When companies buy software testing services, they spend a lot of time on tools, team size and experience. The commercial model often gets less attention, even though it shapes behavior on both sides more than most people expect.
There are two common models: hourly or time-and-materials billing, and fixed engagements tied to scope. Each has a place, but they create very different incentives.
How hourly billing works in practice
Hourly billing looks flexible. You pay for the time you use, scale up when you need more and scale down when things are quiet. For short, unpredictable bursts of work, it can make sense.
Over longer periods, problems appear. The vendor earns more when work takes longer, so there’s little built-in reason to get faster. Reports can grow longer without getting more useful. Test runs get repeated. It’s hard for the client to tell productive hours from busy hours, especially when the work is technical and the client isn’t watching every task.
Budgeting is harder too. Monthly invoices vary, and the finance team has to approve figures that are hard to predict in advance. Every surprise invoice triggers a conversation about whether the hours were justified.
None of this means hourly vendors act in bad faith. Many are honest and efficient. But the model doesn’t reward efficiency, and over time that tends to show.
How fixed engagements work
A fixed engagement ties the price to an agreed scope and team. The vendor commits to covering certain products, platforms and test types, with a defined team and regular reporting. You pay a predictable monthly amount.
The incentives shift. If the vendor finds a faster way to deliver, through better automation or smarter prioritization, that gain doesn’t reduce their revenue. They have every reason to get efficient, and you benefit because the work gets done sooner and better.
Budgeting becomes simpler. Finance knows what QA will cost each month. Engineering leaders can plan around a stable team rather than negotiating hours.
Fixed pricing has its own risk. A vendor who underestimated the work may be tempted to cut corners to protect the margin, testing less deeply or putting junior people on the account. That’s why the scope and reporting terms below matter as much as the price.
What fixed engagements need to work
Fixed models only work with clear boundaries. Without them, both sides end up arguing about what’s included. Before you sign, agree on:
Scope. Which products, platforms, environments and test types are covered? Web only, or mobile too? Functional testing only, or also performance and security?
Change handling. Products evolve. How will new features, new platforms or major redesigns be added to the plan? Is there a process for adjusting scope and price?
Reporting. What reports will you receive, how often and in what format? Who reviews them on your side?
Performance review. How will you measure whether the engagement is working? What happens if targets slip?
Team stability. Will the same people stay on your product? What happens when someone leaves?
The role of a trial period
A trial phase is the best way to calibrate a fixed engagement. Before committing to six or twelve months, both sides work together on real tasks for a few weeks. The vendor learns the product and can estimate effort more accurately. You see the quality of their work and the fit with your team.
At the end of the trial, the scope and price for the longer engagement are based on real experience rather than guesses from a sales call. That reduces the risk of a mismatch later.
A paid trial is better than a free one. When the vendor is paid, they assign the people who would actually work on your account, and you can judge the trial by the same standard you would apply to the full engagement. It also shows how they communicate when something goes wrong, which is hard to learn from a proposal.
When hourly still makes sense
Hourly billing can still be the right choice in some cases:
- A one-off audit or assessment with an unclear scope
- A short burst of extra testing before a major launch
- Very early exploration, before you know what you need
For ongoing QA on a living product, though, a well-defined fixed engagement is usually the better fit.
Some companies combine the two: a fixed engagement for regular testing, plus a small hourly budget for work outside the agreed scope. That can work well, as long as the hourly part stays the exception and both sides agree in advance on what triggers it.
Comparing offers fairly
When you collect proposals, they rarely line up neatly. One vendor quotes a monthly fee, another an hourly rate, a third a mix. To compare them fairly, convert everything into the same picture: what team you get, what they’ll cover, what you’ll receive each month and what it costs over six or twelve months. Add an estimate of your own team’s time spent managing each vendor, since a cheaper option that needs constant supervision may cost more in the end. Once offers are on the same footing, the differences become much easier to judge.
Questions to ask any vendor
When you shortlist a QA outsourcing company, ask them to explain their pricing in one paragraph. If they can’t, the invoices will probably be confusing too. Then ask:
- What exactly is included in the monthly price?
- How do you handle scope changes?
- What reporting do we get?
- Can we start with a trial?
- What happens if we want to end the engagement?
The vendors worth working with tend to be direct about cost. Incisive, for one, states openly that it doesn’t bill by the hour and runs engagements through a paid 30-day trial, then a six-month term, then annual renewal.
Predictability is a feature
In QA, predictability matters. You want a stable team, a stable budget and a stable quality picture you can rely on release after release. The commercial model is a big part of that. Choose one that rewards the vendor for doing the work well, not just for doing it slowly.
