Excel survived because it works
Excel is not elegant, but it is universal. A supplier can send it. A buyer can open it. A product team can edit it. Operations can filter it. A manager can understand it without asking IT for another login.
That is why Excel survived. Not because companies are stupid, but because commerce is fragmented. Every supplier has a different format. Every retailer has different requirements. Every marketplace wants different fields. Every category has its own rules. Excel became the place where all those differences meet.
In that sense, Excel is not the enemy. Excel is the translator everyone already knows.
The problem starts after the file arrives
The issue is not receiving Excel files. The issue is what happens next.
One supplier says “colour.” Another says “main color.” Another says “base_colour.” One file uses centimeters, another uses millimeters. Some values are missing, some fields are free text, some categories do not match, some packaging data is incomplete, and some images are referenced but not attached.
Then people start fixing. They rename columns, correct units, split fields, combine fields, map categories, check mandatory attributes, chase suppliers, and prepare the file for the PIM, ERP, webshop, marketplace, compliance workflow, or analytics tool.
That is where the cost hides. Not in Excel, but in the manual repair work around Excel.
Automation breaks when the input changes
Automation is easy when the input is stable. Commerce is not stable.
Supplier files change. Columns appear. Columns disappear. Product ranges expand. Marketplaces add requirements. Regulations ask for more packaging data. One field suddenly needs to become three. That is why teams keep checking manually. They are not against automation. They just know the data coming in cannot be trusted blindly.
AI does not magically fix that. AI can classify, enrich, detect patterns, and speed things up. But if the source data is messy, incomplete, or unclear, AI still needs a foundation it can understand. Otherwise, you get faster uncertainty. Very modern, not very useful.
This is where digital transformation gets real
Companies love talking about scalable AI, trusted data, governance, and automation. Good. But here is the practical test: can you turn the messy supplier Excel file into usable, validated, structured data without five people rebuilding it every time?
If not, AI will not scale properly. Dashboards will be questioned. Marketplaces will reject fields. PIM systems will contain gaps. AI tools will make assumptions. Teams will keep checking manually because nobody fully trusts the output.
That is not a technology problem. That is a data handover problem.
The boring work decides the outcome
The boring work is mapping, validation, attribute logic, field requirements, exception handling, supplier feedback, category rules, and ownership. Nobody puts that first on a keynote slide.
But this is the layer that decides whether product data can move from suppliers into systems, from systems into channels, and from channels into customer-facing experiences. It also decides whether automation becomes useful in daily operations, not just impressive in a demo.
The winners will not only be the companies with the fanciest AI story. They will be the companies that understand incoming data better, clean it earlier, and stop treating spreadsheet repair as an invisible cost of doing business.
Excel will not disappear
Excel is not going away, and maybe that is fine. The goal is not to pretend every supplier, retailer, brand, and marketplace will suddenly use one perfect system. That is not how commerce works.
The better question is: how do we make Excel less painful?
How do we accept that Excel is still the language commerce speaks, while making sure the data inside it becomes structured, validated, and ready for systems, teams, and AI?
Before commerce can scale AI, it still has to understand the spreadsheet that arrived this morning.
This blog post is a contribution by ProductFlight, written by Thorin Schiffer, Founder, CEO and CTO. ProductFlight helps organizations with AI-driven product data transformation for modern retail. For more information, visit www.productflight.io or stop by the ProductFlight booth at Data Expo.
