud provider’s landing zone. Is this redundant?
No, but it overlaps. An accelerator gives you an account structure and guardrails as a managed service. The Baseline gives you the whole foundation as code you own, including the network design, the identity model, the framework mappings and the pipeline, and the AWS edition ships a documented migration path from AWS Control Tower.
Are we locked in?
No. Your repositories, your cloud accounts, open-source tooling underneath, and no standing access for us after handover. There is no control plane of ours to depend on and no runtime to rent. You can keep every part of what we built and never speak to us again.
What if we only need one piece of this?
Then buy one piece, or none. Blueprints can be added to an existing Baseline one at a time, Managed is optional, and if the gap is narrow enough to fix yourselves, an Assessment will say so. We would rather tell you that than sell you a foundation you do not need.
Where do you lose?
Against a team that treats platform engineering as a core competency and has the time to build it. Against a PaaS when the application is early and nothing is at stake. Against a consultancy when the work is application modernization rather than a foundation. Against a specialist tool when the foundation already exists and only workflow tooling is missing.
How do we find out which option we are?
Start with a BuiltForProd Assessment: a fixed-fee, evidence-based assessment of what you run today against the eight properties of the Standard. You get findings, priorities, a future-state architecture and a migration plan in about two weeks, and the fee is credited up to 50% toward a Baseline within 90 days. If the honest answer is one of the other five options, the report says so.
How do we start?
Book the 30-minute call. You meet an engineer who scopes engagements, not a salesperson; bring what you run and which of the options you are leaning toward, and you leave with a next step, a fixed fee, or an honest no. If you want a reading first, take the two-minute self-check on the Assessment page.
If the comparison lands our way, book a call, start with the Assessment, look at the whole portfolio, or read how ownership and access actually work.
Not sure which of the five you are?
A 30 minutes call with an engineer who scopes engagements, not a salesperson. Tell us what you run and which option you are leaning toward. You leave with a next step, a fixed fee, or a straight answer that one of the other five is better.