A pattern worth naming: in the UAE, a large share of advisory and transformation engagements now bring an outside engineering partner into the work rather than staffing delivery from the advisory firm's own bench. It happens on government programmes, on bank transformations, on contractor and healthcare group modernisations, and on the mid-market work underneath all of it. It is rarely announced, because the advisory firm is the prime and the specialist is a subcontractor, and the client mostly sees one engagement. But the model is common enough that it is worth understanding from the client's side, because the seams between the two firms are where engagements either hold together or quietly drift.

We have an obvious stake in this. BY BANKS is frequently the specialist partner in that arrangement, and an article explaining why the model exists is an article that argues for the thing we get paid to do. We are writing it anyway, for two reasons. The first is that the alternative version, in which advisory firms should simply build the capability internally, is one we think is wrong on the economics and would say so even if it cost us work. The second is that the failure modes below are ones we have been on the wrong side of, and a client who understands the seams manages them better, which produces engagements that go well for everyone including us.

The audience here is anyone commissioning advisory work in the UAE where the recommendation is going to become software: transformation leads, CIOs, programme directors, procurement teams, and the partners at advisory firms deciding whether to bring a specialist in at the start or at the handover. The useful question is not whether a specialist is involved. It usually is. The question is which of the four seams below your engagement is exposed at, and whether anyone has been made accountable for it.

The Four Seams, Read Honestly

Below are the four points at which advisory engagements in this market reach for a specialist engineering partner. For each, what the advisory firm owns, why a specialist tends to be brought in, and the specific way it goes wrong when the seam is not managed. Tap any seam for the detail. These are observations from delivery, not survey findings, and they are patterns rather than rules.

Four seams where advisory work reaches for an engineering specialist

Tap any seam for what each side owns and where it breaks

The four seams are an observational pattern drawn from delivery experience on UAE engagements, not a survey finding, a methodology, or a certified framework. They describe tendencies rather than rules, no specific client, advisory firm, or engagement is described or implied, and any real engagement will differ. This is not procurement, contracting, or legal advice.

Why the Model Exists at All

The straightforward reading is that advisory firms subcontract engineering because it is cheaper than employing engineers. That is part of it, but it is not the interesting part, and on its own it does not explain why the model persists at firms that could easily afford the headcount.

The more accurate explanation is about utilisation shape. Advisory work is intermittent and diagnostic: a team of five for eight weeks, then nothing on that account for a quarter. Engineering work on the same account is continuous and lumpy in a different rhythm: nothing for eight weeks, then twelve months of sustained build. A firm optimised for the first shape cannot hold a bench for the second without either carrying idle engineers or, more commonly, putting whoever is available onto whatever has landed. That second outcome is the one that damages engagements, because the person costing a core banking integration is the person who was free that week.

There is a second driver, which is that the specialist knowledge in this market is narrow and local. Knowing what UAE PASS service provider onboarding actually involves, or what a Mirsal 2 integration exposes, or how long MoIAT ICV verification takes in practice, is knowledge accumulated by having built against those systems and waited out their processes. It does not generalise from having built something similar elsewhere, and it goes stale if you stop. A firm that builds against those surfaces continuously has it. A firm that does so occasionally does not, regardless of how good its engineers are.

The market is moving in this direction visibly. Khaleej Times reported in July 2026 that UAE organisations are increasingly buying specialist expertise in areas like AI, cybersecurity and automation rather than general IT support, and that proven implementation experience has become a primary procurement factor rather than a tiebreaker. The same logic that pushes a client toward a specialist pushes an advisory firm toward one.

The model in one observation

Advisory firms do not bring specialists in because they cannot write software. They bring them in because the shape of advisory demand and the shape of engineering demand do not fit on the same bench, and because the local integration knowledge that decides UAE build timelines is only held by firms building against those systems continuously. The seams between the two are managed well or badly, and that, rather than the existence of the arrangement, is what decides the engagement.

What the Prime Is Actually Buying

An estimate that survives the build

The number attached to a recommendation is the number the client budgets against and defends internally. A benchmark-derived figure is a guess dressed as an analysis. A figure produced by reading the client's actual integration surface and data condition is a different artefact, and the difference shows up as variance twelve months later rather than at approval.

Local integration knowledge that is current

UAE PASS, Mirsal 2, eClaimLink, MoIAT ICV verification, CBUAE reporting surfaces and the rest each carry their own onboarding process and lead time. That knowledge decays when you stop using it. What the prime buys is not the ability to integrate, it is a current answer to how long each one takes.

