Do You Need a Distributor, or a Partner?

The Request Arrives Early

It usually comes in the first meeting, or the second.

The conversation has gone well. They know the market, they know the customers you have spent a year trying to reach, and they have views on your pricing that are better informed than anything in your own plan. Then, before anything has been sold and often before anyone in the room has seen the product work, comes the ask. We would want exclusivity for the region.

Every vendor treats that as a question about how much to give away. It is not, and it cannot be answered yet, because it presupposes something nobody in the room has said out loud.

What are you asking this company to do?

There are two answers and they are not points on a spectrum.

A distributor is your office in the assigned territory. Stock. Credit into the channel. Logistics. Local invoicing. First-line support. Coverage of markets you cannot reach yourself. Local marketing. Demand generation. Those are the functions of a branch you have decided not to open.

A partner or system integrator is the last mile. They take the product to the end user, design it into a project, install it, and hold the client relationship on the ground.

One of those is singular by nature. You do not open two offices in one country. The other is plural by nature, and you want as many of them as you can properly support.

What makes this more than a taxonomy is what happens when the layers get confused. Grant office-level exclusivity to what is really a last-mile player and you have not made a generous decision, you have made an unworkable one. Other integrators will not approach a competing integrator for a price — not out of pride, but because asking exposes the deal they are working on. So they do not ask. They propose something else, and you never learn the opportunity existed.

That is the whole failure, and it is invisible while it happens. Nothing arrives to tell you about it. There is no lost bid, no complaint, no line in a report. There is simply a market in which your product stops being proposed, and a partner whose numbers look reasonable because the deals you never heard about were never in anybody's forecast.

The third part of that sequence describes appointing distribution in exactly these terms without using them: we were looking to outsource the model itself — sales, pre-sales, service, and if it could be made to work, customer success. That is an office-layer ask, and naming it as one changes what you should have expected from it.

One Office, Many Last Miles

Take the singular half first, because it is arithmetic and it is not negotiable.

If your category is a push — new, unfamiliar, dependent on demand being created rather than serviced, requiring investment in stock and people and training before a single order arrives — one serious party is the answer. The entire model depends on somebody doing work across a long sales cycle that will not happen on its own. Split that across two houses and neither can justify the dedicated person the work requires. You have not bought competition. You have bought two disengaged parties instead of one committed one.

Then what two sources into one geography actually do, which is the part vendors discover rather than predict. The buyer plays them against each other. Absent transparency it becomes a price war, margin erodes on both sides, and both parties lose interest in a line that no longer pays them. Nobody has behaved badly. It is the structure, and it is entirely foreseeable.

Ask how large a category has to be before two distributors can each justify a dedicated person and a consultant will give you a number. I will not, because the number is not what decides it. The role is. When demand already exists and the product moves largely by itself, the office function genuinely shrinks and the question reopens — but that is a change in what you are asking somebody to do, which is this article's question arriving a second time rather than a different question.

The plural half is the mirror image, and it is not a licence for chaos.

How many integrators you want in one market is set by geography and by segment. Map who plays where, work out which bases have to be covered, and appoint against that map rather than against whoever turns up. Then police it, because conflict at the last mile is normal and permanent. Deal registration is the first mechanism: the integrator who brings a lead and develops it with you is the one who is rewarded for it, and that has to be true in practice and not only in the partner agreement.

The second mechanism is the one vendors undervalue. Certification, and real installation and commissioning capability. An integrator who has done your work before can price the services in a bid off their own numbers, rather than loading in a vendor-supplied professional services line. That difference is frequently what makes or breaks their deal, which means your certified partners are structurally cheaper than your uncertified ones. Certification is not a training programme. It is a pricing advantage you are handing to the people you want to win.

Policing it is work, and the worst version I have handled was a mega-project where an integrator bid deliberately below everybody else, assuming that once the award was in hand the vendor could be pushed down far enough to make the economics work. A young vendor chasing a reference will sometimes comply. What saved it was that the end user had been part of the budgeting and knew the number was impossible: they asked for financial guarantees the bidder could not produce, then declined the award so the genuine bidders could compete. It cost three months and it cost everybody margin. The project was delivered and it remains one of our best references.

What Your Buying Process Decides

Two questions convert all of that into a structure, and the answers are facts about your business rather than preferences.

How is your product actually bought?

In spec-led project business the work happens long before a tender exists. Somebody has to reach the consultant, the end user, the main contractor and the integrator, and the vendor needs information from all four and a champion inside at least two of them.

A value-added distributor helps here, and it is worth being exact about how. They carry multiple lines, so they see a far larger pipeline than any single integrator does. They run a pre-sales team whose job is to study a project's requirements and propose a complete bill of materials at design stage, which the consultant then vets against budget and specification to produce the final BOM that goes out to tender. Where the consultant has already issued an itemised BOM, the same team proposes value-engineered alternatives with the budget and specification impact spelled out. Their lock on a project is a function of how many of their lines they can get into it.

Which cuts both ways, and vendors forget the second edge. The same process can value-engineer your line out — to bring a system inside a lower budget, or to make room for a component somebody would rather sell instead. A distributor with fifteen lines in a project is not a neutral party on which of them survives.

