The model decision matters more than the vendor
Whether you run a build as an SDaaS retainer or a fixed-scope project is a bigger decision than which vendor you pick. The wrong model will fail on the right team; the right model will forgive a mediocre team. Choose the model first.
When to choose project-based
Project-based works when scope is real. Genuinely pinned. Think: a migration to a new CMS where the target CMS and the content model are decided, a regulatory integration where the spec is written, a one-shot dashboard against a frozen data warehouse.
The tell: stakeholders can sign off a requirements doc without caveats. If anyone says "we'll know more when we start," it's not project-based — it's SDaaS wearing a project hat, and it will go over.
When to choose SDaaS
SDaaS works when the product is still moving and you want to keep it moving. The deliverable isn't a shipped artifact — it's a running capability.
- New product builds where requirements will change every sprint.
- Mature products where the backlog is continuous and prioritization is the hard problem.
- Platform and API work where the consumer teams keep evolving their needs.
- AI product builds — by definition the model, the prompts, and the interface co-evolve.
What you get wrong by defaulting to project-based
Mid-market buyers default to fixed-scope because it feels safer to the finance team. In practice, three things happen by month four:
- Scope creep — every legitimate need becomes a change order, each one negotiated.
- Incentive misalignment — the vendor is rewarded for minimizing work against a spec; the client needs the product to succeed against users.
- Quality debt — corners get cut to hit the fixed date; the debt comes out of the maintenance budget that doesn't exist.
SDaaS eliminates (1) and (2) by design. It solves (3) only if the quarterly KPI review happens.
How NUUN scopes the decision
We scope a pilot project before the retainer starts — three weeks, fixed fee — to build real trust and surface the scope-stability question. If the pilot proves scope will move, we recommend SDaaS. If the scope holds, we recommend a fixed-scope project next. Either way, the client gets a written recommendation with the math — not a sales pitch for the higher-margin option.
Sources & further reading
- NUUN Software & Product Development service pillar
- Software Development as a Service (SDaaS) service page
- Thoughtworks Technology Radar — https://www.thoughtworks.com/radar
- State of DevOps Report — https://dora.dev/research/
- McKinsey — Developer Velocity Index — https://www.mckinsey.com/