Delivery risk moved to a firm that carries it

An advisory firm that recommends a build and does not deliver it carries reputational risk but not delivery risk. Bringing in a partner whose commercial exposure is tied to the build completing puts the risk with the party best placed to manage it. That is a governance decision as much as a commercial one.

Continuity across the recommendation-to-build gap

The most expensive point in these engagements is the handover, where a target state written by one team is inherited by another. A specialist involved during the advisory phase carries its own assumptions into the build rather than inheriting someone else's, which removes an entire class of dispute before it starts.

Where It Goes Wrong

The most common failure is that the specialist is brought in too late. An advisory engagement runs to conclusion, produces a target operating model and a costed roadmap, and only then goes to market for a build partner. Every assumption in that document was made without an engineer in the room, and the build partner's first job becomes reopening the ones that do not hold. That reads to the client as the build partner being difficult, when it is the build partner doing the estimating work that should have happened during the advisory phase.

The second failure is the specialist being hidden. There are legitimate commercial reasons a prime does not want its subcontractor in front of the client, but a client who does not know who is building its platform cannot assess whether that firm is any good, cannot judge the integration claims being made on its behalf, and has no relationship with the people who will still be there in year two. Our own view is that the arrangement works better when it is visible, and we are aware that this view is convenient for us.

The third failure is scope handled at the seam rather than at the top. When a constraint surfaces mid-build, the question of whether it is a change or an omission gets resolved between the prime's engagement manager and the specialist's lead, both of whom have commercial reasons to prefer a particular answer. If the client has no visibility into that conversation, it finds out about the outcome and not the decision.

What Good Looks Like From the Client Side

Ask who is doing the engineering before the recommendation is written, not after. If the answer is that a specialist will be appointed later, the cost estimate in the report is a benchmark, and it is worth treating it as one.

Ask what the specialist has built against specifically. Not sectors, systems. A firm that has completed UAE PASS onboarding can tell you what the review stage involves and roughly how long it took. A firm that has not will describe the capability in general terms. The difference is audible in about ninety seconds.

Ask where the assumptions live. A target operating model that does not record which of its assumptions were tested and which were taken on trust is a document that will be argued over. One that flags them is one the build can start from honestly.

And ask what happens at the handover. Not whether there is one, but who from the advisory phase remains available when the first hard constraint is found. In practice that single question predicts more about how an engagement will go than most of the procurement criteria that sit above it.

How This Sits With BY BANKS, Honestly

The commercial stake is direct and worth stating plainly. BY BANKS is often the specialist engineering partner in exactly the arrangement described above, both as a subcontractor to advisory and consulting firms and directly to clients. An article arguing that advisory engagements are better with an engineering specialist involved early is an article arguing for work we are paid to do, and readers should weigh it accordingly.

We are making the argument anyway because we think the alternative is worse for clients, and because the failure modes above are ones we have been on the wrong side of rather than ones we have only observed. We also accept the parts of this that cost us. Some engagements genuinely do not need a specialist: where the recommendation is to buy a configured product, where the integration surface is small and well-documented, or where the client has a capable internal engineering function that has been building against the same systems. In those cases bringing in a specialist adds a margin and a coordination cost for no corresponding reduction in risk, and saying so loses us work we would otherwise be considered for.

The boundary stays clear. 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, we do not provide strategy, audit, tax, legal, recruitment, staffing, or regulatory advisory services, and we are not a regulated entity in any sector we serve. On any engagement, whether we are engaged directly or through a prime, the buyer owns its commercial, technology selection, regulatory, and compliance decisions and the responsibility for their implications.

Nothing in this article describes any specific client, advisory firm, consultancy, or engagement, and no arrangement, dispute, or failure described here should be read as referring to any identifiable party. The four seams and the failure modes are an observational pattern drawn from delivery experience, not a survey finding, a methodology, a certified framework, or specific advice, and any real engagement will differ by client, sector, scope, and contracting structure. The reference to UAE organisations increasingly seeking specialist technology partners is drawn from Khaleej Times reporting published in July 2026, as published. Integration systems named in this piece, including UAE PASS, Mirsal 2, eClaimLink, MoIAT ICV verification and CBUAE reporting surfaces, are referenced descriptively; BY BANKS is not affiliated with, accredited by, or endorsed by TDRA, Digital Dubai, Dubai Customs, the Dubai Health Authority, MoIAT, the Central Bank of the UAE, Khaleej Times, or any other authority, publication, or organisation referenced in this article. Onboarding processes and lead times for those systems change; readers should rely on the issuing authority for current requirements. This article is not procurement, contracting, governance, or legal advice, and organisations should obtain qualified advice for their specific circumstances. Public sources used in this piece are listed on our Sources and Data page.