Every warehouse system purchase is made by someone doing it for the first or second time, across a table from someone who does it every day. That gap decides more outcomes than the feature comparison does.
Bill Carlin framed it better than I had on episode 210 of Disruptive Minds in May 2025. Buying a WMS is like buying a car, except the dealer has been to the lot eight times today and you have been twice in your life.
Every quote block below is me, on the episode, condensed for reading.
Asymmetry of Information
Most of the folks in there either haven't done it, or they might have had one at bat, maybe at a different business, maybe a completely different industry. So the experience doesn't necessarily translate. And the vendors do this every day. The asymmetry starts before the demos, in how the list of vendors got assembled in the first place.
The asymmetry doesn't stop at the sales team, and this is the part buyers get wrong in the other direction:
What I've seen a lot on the vendor side is they know the software. They don't necessarily know how the software is used operationally. They haven't worked in a fulfillment center. They haven't worked for a 3PL. They know how the software works, and bridging that gap is a hard one to do.
The vendor is answering a narrower question than the one you asked, with no bad faith involved. You want to know whether this runs your building. They can tell you what the software does.
Put-Away Is the Tell
If you get one technical question to size up a system, this is the one I'd spend it on:
A lot of the lower-end warehouse management systems operate on the concept of you tell us where the SKU goes and we have a static assignment. That's one end. Then you have the other end of the spectrum where there's a lot of logic that can do things on the fly and really adapt with you. Usually that's a pretty good proxy for the sophistication and adaptability of the system. You'll see pretty quickly just how mature they are, or what their ambitions are.
Put-away is where the system either makes a decision or asks you to have made one. Everything downstream inherits that. Rules at receiving that account for order profiles make picking easier for the rest of the item's life in the building, and a static bin map means every efficiency you want later has to be bolted on somewhere else.
Ask what happens at put-away and you learn what the vendor thinks a warehouse system is for. No feature matrix will tell you that.
The trap is that this also demos beautifully, and I've watched it close deals. The decision criteria it depends on usually live in one person's head, with nothing in any system for the software to read.
The Two Groups Nobody Budgets For
Bill raised the downstream effect people skip: a 3PL changing systems is also changing its clients' experience, and its employees' jobs. The second one has a cost that shows up on a report:
Moving to a system can be a massive change. Now there are all these verifications that weren't there before. The person's like, why did this change? Especially if they're being compensated or rewarded based on number of packages per hour. Whoa, you just crushed me. Why? So maybe there's a compensation model that needs to change too.
Bill had lived it. A packer going from roughly forty orders an hour to twenty-five after an upgrade, with the drop landing on her performance numbers and her pay, while the extra checks she was now performing showed up nowhere. That is the reporting fault line in miniature, where each function optimizes the number it can see.
Add verification steps and leave the productivity metric where it was, and you have cut somebody's pay for following the new process. The floor learns from that.
The offsetting argument is real and worth making to the floor rather than about them. One person packing alone gets very fast and very accurate. Add a second and the error rate climbs, and mispicks cost a customer relationship rather than a minute.
The Person Selling You Is Not the Person Supporting You
Not every vendor is the same. The maturity of the vendor plays a big part. Support is another component of that relationship. If they're more focused on your type of business, they might be more helpful to answer your questions and get you back on the rails. Versus some out there that support a massive cross-section of industries and are looking for you to come to them with a defect and the steps to reproduce it. Not the question of, hey, I'm trying to get this out by five, how do I do it?
Bill's practical test is the best one I've heard: email their support address a few times during evaluation and time the replies. If you can't, find a current customer and ask them instead of asking the salesperson.
Support quality is the one part of a vendor relationship you can test for free before signing. It's also the part you meet on the morning nothing ships.
A related move worth paying for: get their trainers in front of you before the contract rather than after. Their trainers are who you'll actually depend on, and finding out then is better than discovering it through a reference call that was arranged for you.
What's Changed Since May 2025
The framework I describe on this episode is the System Fit Sprint now, with a published price and scope. The argument hasn't moved.
Bill closed by noticing that we booked forty minutes to talk about technology and spent it on people, processes and data quality. That's the honest summary of the work, and the reason a vendor conversation held before that work is finished tends to produce a contract rather than a fit.
What you sign up for is five years with a company, its support desk, and its idea of what a warehouse is. One of those three shows up in the demo.
