More teams are trying to buy execution differently. They want faster delivery, less overhead, and less time lost in meetings, revisions, and unclear ownership. At the same time, AI has changed the economics of first-pass production. Drafting, research, and iteration are cheaper than they used to be. That makes the old agency retainer harder to justify in cases where the buyer does not actually need open-ended capacity. But it does not mean retainers are always wrong. It means the buying decision should be more deliberate.
Core idea
The useful distinction is simple: an agency retainer usually sells access to a team’s time, while results as a service sells a defined business-ready output.
That difference affects almost everything downstream.
| Dimension | Agency retainer | Results as a service |
|---|---|---|
| Buyer is paying for | Ongoing access to a team | A defined business-ready output |
| Best fit | Priorities will shift and scope will evolve | The finish line and acceptance criteria are clear |
| Change during the engagement | High and expected | Lower and more tightly managed |
| Client management load | Often higher | Often lower if the provider owns execution well |
| Main failure mode | Scope drift without a real finish line | False certainty when success is still poorly defined |
In a retainer model, the buyer is paying for ongoing availability. The scope may move. Priorities may change. Part of the value is flexibility. That can be appropriate when the work is exploratory, politically complex, or hard to define in advance.
In a results-as-a-service model, the buyer is paying for a specific outcome with clear acceptance criteria. The provider is responsible for getting from problem to usable deliverable without charging for every intermediate step. The value is not “hours worked.” The value is that the buyer receives something they can launch, use, or act on.
That is why the comparison should not start with price alone. It should start with what the work actually is.
If the need is “help us think through several shifting priorities over the next quarter,” a retainer may be the right tool. If the need is “build the landing page, fix the analytics setup, or ship the workflow automation,” the buyer may be better served by paying for the result rather than the process around it.
Where this helps in practice
Results as a service is strongest when the problem is clear enough to define a useful finish line.
Common examples:
- a landing page that needs stronger positioning, tighter structure, and production-ready delivery
- an analytics setup that needs a tracking plan, event model, and dashboard the team can trust
- a workflow that needs to remove manual handoffs without creating new operational confusion
- a content system that needs clear page architecture instead of vague publishing effort
In each of those cases, the buyer can usually describe what “done” should mean. That matters because the provider can be held accountable to a standard rather than simply continuing to bill while the project evolves.
This model also helps buyers who are specifically trying to reduce management drag. Many service engagements become expensive not because the hourly rate is extreme, but because the client team has to spend too much time clarifying, reviewing, following up, and keeping momentum alive. When the engagement is structured around a result, the provider has a stronger incentive to absorb that execution burden instead of exporting it back to the client.
Agency retainers are stronger in a different set of conditions.
They make more sense when:
- the work is genuinely ongoing and priorities will change frequently
- multiple stakeholders need regular access to a team across strategy, execution, and review
- the company is buying bandwidth as much as output
- the problem is still being defined and the right solution will likely shift over time
For example, a business navigating a larger rebrand, an evolving go-to-market motion, or a complex multi-channel growth program may need a retainer precisely because the questions are not yet stable. In that situation, forcing a narrowly defined outcome too early can create false certainty.
What teams usually get wrong
The first mistake is comparing line items instead of comparing operating models. Buyers often ask whether the retainer cost is higher or lower than the project fee. That matters, but it is not the first question. The first question is whether the model matches the shape of the work. A cheaper mismatch is still a mismatch.
The second mistake is treating flexibility as free. Retainers are often sold as safer because they can adapt. Sometimes that is true. Sometimes it simply means the scope stays vague, decisions stay deferred, and the client keeps paying while the work remains unfinished. Flexibility is valuable only when the underlying need actually requires it.
The third mistake is buying an outcome when the buyer cannot define success. Results as a service depends on clarity. If the team cannot say what deliverable is needed, what constraints matter, and how quality will be judged, an outcome-based structure becomes fragile. The provider may still ship something, but both sides are now exposed to frustration because the target keeps moving.
The fourth mistake is assuming AI eliminates the need for either model. It does not. AI changes production speed, but the buyer still has to decide how accountability works. Someone still owns quality, prioritization, and final usability. Faster drafting does not make a vague service model clearer.
The fifth mistake is overlooking coordination cost. Some engagements look manageable on paper and then quietly consume internal time through status meetings, Slack follow-up, revision loops, and handoff confusion. Buyers should account for that operational cost, not just the invoice.
How to think about it
Use five questions to decide which model fits better.
1. Is the desired outcome specific and testable?
If you can define the output clearly, results as a service becomes much more viable. A useful definition might include the deliverable, the business goal, the constraints, and the review standard.
If you cannot define success beyond “help us improve things,” a retainer is usually safer.
2. How much will priorities change during the engagement?
Frequent change favors retainers. Stable priorities favor outcome-based work.
This is where many teams get misled. They think their problem is defined because they know it is important. But importance and stability are not the same thing. A project can be urgent and still be poorly scoped.
3. Who should carry the execution burden?
If the buyer wants to minimize internal coordination and buy a finished output, results as a service is usually the better fit. If the buyer wants a team to plug into ongoing decision-making week after week, a retainer may make more sense.
That is a practical distinction, not a philosophical one. Some companies need embedded support. Others need a defined problem solved without adding another management layer.
4. What is the cost of getting it wrong?
High-stakes work changes the equation. If the output affects conversion, reporting quality, or operational reliability, the provider needs strong QA and clear ownership. In those cases, the buyer should look less at surface-level speed and more at how the model handles review, acceptance, and correction.
An outcome-based model can work very well here if the acceptance criteria are explicit. A retainer can also work, but only if it does not blur who is actually responsible for shipping a trustworthy result.
5. Is the provider demonstrating capability or selling promise?
This question matters in both models. A provider should be able to show signs that they can execute the kind of work they are proposing to do. That might mean the quality of their own site, the clarity of their process, the specificity of their recommendations, or examples of comparable outcomes.
If the model relies on vague assurances like “we will iterate until it feels right,” the risk goes up, whether the pricing is monthly or project-based.
The practical rule is straightforward:
- buy a retainer when you need adaptable capacity across shifting work
- buy results as a service when you need a defined outcome delivered with accountability
Neither model is inherently more sophisticated. They solve different buying problems.
Takeaway
The better model is the one that matches the shape of the work.
If your priorities are unstable, your stakeholders are still aligning, and you genuinely need ongoing bandwidth, an agency retainer can be appropriate. If you know what needs to be delivered and you want less management overhead between problem and finished output, results as a service is often the stronger choice.
The mistake is not choosing one model over the other. The mistake is buying time when you really need a result, or buying a result when the work is still too undefined to scope honestly.
That is the decision lens that matters: not which model sounds more modern, but which one gives you the clearest path to usable business progress.