Most retail technology failures are not technology failures. They are operating model failures dressed up as technology decisions.
A POS modernization fails because the operating model assumes associate behavior that does not exist. An ERP migration creates data debt because no one decided whose system of record wins. An ESL rollout stalls because the network infrastructure in the pilot store does not match the network in 250 other stores.
I have walked into stores mid-pilot and watched associates use printed paper cheat sheets to work around a new POS workflow that nobody had tested against actual seasonal traffic. The system was live. The procedure was rejected. The program called that a successful go-live.
Where retail technology programs actually fail
The vendor names change. The platform generations change. The pattern holds.
Pilots run in flagship stores that do not reflect the real estate. A pilot in a well-lit flagship location with a dedicated IT resource on site, a fresh network, and a cooperative store manager produces results that will not survive contact with a strip-mall store running on a ten-year-old switch, staffed by seasonal associates, under a district manager who has not been told the program is happening. Rollout hits integration failures in 40% of stores. Nobody can explain why. I wrote about this pattern specifically in the Fail Fast post because I have seen it play out across ESL, POS, and network refresh programs.
Data model assumptions that do not survive acquisition integration. When one retailer acquires another, the two ERPs do not merge, even when the vendor deck says they do. I lived this during a full ERP conversion post-acquisition across roughly 190 stores in two countries. The real question was never whether the new system worked. It was which system of record wins for what, and who had the authority to make that call without a three-month escalation.
Vendor selection driven by procurement, not operating model fit. RFP scoring produces a number. The number picks a vendor. The vendor was evaluated against feature checklists and reference calls that the vendor curated. The operating model assumptions embedded in the platform get discovered during implementation, not during selection. By then the contract is signed.
Change management treated as an afterthought. Associate training, store readiness, comms cadence, and the actual adoption curve all sit outside the implementation budget until the week before cutover. At which point the program owner discovers that the people who will use the system every day do not know what it is.
How I work with retail operators
I have sat in the technology seat at a 265-store retailer, and before that at a 190-store group across the US and Canada. I run programs. I am not an outside observer.
That matters because the work I do as an advisor is the work I would do if I were the CIO. It is not a slide deck. It is structured around a specific question: what decisions need to be made in the next 90 days to keep this program credible, and who has the authority to make them?
The engagement pattern starts with the operating model, not the technology. How do stores actually run. Where do corporate and store decisions diverge. What do the seasonal peaks look like. Which operational constraints shape what is deployable. The technology architecture follows from those answers, not the other way around.
Retail-specific capabilities
ERP modernization and migration. I led a full ERP conversion post-acquisition across roughly 190 stores in two countries, and advised on roadmaps for others. The hard part of retail ERP is rarely the general ledger. It is pricing, merchandising, inventory, and promotions living across the ERP, POS, commerce platform, and merchandise systems at the same time. The integration design decides whether the migration ships on time, or accumulates two years of data debt before stabilization.
POS and payments modernization. I led a POS and payments program across 300+ stores that delivered $500K in annual payment savings and cut M&A onboarding timelines in half. The work was not the POS replacement. It was the payments architecture, the 3DS integration, the mobile POS workflow that associates would actually use in a furniture store aisle, and the rollback plan that nobody budgeted for and everyone needed.
Electronic shelf labels and digital pricing. I deployed Vusion ESL across 300+ locations, reaching near-100% pricing accuracy and saving 18 to 22 labor hours per store per week. ESL looks simple in the demo. In production, the decision that matters is how pricing propagates from ERP to POS to web to shelf edge, and what happens when those systems disagree at 3pm on a promotional weekend.
Omnichannel and commerce platform. SFCC at an apparel manufacturer, order management across physical and digital inventory, and the data-model decisions that decide whether BOPIS, ship-from-store, and endless-aisle actually work or just appear to work during the demo. Inventory truth is a data problem before it is a platform problem.
Store operations technology. Samsung Tab Active devices, VDI, associate experience, and the difference between a hardware-driven problem and an adoption-driven problem. These are usually adoption problems. I have stood in enough stores during go-live weekends to know where the workarounds start.
Data, analytics, and PIM. I have evaluated InRiver with Ntara as implementation partner during PIM selection work, worked with Centric PLM in apparel manufacturing, and led master data work that precedes any serious retail analytics investment. The reporting layer is the last 10% of the work. The other 90% is master data and integration.
Who I work with in retail
My advisory engagements in retail are with multi-site operators between 50 and 500+ stores, typically in one of three situations. Post-acquisition integration, where the new parent has two technology estates and needs a rationalization plan. Pre-modernization, where leadership has recognized the existing stack cannot support the next three years of growth and needs a credible roadmap before committing spend. Active program oversight, where a large implementation is in flight and the executive team needs independent visibility into what the implementation partner and the vendors are actually delivering.
The buyer profile is typically a CIO, VP of Technology, or COO. In smaller chains the CEO is often involved directly. In PE-backed retailers the operating partner is usually the first call. Every engagement runs independent of software vendors and systems integrators, which is unusual in this category and is the reason most of these calls happen in the first place.
Case studies
Two engagements that illustrate how this work lands in practice. Both ran across 300+ stores, both had measurable business outcomes, and both are documented with the specifics.
POS and Payments Modernization
300+ stores. $500K in annual payment savings. M&A onboarding timelines cut in half.
Case studyDigital Price Tags Across the Estate
Vusion ESL across 300+ locations. Near-100% pricing accuracy and 18 to 22 labor hours saved per store per week.
Frequently asked questions
How long is a typical retail technology engagement?
A focused roadmap sprint runs 2 to 4 weeks. A full current-state assessment plus target-state roadmap runs 6 to 10 weeks. Program oversight engagements typically run 3 to 6 months, sometimes longer for multi-year ERP programs. If someone is selling a 12-week assessment as a generic product, ask what decision it is actually supposed to enable.
Do you work with retailers under $500M in revenue?
Yes. In some ways smaller multi-site retailers benefit more from senior advisory than larger enterprises do. Large enterprises have internal architecture teams and vendor management functions. Mid-market retailers run the same complexity with a fraction of the bench strength, and a few right decisions in the first 90 days of a program materially change the outcome.
Can you support selection of a specific vendor (STORIS, SAP, NetSuite, Microsoft Dynamics, Oracle Fusion)?
Yes. I have evaluated and worked alongside all of them in different contexts. No referral fees, no vendor incentives, no implementation partner relationships that shape the recommendation. That independence is rare enough in this category that it is often the reason a retailer calls.
Do you deliver or just advise?
Advisory with delivery governance. I oversee programs and hold vendors accountable. I do not replace the implementation partner or build the software. The distinction matters: my value is senior judgment on decisions that are difficult to reverse, not extra hands on a Gantt chart.
The question is not whether your retail technology stack will be modernized in the next three years. It is whether the next modernization produces a credible platform for the business, or another eighteen months of debt and a second selection cycle that nobody wants to run.