From the field · Event notes

Notes from the Retail Shopping Summit 2026: where do AI agents fit in retail's new codes?

·4 min read·Skyloop Cloud

On 22 October 2026 we were at the Retail Shopping Summit (Perakende Alışveriş Zirvesi) at Fişekhane in Istanbul. The Skyloop and ZZeti team spent the day at the Nebim booth, together with the Nebim team. The booth's headline summed up the summit's agenda as well: ERP, retail and agentic AI solutions that carry a business into tomorrow.

This article covers the frame of the summit, why we were there, and the four questions every conversation about AI in retail comes back to. It is not meant as an event recap. It is a plain account of what retailers expect from AI and how that expectation is met on their own data.

The summit: retail's new agenda under the theme "New Codes."

Organised by Bi'Sektör, the Retail Shopping Summit was held for the third time this year, under the theme "New Codes." The programme put the effect of AI on business processes, new growth models, changing consumer behaviour, customer experience and sustainability on the same stage. The audience matched it: retail brands, shopping-centre managers, investors and technology providers.

What the theme means to us is clear. The code of competition in retail is no longer only store count or the campaign calendar. When two companies look at the same data, the one that notices earlier and acts faster wins.

The Nebim booth at the Retail Shopping Summit 2026: team photo under the headline ERP, Retail and Agentic AI solutions
Retail Shopping Summit 2026, Fişekhane, Istanbul — at the Nebim booth.

Why were we at the Nebim booth?

Nebim V3 is one of the most widely used ERP platforms in Turkish retail. Sales, stock, cost and customer data live there. ZZeti Zeka is an enterprise AI platform that connects to that data directly through an MCP connector and runs on the company's own infrastructure. Skyloop is ZZeti Zeka's authorized reseller and implementation partner in Türkiye.

So if AI in retail is the topic, the right place to stand is next to the ERP. For agents to be useful they first have to reach the data and then be able to do something with it. That is what "agentic AI" on the booth meant: not an assistant that answers questions, but a system that spots a deviation and opens a task for the right person.

Four questions retailers ask about AI

When AI comes up in retail, the conversation usually arrives at the same four questions. Our answers are short:

  • Where does it make money? Where the P&L leaks: lost sales, stock waiting in the wrong store, late markdowns, cost increases that reach the price late, and customers drifting away.
  • Does my data leave the company? It does not have to. The platform runs on the company's own server or its private cloud in Türkiye, and so do the models.
  • How does it connect to my ERP? Through a ready-made MCP connector for Nebim V3. The agent reaches the data through named tools and every call is logged.
  • Who takes the action? The agent prepares the recommendation and assigns it as a task; the store manager, the regional manager or headquarters decides. The result is measured in the same place.

What does agentic AI mean in retail?

For us it is a four-step loop: detect, diagnose, assign and measure. Every day the agent scans the data, finds the point that deviates from what was expected, writes down the likely cause, assigns the work to its owner and tracks the result. Access follows the hierarchy: a store manager sees their store, a regional manager their region, headquarters everything.

In retail we apply this loop in six use cases:

  • Smart product allocation
  • Demand forecasting
  • Dynamic pricing and campaigns
  • Cash flow optimization
  • Personalized sales
  • Operational efficiency

After the summit

Thank you to everyone we met at the summit and talked to at the booth, and to the Nebim team for hosting us.

If you would like to see which use case would pay back fastest on your own Nebim V3 data, we suggest starting with a single use case and a metric defined up front. Our retail page explains how the six use cases work in detail.

Takeaway

One of retail's new codes is AI agents that do not stop at looking at data but act on it. For that to work, the agent has to connect to the ERP, the data has to stay inside the company, and the decision has to reach the right person.

Frequently asked questions

When and where was the Retail Shopping Summit 2026 held?

The summit (Perakende Alışveriş Zirvesi) took place on 22 October 2026 at Fişekhane in Istanbul. It is one of Bi'Sektör's event series and was held for the third time this year, under the theme "New Codes."

How did Skyloop and ZZeti take part?

The Skyloop and ZZeti team was at the Nebim booth together with the Nebim team. The booth's headline was ERP, retail and agentic AI solutions.

How does ZZeti Zeka work with Nebim V3?

ZZeti Zeka connects to Nebim V3 data through a ready-made MCP connector. Agents read sales, stock and customer data on the company's own infrastructure, detect deviations, write down the cause and assign the work to the right person.