How to Choose the Right WMS

WMS replacements are expensive. Not just in software licensing — in disruption. A poorly chosen system goes live, fails to meet expectations, and creates pressure to switch again within three to five years. By that point the business has spent money on implementation, training, and integration twice, and the operations team has lost trust in IT's ability to deliver.

According to Gartner, the average WMS implementation runs 20% over budget and 30% over schedule. The causes are usually not technical. They are misaligned requirements, underestimated integration complexity, and insufficient change management. The best way to avoid these outcomes is to invest in a rigorous selection process before signing anything.

This guide is written for operations directors, supply chain VPs, and IT leaders who are evaluating WMS options — probably for the first time, possibly for the second time after a disappointing first attempt. The framework here is practical, not theoretical. It reflects the evaluation processes we have seen work in real enterprise environments.

The single most common mistake in WMS selection is talking to vendors before you have a clear picture of what you need. Vendors are very good at demo-driven selling. If you go into a demo without specific requirements, you will end up buying the product that was demonstrated most effectively — not the one that fits your operation.

Start with a process walk. Spend time with your warehouse supervisors and operators mapping every movement type that happens in your facility: inbound receipts, putaway, replenishment, picking (zone, wave, batch, or discrete), packing, despatch, returns, cycle counting, and any specialist processes like kitting, cross-docking, or value-added services. For each movement type, capture the volume (orders per day, lines per order, units per line), the complexity (number of special handling rules, exception rate), and the current pain points.

This process walk typically takes three to five days for a single-site operation. It is not glamorous work, but it is the most valuable input into any vendor evaluation. You will identify requirements you did not know you had, and you will discover which capabilities your current system actually uses versus which ones your team has worked around.

Once you have your requirements, score each shortlisted vendor across twelve dimensions. Weight each dimension according to your priorities — a 3PL operator will weight multi-client capability heavily; a pharmaceutical distributor will weight lot traceability and compliance; an eCommerce retailer will weight carrier integration and returns.

The twelve dimensions are: (1) functional fit against your process walk; (2) integration depth with your ERP; (3) mobile and scanning interface quality; (4) reporting and analytics; (5) total cost of ownership over five years; (6) implementation methodology and timeline; (7) vendor financial stability and roadmap; (8) customer references in your sector; (9) configurability without custom code; (10) support model and SLAs; (11) cloud architecture and uptime guarantees; (12) data portability and exit provisions.

Do not skip dimension 12. Data portability — the ability to export your transaction history, bin configurations, and master data in a standard format — is easy to overlook when you are excited about going live, but it is critical if you ever need to switch systems again. Some vendors make this deliberately difficult.

The quoted licence fee is rarely the largest cost in a WMS deployment. The costs that surprise buyers most are: integration development (connecting the WMS to the ERP, carrier systems, and customer portals); data migration and cleansing; hardware (scanners, label printers, mobile devices); training; and the productivity dip during go-live.

In our experience working with mid-market operations (50,000 to 500,000 sq ft, 50 to 500 staff), the total first-year cost of a WMS implementation typically runs between 2.5x and 4x the annual licence fee. Year-two costs drop significantly once the implementation is complete, but ongoing costs — licence renewal, support, configuration changes, and integration maintenance — should be modelled over a five-year horizon to give an honest TCO picture.

When building your business case, benchmark your projected savings against realistic baselines. Labour savings are the most quantifiable: if your current pick accuracy is 98% and you can demonstrate a target of 99.5%, model the reduction in returns handling, customer credits, and re-pick labour. Inventory savings are real but harder to attribute directly to the WMS versus other changes. Be conservative.

Every WMS vendor will give you reference customers. They will be the ones who had good implementations. Ask for references in your sector and size range, but also ask for references who had difficult implementations — and how those were resolved. A vendor who has only good stories to tell is either new or not being honest.

When you speak to references, ask: how long did go-live actually take versus the plan? What was the biggest surprise? How responsive is the support team when something goes wrong? Would you choose this vendor again? Have you moved to a competitor since? That last question is easy to verify: ask the vendor for a list of churned customers in the last three years. Some will not provide it, which is itself informative.