WMS Implementation Cost – Part Two

What Actually Drives the Cost of a New WMS (Part Two)

This is the second part of a long, honest look at what it actually costs to implement a new WMS. We wish it could be shorter. In the spirit of transparency, it can’t be.

Part one covered the elements that come together to make up a meaningful price. If you’re upgrading any time soon, start with part one, which will at least arm you with some awkward questions to ask anyone who gives you a price too quickly. It’s also worth reading how to write an RFP that gets you the WMS you actually want . This part is about how to get a realistic price. Let’s spill the beans.

What makes your number bigger than their number?

You’re not going to price out your project from a couple of articles. But you can work out, fairly quickly, which end of the spectrum you’re likely to be on. These are the conditions that move the final number.

More than one site

Any second site usually costs significantly less than the first, and ten sites cost less again per site. But multi-site brings its own work: shared master data, inter-site transfers, consistency of process across the whole operation, and the sometimes political question of whose way of working becomes the standard.

SKU characteristics, more than SKU count

Fifty thousand identical boxes is a far simpler operation than four thousand items carrying batch tracking, expiry dates, serial numbers, catch weight, temperature requirements and a sprinkling of hazardous goods. On cost, complexity of stock beats volume of stock almost every time.

The real number of systems your WMS has to talk to

Count them all. Then add the things that are technically spreadsheets but function as systems, because they count too, and they’re usually the ones uncovered after everything’s been settled. More information, earlier and more clearly, helps everyone.

Automation

Conveyor, sortation, AS/RS, robotics, pick-to-light, voice. Each one is an interface, and each interface is additional scope.

How far your process is from ‘standard’

Standard is cheap, because it’s functionality you turn on and off during configuration. Bespoke inside your WMS is expensive, because it’s development, and it stays expensive forever because of the testing and upgrades it needs. This is the single biggest lever you personally control. It’s genuinely worth asking which of your requirements are real and which are just how you’ve always done things. And if you truly must have things your way, look into the alternatives to customising your core WMS, because they do exist.

The state of your data

Covered in more detail below, because it needs to be. Sorry.

Countries and languages

Multiple tax and customs regimes? Multiple languages on the Ops floor? Multiple sets of local compliance? All of it is real work, which means real cost.

Replacement or greenfield?

Replacing a live system in a running warehouse is sometimes harder than starting fresh, because you have to cut over without stopping. A new building, with no legacy and no live stock, is the easiest version of this kind of project.

What your peak looks like

A steady operation and a spiky one need very different amounts of testing, resilience and go-live planning. If your December is ten times your June, that single fact needs to shape your timeline as well as your cost.

Unexpected invoices

Not everyone runs the No Surprises policy we stick to. Every experienced buyer should know where overruns actually come from. In our experience, it’s only ever a few places.

Scope that appeared halfway through

Something your business does that nobody on the floor mentioned during design, usually because they don’t think of it as a process. It’s just what they do every third Thursday. Avoid it by being explicit and walking your provider through every step of your operation, especially the parts you think are blindingly obvious.

The integration nobody thinks is an integration

Almost always a customer requirement, a carrier, or a file that gets emailed somewhere to do something. Nobody hides these. It just didn’t occur to anyone they were part of the system landscape. Map out all your processes before you ask for quotes.

Data in worse shape than anyone realised

This one’s a biggie. Everyone knows their master data is imperfect. Very few know how imperfect until someone tries to load it into a new system. What we see a lot: locations that don’t physically exist, products with three competing units of measure, stock nobody has counted since before the pandemic. It’s closely tied to what poor inventory accuracy is really costing you [reduce-warehouse-errors].

Hardware that actually works

Some devices can’t read a label at minus 22 degrees. Sometimes there’s a signal blackspot behind the racking. Occasionally we meet a printer older than most of the pickers.

Peak, and any freeze around it

Real talk: a project that slips into your peak window will not slip by two weeks. It will slip to January, because almost nobody chooses to go live in November. That changes the cost of delay significantly.

None of these is unavoidable. All of them are cheaper to find in week two than in week twenty. That’s exactly what a proper discovery phase is for, and why you should be wary of anyone who’ll quote without doing one.

How to get a real number out of anyone (including us)

If you take one thing from this article, take this: the quality of the quote you get is decided almost entirely by the quality of the information you share. Offer the following and you’re far more likely to get something meaningful from any vendor worth talking to.

• Sites: how many, how big, what type, owned or leased, and whether more are coming.
• Order profile: average lines per order, average units per line, and the split between pallet, case and each picking. It drives the design.
• SKU count and characteristics: batch, serial, expiry, temperature, weight, anything hazardous. Mention any stock you mix in a specific way or keep quarantined.
• Stock actions: if you do kitting or similar, say so.
• Volumes: an average day versus your worst day, and the shape of your year. Talk about your real peak, not a hypothetical one.
• Systems inventory: everything the WMS will exchange data with. Include the spreadsheets. Especially the spreadsheets.
• Automation: what you have, what you’re committed to, what you’re considering.
• People: headcount, shift patterns, agency proportion, turnover, languages spoken on the floor.
• Current state: what’s broken, what it’s costing you, what you’ve already tried.
• Timing: your deadline and the reason behind it. A date driven by a lease expiry is a very different constraint from one driven by ambition.

Then judge the responses. A vendor who reads all that and comes back with a number five minutes later probably hasn’t read it. The one you want reads it and asks more questions, because those questions are the difference between a realistic quote and a guess. (If you want a steer on what ‘good’ looks like before you start, our piece on benchmarking your operation is a sensible place to begin.)

The cost of doing absolutely nothing

The information here is probably making your head spin. It’s a lot. So it’s worth recognising that doing nothing is also a decision, and it has a tangible cost. It’s a decision to keep doing what you’re doing, even if you started reading this with a very big reason not to.

That cost is spread across overtime, agency cover, stock write-offs, mispicks, customer credits, the space you’re renting because you can’t fully use the space you’ve got, and the growth you’ve had to turn down because you can’t service it. It’s much harder to see than the number at the bottom of a quote, which is precisely why ‘staying where you are’ tends to win by default.

You don’t need us to work those costs out. You need your own numbers, carefully collated, and a sense of what good looks like in your competitors’ warehouses. But if you’d like a real number for upgrading your operation, bring us what you’ve got and we’ll tell you what’s missing. We don’t do sales scripts, we don’t do standard demos, and we don’t have a pricing page with three tiers and ‘Let’s talk’ where the price should be.

Book an obligation-free discovery call and we’ll show you your real potential, rather than a work of fiction that only slows your progress down.

Next in the series: how long, realistically, a WMS project actually takes.