Most discovery work in this market is sold as the first phase of a build. It is priced accordingly, sometimes discounted or waived against the build that follows, and the assessing firm is the firm that expects to do the work. That arrangement is normal, it is not dishonest, and everyone involved understands it. But it has a structural consequence worth being direct about: an assessment run by the party that will build the outcome has one ending it is paid to reach, and the client cannot easily tell whether the recommendation survived that pressure or was shaped by it.
A standalone advisory engagement is the same work with the commercial link removed. Fixed price, fixed scope, a documented output the client keeps, and no build attached to it. The difference is not the method. It is that four endings are available instead of one, and three of them are worth less to the assessing firm than the one everybody expects.
Our stake here is unusually visible and worth stating up front. BY BANKS both runs paid assessments and builds software, which is precisely the arrangement this article says creates a pressure. We are writing it because we are moving toward offering the assessment as a product in its own right rather than only as a route into a build, and an article that sets out what that means, including what it costs us, is a more useful thing to publish than one that only describes the upside. Readers should weigh it knowing we sell both.
The audience is anyone about to commission a discovery, an operations assessment, or a technology review in the UAE: COOs, CIOs, finance directors signing the budget, and founders about to spend a large share of a year's investment on a platform. The useful question is not whether the assessing firm is any good. It is which of the four endings below they are able to reach, and what happens to their commercials in each.
The Four Endings, and What Each Is Worth
Below are the four ways a genuine assessment can end. For each: what it looks like, what the client actually gets, and what the ending costs the firm doing the assessing. Tap any ending for the detail. Only one of the four leads to a build.
Four endings a genuine assessment can reach
Tap any ending for what the client gets and what it costs the assessor
Why the Commercial Link Changes the Output
The mechanism is not that firms lie. It is subtler and more ordinary than that. When the assessment is the front end of a build, ambiguity resolves toward the build. A workflow that is 70% standard gets described as distinctive. A product that would cover most of the requirement gets assessed against the 20% it misses rather than the 80% it covers. A prerequisite that is not in place gets phased around instead of waited for. None of those is a false statement. Each is a judgement call, and the judgement calls accumulate in one direction.
Remove the link and the calls stop leaning. That is the entire argument for the standalone version, and it is also why the standalone version has to be priced as a real engagement rather than a loss leader. An assessment that is free or heavily discounted against a subsequent build has the commercial link in it whether or not it is named in the contract.
There is a second effect, which is about what gets written down. When the assessing firm expects to build, the output can be thinner, because the detail lives in the heads of people who will still be involved. When they expect to hand it over and leave, the output has to stand on its own: the assumptions have to be explicit, the systems audit has to be legible to someone else, and the costing has to be defensible to a third party who might be the one delivering it. That produces a better document, and clients notice the difference.
The distinction in one observation
An assessment tied to a build has one ending it is paid to reach. A standalone assessment has four, three of which return less money than the fourth. The value of the standalone version is not that the people are more honest. It is that the structure stops rewarding a particular answer, and the output has to be written for someone other than the person who wrote it.
What the Output Actually Contains
A map of how the operation actually runs
Not the process as documented, the process as performed, including the spreadsheets, the WhatsApp threads, and the person who knows the exception. This is routinely the part clients say they got most from, because it is visible to the whole leadership team for the first time.
An audit of what the current systems will and will not expose
What the existing stack can be integrated with, at what cost, and where the data is in a condition that supports what is being proposed. This determines more about feasibility than the requirements do, and it is the part most often assumed rather than checked.
Costed routes, with the assumptions named
Usually more than one, with what each assumes and what would have to be true for it to hold. A single recommendation with a single number is a weaker artefact than two routes with their assumptions exposed, because the second lets the client argue with the reasoning rather than only the conclusion.
A document that works without the authors
Written to be handed to a board, a procurement team, or a different delivery firm entirely. If the output only makes sense with the people who produced it in the room, it is a sales artefact rather than an assessment.
Where a Standalone Engagement Is the Wrong Choice
It is not always the right answer, and pretending otherwise would be the same failure this article is about.
If the decision is already made and correct, an assessment is a delay with a fee attached. Some organisations know exactly what they need, have the internal capability to specify it, and should go straight to procurement. Paying for a document that confirms a good decision is a governance comfort, not a business outcome, and it is worth being honest about which one is being bought.
If the problem is small and bounded, the cost of assessing it can approach the cost of solving it. Below a certain size, the sensible move is to build the thing, get it wrong somewhere, and fix it, which is cheaper than de-risking it in advance.
And if the client genuinely wants a partner rather than an opinion, splitting assessment from delivery adds a seam for no benefit. The value of the standalone version is independence. If independence is not what is being bought, the commercial link is not a problem, it is the point.
How This Sits With BY BANKS, Honestly
The conflict is direct and we are not going to talk around it. BY BANKS runs paid assessments and also builds software. The article above argues that an assessment run by the firm that will build has a pressure on it, and that describes us in every engagement where the client goes on to commission a build with us, which is most of them. Naming a pressure does not remove it, and no reader should conclude that our assessments are immune to it because we have written about it.
What we can say concretely is what it costs when we reach the other endings. When an assessment concludes that a product covers the requirement, we return a shortlist and integration costs and forgo the build, which is the larger number by a wide margin. When it concludes the problem is process rather than software, we say so and the engagement ends there. When it concludes the timing is wrong, we date the plan and wait, sometimes for a year, sometimes permanently. Each of those has happened, each costs us the build, and a firm that has never reached those conclusions is not running assessments.
We are also aware of the obvious counter: a firm can reach the honest conclusion occasionally and still lean toward the build in the ambiguous cases, and no article demonstrates otherwise. The only real protections available to a client are structural. Pay for the assessment properly so it is not subsidised by the build. Ask, before it starts, what the assessing firm would have to find in order to recommend against building. And keep the output in a form you could hand to somebody else, because the ability to do that is what makes the independence real rather than asserted.
This article describes engagement structures in general terms and is an observational framing drawn from delivery experience rather than a methodology, a certified framework, or specific advice. No specific client, engagement, advisory firm, consultancy, product, or competing supplier is described or implied, and no outcome described should be read as a representation about any particular engagement or as a guarantee of any result. The four endings are a simplification; real assessments frequently produce mixed or partial conclusions, and any assessment depends on the specific business, sector, systems, data condition, and starting position. Commercial terms, scope, duration, and pricing for any BY BANKS engagement are set out in the relevant proposal and contract and are not described here. BY BANKS is an independent software engineering company based in the UAE. We design and build software and hand it over. We are not a management consultancy, an auditor, or a regulated entity in any sector we serve, and we do not provide strategy, audit, tax, legal, recruitment, staffing, immigration, investment, or regulatory advisory services. On any engagement, the buyer owns its commercial, technology selection, regulatory, and compliance decisions and the responsibility for their implications. This article is not procurement, investment, governance, or legal advice; organisations should obtain qualified advice for their specific circumstances. Public sources used in our published work are listed on our Sources and Data page.
Ready to Build Something?
If this resonated, let's talk about how we can apply these ideas to your business.
Start a Conversation