In transactional volume business the calculation is different and the contribution is real: stock holding, logistics, local invoicing, warranty float, and credit into the channel.

Credit is the one vendors underweight. An integrator depends on you for one project; a distributor deals with you across many, which makes their exposure — and therefore your leverage — considerably larger, because a default on a small case can jeopardise delivery on a mega project they cannot afford to lose. Their scale cuts the same way: a default with any one vendor flows straight into their credit rating and their insurers and affects their standing everywhere, which is not true of an integrator. And then the timing difference, which never appears in anybody's model. Project payments are milestone-based and tied to progress. Distributor terms are time-bound. Appointing a distributor transfers your exposure to project delay onto somebody else, and in this region that is not a detail.

Do you need one at all?

Sometimes the answer is no, and it is not a failure. If you are small enough that no distributor worth having wants to take you on, you are better off with none than with one who will not work for you. If you sell software with no physical goods movement, the integrators already handle everything a distributor would have handled. And at entry into a push category, direct to integrators may simply be the correct structure.

That is what we did at Prysm. The category was project-based, nothing needed stocking because units shipped custom against each project, there was no history and no business case for a layer, and the order-to-install cycle ran long enough that we could service demand inside six to twelve weeks. We were creating a market brick by brick rather than servicing one that existed.

It worked, and the reason it worked is the argument nobody makes for going direct. We were on top of every opportunity — objections handled in the room, pricing on the spot, technical support without a hand-off. And nobody in the chain had the option to win while we lost. A distributor can win a deal by substituting your line for another one in their portfolio, and letting your technology slide out of a project costs them very little when they have other lines going into the same job. Direct to integrators, that option does not exist.

The structure changed when the product did. Gen2 was a standard box that could be stocked, in a category that had turned from push to pull, where demand needed servicing quickly and in-region. That is the moment a distributor became the right answer. The moment it becomes the right answer for you will be a change in how your product is bought — not a change in how much reach you wish you had.

When the Company in Front of You Is Both

Here is the complication the model does not survive contact with. Many of the strongest houses in this region are both. They carry lines and they integrate. The company asking to be your distributor on Tuesday is bidding a project as an integrator on Wednesday, and both of those are real businesses with real capability.

Which is fine, and workable, and it has exactly one rule attached.

Whoever holds your office layer does not compete for new business against the people they are supposed to be supplying. Not as a matter of etiquette — as a matter of viability, for the reason in section one. The moment your distributor is a bidder, every other integrator in the market quietly stops pricing your product, because getting a number out of them means showing a competitor the deal. You will not be told. You will simply stop being proposed.

Service is the exception, and it is worth granting deliberately rather than leaving ambiguous. We had a distributor bid for and hold service contracts directly and it suited everybody, because service is not what the channel is fighting over. New sales are.

So the question to ask a house that does both is not whether they can keep the two businesses separate in principle. It is narrower and more answerable. Which of my two layers are you, in writing, for new business? Where does your integration arm sit in relation to my other partners? And what happens the first time your own bid team wants to quote my product into a project one of my certified integrators has been developing for eight months?

That last one has a right answer, and it is a deal registration record with a date on it.

What I Got Wrong

We changed the layer and never changed the motion.

Moving to distribution at Gen2 was the right call on the facts — a stockable box, a category turning from push to pull, demand that needed servicing in-region faster than we could service it ourselves. What I did not do was rebuild the selling motion underneath the new structure. The whole point of going direct at entry had been that we were in every opportunity ourselves: objections handled in the room, price on the spot, nobody in the chain holding an option to win while we lost. Handing the office layer to somebody else removed all three of those, and we did not replace them with anything.

That article already carries the sentence, so I will not retell it: we never ran the motion we had restructured to run. What is worth adding here is why it happened, because it was not oversight. The bandwidth was not there, and the structural change was easier to execute than the behavioural one — a contract is signed in a quarter, and a selling motion is rebuilt over two years by people who are also carrying a number. Both of those are true. It was still the thing we should have done, and the cost was that we bought a layer and kept selling as though we had not.

There is a related admission I can only make partially. The business we were in could have run direct to integrators here, as it did in the United States and Europe. Managing local logistics, credit and service support across several Gulf countries made distribution the more cost-effective structure for us, and I still think that was right. But I never tested it. We reasoned our way to the answer from cost, at a point when we had two years of evidence that the direct model worked in this market, and I never went back and asked the question properly with the evidence in front of me.

And the layer question I have never had to answer in anger is the one in the section above. I have never appointed a house that was both and then watched the two arms collide, because we excluded named global integrators at the outset and the situation did not arise. So read that section as designed rather than survived.

If You Are About to Appoint

The request that opened this article is still sitting there unanswered, and it should be, because it was asked in the wrong order. Decide what you are buying and how much of it your market can carry, and the question of what to grant becomes a much smaller conversation — one with a defined shape and a known price, rather than an argument about trust conducted over an eighteen-month misunderstanding.

Connektions MEA works with vendors on channel structure before it is signed — the layer, the territory, and what the agreement has to say. That work is described under Entry Blueprint.