There is no sensible universal price for software testing consultancy because the thing being bought can range from a short independent challenge to months of specialist work. Comparing the price without comparing the problem is how organisations end up arguing about day rates while quietly buying the wrong service. Establish what testing service you are actually buying before comparing bids.
The useful question is what is driving the cost, what decision the work is meant to improve and whether the proposed effort is proportionate to that decision. Consultancy pricing gets strange when the proposed team has more role titles than people.
If you want a price for a real testing problem rather than a brochure-shaped package, tell us what is happening and we will scope the work around the answer you actually need.
First decide whether you are buying judgement or capacity
If you already know exactly what needs testing, have usable environments, clear scope and a sensible test model, you may simply need people to execute the work. Paying consultancy rates for routine execution is usually an expensive way to avoid calling a tester a tester.
Consultancy earns its price when the difficult part is diagnosis, challenge or judgement: working out why testing is failing, whether the supplier is good enough, whether release evidence is trustworthy, what capability is missing, or which activity should stop.
The real cost drivers
Complexity matters. A review of one web application is different from a programme with multiple suppliers, legacy platforms, data migration, regulatory constraints and a release that can stop customers getting paid. The amount and quality of evidence also matters: clean accessible evidence is cheaper to assess than six repositories, four dashboards and a folder called FINAL_FINAL2.
Urgency matters too. Asking for an independent view next month allows sensible sampling and access. Asking on Thursday whether Friday’s release is safe means the work has to focus aggressively on what can still change the decision.
Specialist skill can legitimately cost more. Performance engineering, accessibility, security, complex migration or deep automation architecture may require expertise you should not expect from a generalist review.
Choose a commercial model that fits the question
Time-based pricing can make sense where the scope is genuinely uncertain. Fixed-fee work can make sense where the question and boundaries are clear. A short diagnostic followed by an optional implementation phase can be useful when nobody yet knows whether the fix is two days of work or a six-month change.
A cheap day that does nothing is still expensive; an expensive day that kills a six-month mistake can be a bargain.
None of those models is automatically better. What matters is whether you can see what you are paying for, where the engagement can stop and what has to be true before more work is authorised.
Do not pay by the kilogram
Number of consultants, number of workshops, number of pages and number of governance forums are not measures of value. They are measures of things that are easy to invoice.
A five-day review that prevents a pointless six-month test expansion can be excellent value. A cheap twenty-person team executing the wrong regression pack can be very expensive while remaining impressively busy.
How to compare two proposals
Ask both suppliers to state the problem they think they are solving, the evidence they need, who will actually do the work, what decisions their output will support and which parts of the proposal can be removed without damaging the answer.
Also ask what happens if they discover early that the problem is simpler than expected. A consultancy should be able to reduce scope. If every path through discovery somehow leads to the full programme, discovery may be serving the pipeline rather than the client.
A report is not a pricing unit
Written findings can be useful where you need an auditable record, board evidence, supplier challenge or clear priorities. But fifty pages do not justify a higher price than five. The value is in the judgement and evidence, not the amount of toner the recommendation could consume. If removing one workshop causes the whole proposal to collapse, the workshop may have been load-bearing commercially rather than intellectually.
Pay for a clearer decision, a better diagnosis or specialist capability you genuinely need. If you cannot explain what will be clearer when the work ends, the price is not yet the main problem.