Why is an enterprise logo irrelevant to a forty-person forwarder?

That a large company bought something. It is real evidence, and it is evidence about them.

Think about what sat behind that deployment. A project manager whose whole job was the rollout. An in-house team who built the connection to their own system. A contract with a pilot phase inside it. Contracted rates on most lanes, so the software had something to price from on day one.

Now picture the firm asking the question. Forty people. Nine of them quote. Enquiries arrive as email, WhatsApp and a photograph of a rate sheet. Half the lanes are priced from an agent's reply rather than a contract. The person who would run the rollout also runs pricing, operations and the bank.

The two firms bought the same words on the same website. They did not buy the same thing.

Public rules already say this out loud in other markets. The United States Federal Acquisition Regulation, which governs how the American government buys, requires that an evaluation of past performance consider “the currency and relevance of the information, source of the information, context of the data, and general trends in contractor's performance”. Relevance sits in the same sentence as currency. A reference that is not comparable is not disqualifying, it is simply not evidence about you.

Advertising rules make the related point from the other side. The United States Federal Trade Commission's endorsement guidance says that where an advertiser cannot show that an endorser's experience represents what people will generally achieve, the advertisement must make the generally expected results clear. A case study built on a firm ten times your size is that situation, and nobody puts the disclosure on the slide.

Headcount is the shorthand. It is not the thing that decides whether their experience transfers.

The European Commission's SME definition draws its lines at fewer than ten staff, fewer than fifty and fewer than two hundred and fifty. A forty-person forwarder is a small enterprise on that scale and a rounding error on an enterprise reference. That is a useful sanity check, and it is still only one of six things that have to match before a reference tells you anything.

You will not get all six. Three is a good reference and four is rare. What you are avoiding is the one that matches on none and is offered as though it matches on all.

You have about twenty minutes with somebody doing you a favour. Spend it on the year, not on the demo.

The reference has already seen the product work. So have you. What they have that you cannot buy is a year of ordinary weeks with it.

• 01How long from signing to the first quote that went to a real customer. Not the first login.
• 02What went wrong in month two. Everybody has a month two. If they say nothing did, ask about month six.
• 03How many hours of your own people did it actually take, and in which weeks.
• 04What did you have to clean up before it was useful. Rate cards, port names, customer lists.
• 05What does your pricing desk do now that it did not do a year ago. Their answer to C3, from the other side.
• 06What do you still do outside it, and does that annoy you.
• 07If you were starting again, what would you do differently.
• 08Would you buy it again at the price you are paying. Wait for the pause.

Then one more, and it is the one the checklist keeps in reserve as the dodge follow-up: has anyone stopped using this, and could I speak to them. Ask it of the vendor, not of the reference. A vendor who names a firm that left, and says why, has told you more about themselves than any reference call will.

Two rules for the call itself. No vendor on the line, in any form, including a recording. And ask for the reference's own numbers rather than the vendor's: enquiries a week before and after, in their words, however rough.

None of the four is dishonest. All four leave you with less than you think.

Trap one. The hand-picked champion

Every vendor has one customer who loves them, usually the earliest, often on a discount. That call is pleasant and worthless on its own. The fix is not to refuse it. Ask for a second name after you have taken the first, and notice how long the second takes to arrive.

Trap two. The logo that is really a pilot

A famous name on the slide can be one office, one lane, three people and a trial that ended in March. Ask three questions of any logo: which office, how many people use it today, and since when. A vendor who cannot answer those about their own showcase account is telling you something.

Trap three. Confidentiality as a wall

“Our clients prefer to stay confidential” is sometimes true and always testable. Ask them to go and ask one. A vendor with happy customers your size comes back in a week with a name and a time. A vendor without them comes back with another case study.

Trap four. The reference who is not only a customer

Ask directly whether the reference is an investor, a reseller, a partner or a paid advisor. It is a normal question and the answer is usually no. When it is yes the call is still worth having. You read it as a partner's view rather than a peer's view of their own year.

Three lines, the same day, in the reference's own words. You will remember whether you liked them and none of the substance.

Then use it. D1 is one of two questions on the sheet that is easy to verify and hard to fake, and both are wasted if you take the offer and never make the call. A named introduction you did not follow up is worth exactly as much as the logo wall.

Then say so, and change what you accept. Do not change the standard.

A company with six live forwarders cannot produce a comparable reference for every kind of buyer. Insisting anyway turns a good question into a rule that says buy from the largest supplier.

The same procurement rule that demands relevance says what to do about it: an offeror without a record of relevant past performance “may not be evaluated favorably or unfavorably on past performance”. Neutral, not disqualified. That is the right posture for a young vendor, and it puts the weight back on the questions that need no customer base: A1, A3 and D2.

What to accept instead, in order of usefulness: a firm your size in a different market, a firm in your market at twice your size, a trial that ended without a sale where the vendor will tell you why, or the vendor's own operations run on their own product. What not to accept from anybody is a logo offered as though it answered the question.

And for completeness, since the checklist is meant to be used on us too, here is our own answer in the form this page asks for. Infinity Logistics quoted 70% more shipments with the same five-person team, comparing Q4 2024 with Q4 2025. That is one firm, one market and a five-person pricing desk. If it is close enough to your shape to matter, the right response is not to believe it. It is to ask us for the phone call.