Retail
AI-powered cement for gaps between classic systems
A glance into designing and building transformative AI solutions for retailers.
Podcast Audio Transcript
Mimi: Hello listeners, this is Mimi; thank you for tuning in to yet another exciting and informative podcast from us at Infosys BPM. Today’s episode is for anyone who’s ever wondered - retailers have invested heavily in ERP, supply‑chain platforms, and planning tools, so why do some stubborn problems just refuse to go away?
To help us unpack that, I’ve got here Mehul Goyal, who leads Digital Solutions for Retail at Infosys BPM.
Mehul: Great to be here. And I’ll say this upfront—ERP didn’t fail. But retail has a few structural realities that classic systems were never designed to handle.
Mimi: Let’s start off on that note then, Mehul. Retail looks highly digitized on the surface – it is one of the most digitized industries, some would say. What’s still missing?
Mehul: Retail is digitized, but not digitally resolved. Traditional systems are great systems of record—they capture transactions and enforce control. But retail operates in what I call the messy middle: the space between process definition and real‑world decision‑making.
Three forces dominate this space:
- Silo friction across functions
- Decision latency as the business changes faster than plans
- Exception gravity, where deviations become the norm
Mimi: Good to explain that Mehul. Can you give us a concrete example?
Mehul: Sure – let’s take product onboarding as a process. Despite implementation of PIMs and ERPs, the process still involves emails, spreadsheets, and manual validation. These systems store the final answer, but they don’t manage the negotiation of truth between retailers and suppliers.
The result is human glue at scale, and long go‑to‑market timelines driven by exception gravity.
Mimi: Okay, thanks for explaining. Let’s explore different parts of the retail business with a similar lens. Merchandise planning, for example, why has it remained so challenging?
Mehul: Because planning systems still assume static plans in a world of live reality. Demand shifts daily, but decisions are locked into monthly or quarterly cycles. Planners are forced to choose between over‑buying and under‑buying—leading to markdowns or missed sales.
Mimi: Pricing often feels more like art than science. Why would you say that is?
Mehul: Traditional pricing tools offer visibility without strategy. Price is treated as a field, not a lever. Promotions are planned in silos, which creates silo friction and exception gravity when reality diverges from plan.
Mimi: And omnichannel?
Mehul: Retailers often have visibility without agency. They can see inventory, but systems don’t decide what to do with it—leading to cancellations, delays, and idle stock. Store labor systems face the same issue, enforcing compliance but ignoring local demand signals.
Mimi: Let me quickly recap. Across onboarding, planning, pricing, inventory, and stores, similar themes repeat:
- The messy middle
- Silo friction
- Decision latency
- Exception gravity
- and then you also mentioned Visibility without agency
Mehul: AI targets precisely these gaps. It helps manage the messy middle by supporting decision‑making, not just data storage. Agentic solutions reduce decision latency through continuous sensing and simulation, break silo friction by orchestrating cross‑functional decisions, and manage exception gravity by learning from deviations.
Companies like RELEX and o9 deserve credit for building retail‑native decisioning platforms that reflect how retail actually works. We’ll leave it to the audience to imagine what such sophistication might cost.
Mimi: Are there any final takeaways?
Mehul: Retail’s hardest problems were never missing software—they were missing judgment at scale. AI and agents allow retailers to institutionalize expertise, rather than rely on heroics.
Mimi: Thank you for joining us Mehul, and thank you for sharing your insights, I’m sure we can look forward to exploring a range of advancements soon.
Mehul: Fantastic, it was great to be here Mimi. I look forward to speaking to you soon
Mimi: Of course! If you have found this episode valuable, please subscribe and share it with your network. Our podcasts are available on Apple podcasts Spotify and several others. Please also remember to share and like it on social media. If you do have any queries, reach out to us through the email address on the podcast description. You can also watch our website via the Infosys BPM website www.infosysbpm.com for more exciting podcasts coming up. Once again, thank you for tuning in, stay safe and stay sharp and thank you to our guest. Have a great day!
