Which vendors deserve your practice

Before scoring anything, settle one question, because getting it wrong produces confident nonsense rather than a rough answer.

Does your product flow through this partner toward the customer? If it comes from them to you, they are upstream and this guide applies. If it goes from you through them to their customer, they are downstream and the other guide applies. If nothing passes through them at all and they refer or co-sell while the customer buys from you directly, they are lateral.

It is a property of the relationship, not of the company. The same MSP can be downstream when they buy and bundle your product, and lateral when they refer a client who buys from you directly. A vendor you carry in one product line can be a co-sell peer in another.

This matters because scoring questions do not survive being pointed the wrong way. Ask a distributor you buy from about their install penetration of your product and you get a meaningless answer, and it will not come back as an error. It will come back as a low score on a world-class supplier. Direction has to be stated, never inferred.

You are not managing a partner list. You are allocating a practice.

Every vendor line you carry consumes something genuinely scarce: certifications, engineer time, sales enablement, a slot in your own pitch, and the attention of whoever owns the relationship. Agencies over-commit to too many platform partnerships. MSPs agonise over which stack to standardise on. Resellers pick lines and live with them for years.

And the stakes are asymmetric compared with the vendor side. A vendor scoring an ecosystem of a hundred partners can afford to be wrong about several. A twelve-person agency betting its practice on the wrong platform has made one decision that shapes three years of hiring.

The obvious ranking is by margin, or by whichever brand is biggest. Both fail, in opposite directions.

Margin is a term, not a property. It is negotiable, it changes at renewal, and it tells you nothing about whether you can deliver the thing, whether the vendor will send you work, or whether they will compete with you in two years. A generous discount on a line you cannot staff is not a good deal.

Biggest brand fails the other way, and it is the more expensive mistake because it feels prudent. Which brings us to the trap that catches this side of the table specifically.

Score each vendor or platform on six things, twice: what is demonstrated today, and the realistic ceiling if the relationship were properly worked. Mark N/A where a dimension genuinely does not apply.

Discovery is often N/A for a self-serve platform, because there is nobody to have the conversation with. No briefings, no deal intel, no human who knows your market. That is not a failing grade, it is the product working as designed.

Amplification is frequently N/A for infrastructure you build on but never mention to a client. Nobody chose you because of your hosting provider.

Implementation is N/A only for a pure product line with no delivery angle at all, which is rarer than it sounds.

Scoring any of those as a 1 would rank an entire category of perfectly good suppliers last for no reason. N/A is an answer, not a zero.

The vendor-side guide warns about chasing the impressive partner who never activates. Up here it inverts, and it stings more, because now you are the small one.

You can produce a beautiful score for a major platform, build a full plan around becoming a serious partner, and get nowhere. Not because the platform is bad. Because to them you are a rounding error. Your certifications do not move their numbers, your deal volume does not register, and your account manager has four hundred other partners.

So run the honest second assessment alongside the score: what do you look like to them? Your growth trajectory, the clients whose names would actually impress their sellers, the volume you can commit, the bandwidth you have to support a serious relationship. A high score combined with low attractiveness does not produce a partnership. It produces a pursuit you want far more than they do.

The useful version of this question is the one a good vendor is silently asking anyway: why should they prioritise you over the other four hundred partners asking for time? If you cannot answer it, the score is not wrong, but it is not actionable either.

This is the risk that has no real equivalent on the vendor side, and it is the one this audience raises first.

A supplier can go direct. A platform can ship the service you built a practice on. A vendor can acquire, or be acquired by, someone who competes with you. It rarely happens in month one, when the relationship is shallow and walking away is free. It happens once you have certified six engineers and built a delivery methodology around them.

Which means risk on its own is not the number you want. Depth is the multiplier:

Exposure = risk x depth.

A high-risk vendor you barely transact with is noise. A moderate-risk vendor you are deeply wired into is the expensive surprise, and it will never feel urgent, because moderate risk does not, right up until it does.

The method already measures depth. That is what Implementation and Utilization are. So the alarm is not who looks threatening in a meeting. It is who scores high on I and U while showing real signs of moving into your space.

Hold a discipline about evidence, though, or this becomes paranoia. Almost every good supplier is adjacent to you; that is why the relationship works. Adjacency is not competition. Substitution is. The question is not whether they are near your space, it is what they actually shipped, announced, hired for or acquired, and specifically why a client would choose it instead of you.

Take your current lines plus the ones you are considering. Six dimensions, scored twice, N/A where it does not apply, one line of reasoning each.

Two things to fix before you start. Write down which direction each relationship runs, because half your list will be upstream and some of it will be lateral, and mixing them silently is how you get a false answer. And be honest about your own position, since the attractiveness question in section 6 changes what a high score is worth.

Keep commercial terms in a separate column. Margin, rebates, MDF amounts and deal-registration protection all matter enormously to the decision. They are just not measures of the relationship, and mixing them into the score is how the best discount ends up ranked first.

The harder problem is the one after the first pass: keeping it honest as tiers change, competitors emerge, and a supplier you trusted starts drifting toward your space. A ranking built once and left alone is worse than useless, because it still looks current. That needs real usage data, not a memory of last year's QBR, and it needs the same exposure check, risk multiplied by depth, run again every time a vendor's status changes rather than once at signing.

A defensible answer to which lines get investment next year, in the same terms for every vendor, which is the argument you need when somebody senior has a favourite.

A shortlist of relationships whose ceiling sits far above their current contribution. Upstream that usually means a tier you have not pushed for, enablement you have not taken, or demand the vendor routes that you have never asked to receive.

And a much clearer view of concentration risk. Once depth is scored rather than felt, the question of whether you are dangerously reliant on one supplier stops being a worry and becomes a number you can act on.

The same six dimensions work in a third configuration, for co-sell peers who share your customers without anything passing through either of you. And they work outside technology entirely, which is where the next guides go.

For the framework itself, see RADIUS, and for the today-versus-potential mechanic, the activation gap.