17 april. 2026
Retail talks about AI and marketing but the problem still lies in product data
While the retail industry is investing heavily in digitalization and e-commerce, product data between brands and retailers is still often handled manually resulting in significant consequences for speed, quality, and cost.
In recent years, retail has accelerated its digital transformation. Webshops have become more advanced, marketing more data-driven, and the customer journey more complex. Yet one area still significantly lags behind: the management of product data between brands and retailers.
In practice, the exchange of product information, images, and formats is still often done via Excel sheets, emails, and manual corrections. Each retailer has its own requirements for image sizes, metadata, descriptions, and categorization, meaning the same product must be adapted repeatedly. For many organizations, this has become an accepted condition—but the consequences are greater than most realize
Manual processes slow down the business
When product data is handled manually, errors and delays are inevitable. Product launches are postponed, assortments go online late, and teams spend a disproportionate amount of time on operations rather than development.
At the same time, complexity continues to grow with:
- More sales channels
- More markets
- More partners
What worked with a limited number of products and retailers simply does not scale.
Product data impacts the entire value chain
The problem does not stop with the e-commerce team. Poor or inconsistent product data has ripple effects across the entire value chain:
- Marketing struggles with incorrect or incomplete product feeds
- Retailers spend time correcting supplier data
- SEO and conversion suffer due to insufficient product descriptions
- Supply chain and commercial teams lose visibility
Despite this, product data is rarely positioned as a strategic responsibility. It often sits in an organizational gray area between IT, e-commerce, and supply chain.
AI and automation are being used in the wrong places
There is a lot of talk about artificial intelligence in retail. The focus is often on campaigns, marketing, and personalization. However, some of the most obvious efficiency gains are found in more practical, operational processes.
Automation of product data, image formatting, and data flows between brands and retailers can significantly reduce manual work, improve data quality, and shorten time-to-market. Yet this area is often given low priority because it is perceived as technical and less visible.
Time to rethink product data as a strategic discipline
If retail truly wants to realize the benefits of digitalization, it requires a break from the manual workflows that still characterize product data management.
Product data is not just a technical issue—it is a business-critical discipline that affects collaboration between brands and retailers, speed to market, and the quality of the customer experience.
The question is not whether retail can afford to automate product data.
The question is whether it can afford not to.
17. april 2026
Detailhandlen taler om AI og marketing, men problemet ligger stadig i produktdata
Mens detailhandlen investerer massivt i digitalisering og e-commerce, håndteres produktdata mellem brands og retail stadig ofte manuelt med store konsekvenser for hastighed, kvalitet og omkostninger.
Detailhandlen har gennem de seneste år accelereret digitaliseringen. Webshops er blevet mere avancerede, marketing mere datadrevet, og kunderejsen mere kompleks. Alligevel halter ét område fortsat markant: håndteringen af produktdata mellem brands og retailere.
I praksis foregår udvekslingen af produktinformation, billeder og formater ofte stadig via Excel-ark, mails og manuelle rettelser. Hver retailer har sine egne krav til billedstørrelser, metadata, beskrivelser og kategorisering, hvilket betyder, at det samme produkt skal tilpasses igen og igen. For mange organisationer er det blevet accepteret som et vilkår, men konsekvenserne er større, end de fleste regner med.
Manuelle processer bremser forretningen
Når produktdata håndteres manuelt, opstår der uundgåeligt fejl og forsinkelser. Lanceringer bliver skubbet, sortimenter kommer sent online, og teams bruger uforholdsmæssigt meget tid på drift frem for udvikling.
Samtidig bliver kompleksiteten kun større i takt med:
- Flere salgskanaler
- Flere markeder
- Flere samarbejdspartnere
Det, der fungerede, da man havde få produkter og få detailhandlere, skalerer ganske enkelt ikke.
Produktdata påvirker hele værdikæden
Problemet stopper ikke ved e-commerce-teamet. Dårlig eller inkonsistent produktdata har afledte effekter i hele værdikæden:
- Marketing kæmper med forkerte eller ufuldstændige produktfeeds
- Retailere bruger tid på at rette leverandørdata
- SEO og konvertering lider under mangelfulde produktbeskrivelsere
- Supply chain og commercial teams mister overblik
Alligevel er produktdata sjældent placeret som et strategisk ansvarsområde. Det befinder sig ofte i et organisatorisk grænseland mellem IT, e-commerce og supply chain.
AI og automatisering bruges de forkerte steder
Der tales meget om kunstig intelligens i detailhandlen. Fokus ligger ofte på kampagner, marketing og personalisering. Men nogle af de mest oplagte effektiviseringsgevinster findes i de mere lavpraktiske processer.
Automatisering af produktdata, billedformater og dataflow mellem brands og retailere kan reducere manuelt arbejde markant, øge datakvaliteten og forkorte time-to-market. Alligevel er området ofte lavt prioriteret, fordi det opfattes som teknisk og mindre synligt.
Tid til at gentænke produktdata som strategisk disciplin
Hvis detailhandlen for alvor vil høste gevinsterne af digitalisering, kræver det et opgør med de manuelle arbejdsgange, der stadig præger produktdata-håndteringen.
Produktdata er ikke blot et teknisk spørgsmål, det er en forretningskritisk disciplin, der påvirker samarbejdet mellem brands og retailere, hastigheden i go-to-market og kvaliteten af kundeoplevelsen.
Spørgsmålet er ikke, om detailhandlen har råd til at automatisere produktdata.
Spørgsmålet er, om den har råd til at lade være.
