Industry
Retail
Omnichannel CX and Intelligent Innovation – The Future of Retail
HIGHLIGHTS
- Modernisation is essential for DIY retailers, but successful migration does not always guarantee business confidence. Protecting product meaning helps ensure that product information remains trusted, useful, and aligned to business intent long after systems change.
- Retailers rarely struggle because information disappears completely. More often, information remains available but no longer behaves as the business expects.
- If the rules, relationships, and approval controls behind that information change without clear ownership, the outputs may still be generated, but they become harder for business teams to explain, trust, and act upon.
On this page
On this pageinpage
When data survives but meaning doesn't
The big picture Modernisation is essential for DIY retailers, but successful migration does not always guarantee business confidence. Protecting product meaning helps ensure that product information remains trusted, useful, and aligned to business intent long after systems change. |
DIY and home improvement retailers manage some of the most complex product landscapes in retail. A single range can span thousands of products, multiple customer missions, specialist supplier inputs, seasonal peaks, and local market variations. That means, modernising the platforms behind this landscape is essential for improving customer experience, supporting omnichannel operations, strengthening resilience, and preparing the business for AI-enabled retail.
The difficulty is that a migration can look healthy on paper while the business quietly loses confidence in its data. Data gets migrated, validations are passed and the dashboard turns green. Then a trading manager finds a range sitting under the wrong category, or a store colleague notices that a product page shows the wrong image.
We saw this first-hand on a recent multi-market product data migration for a home improvement retailer. The weekly progress updates looked encouraging. Most batches had loaded, and the issue log was steadily shrinking. Very few records were missing, but values such as material or fixing type no longer matched the approved lists. Some images were tagged with the wrong content type, and a handful of special characters in free-text fields were not handled correctly by the new platform. A few days later, an upstream export dropped category-specific attributes for part of the final delta load. The products still arrived, just without some of the features customers rely on. Because reconciliation was running, the gap was caught, contained and reloaded before go-live. Without it, the dashboard would have stayed green.
Retailers rarely struggle because information disappears completely. More often, information remains available but no longer behaves as the business expects. |
Why meaning matters
Product information sits at the centre of retail operations. Customers use it to discover and compare products. Merchants use it to build ranges. Suppliers use it to enrich content. Digital channels rely on it to power search, recommendations, and product pages. When that information becomes less reliable, the impact extends far beyond technology teams.
Someone tiling a bathroom needs adhesive, grout and sealant alongside the tiles, ideally shown together with accurate stock. If the grout ends up in the wrong category, or the "suitable for wet areas" attribute is dropped, the project basket falls apart.
The same challenge appears in merchandising and supplier operations. If a hierarchy no longer reflects how trading teams plan the range, products can lose relevance or appear in the wrong place. If enrichment rules or approval workflows change, suppliers may face repeated rework, unclear reasons for rejection , or slower product setup. For retailers focused on speed to market, this can become a hidden operating cost.
There is also a governance and AI dimension. Retailers increasingly rely on product information to support analytics, recommendations, automated enrichment, and AI-assisted decisions.
If the rules, relationships, and approval controls behind that information change without clear ownership, the outputs may still be generated, but they become harder for business teams to explain, trust, and act upon. |
Meaning can drift
Product meaning is the business understanding attached to product data - how a product should be categorised, described, localised, enriched, governed, approved, and used across channels and operations. Protecting product meaning, therefore, means protecting not only the data values, but also the business logic and context that make those values useful.
That risk is what we call product-meaning drift. It occurs when product data remains available, but the context behind it weakens. A category may still exist, but no longer reflects merchandising logic. A translation may no longer fit the local market. An asset may be linked to the wrong product. A rule may continue to operate but behave differently after modernisation. The information is still present, but the business understanding behind it becomes less dependable.
For DIY retailers, this matters because product information does far more than populate product pages. It supports search, recommendations, supplier workflows, compliance checks, store operations, reporting, automated enrichment, personalisation, and AI-assisted decisions. If product meaning drifts during modernisation, the impact can be felt across the business long after the programme is considered complete.
Where meaning can drift |
What the business may see |
What is at risk |
Category and hierarchy |
Products appear in the wrong place |
Findability, merchandising consistency, digital shelf quality |
Attributes and values |
Specifications are missing, weak, or misleading |
Product comparison, customer confidence, operational efficiency |
Translations and local content |
Product pages feel inconsistent across markets |
Customer experience, launch readiness, market consistency |
Content and assets |
Images, documents, or descriptions are missing or mislinked |
Conversion quality, digital shelf trust, colleague productivity |
Supplier workflows |
Setup slows down and rework increases |
Vendor experience, speed to market, operating efficiency |
Rules and AI signals |
Data is present but less reliable for decisions |
AI trust, governance sign-off, regulatory confidence |
Building confidence during modernisation
Retailers can reduce product-meaning drift by treating product information as a business asset throughout the modernisation process, rather than simply as a technical migration output.This approach need not slow delivery. It gives business, data, and technology teams a shared view of the product context that must be protected while systems, workflows, and rules change around it.
Defining what matters most is the first step. This includes category logic, key attributes, market rules, translations, assets, supplier workflows, compliance requirements, and approval logic, all of which are critical to daily operations. The goal is to agree on what “good” looks like before the change, not after problems appear.
Validation is equally important. A product may pass a migration check but still fail when a customer searches for it, a supplier enriches it, a store colleague uses it, or an AI use case depends on it. Testing should focus on real business journeys, including project-based shopping, seasonal ranges, local market requirements, supplier enrichment paths, and edge cases across categories and languages.
Business confidence comes from visibility, not assumptions. Before sign-off, teams should be able to show what matched, what changed, what was corrected, what was accepted, and what still needs ownership. Confidence does not come from a migration status marked complete. It comes from knowing which customer, supplier, merchandising, compliance, and AI-related journeys are safe to resume and which risks still require action.
Rule definitions in source systems often hold years of validation, merchandising and compliance decisions. There can be some which follow a fixed or simple pattern which would be relatively easy to transform and migrate. However, there can also be certain patterns which cater to complex scenarios. AI can help surface these and propose target logic. In AI assisted complex rule transformation scenarios, a business or data owner should approve each recommendation, and the reasoning should stay traceable. The work should continue after cutover, because ranges, suppliers and markets will keep changing.
Retailers should therefore treat product meaning as an ongoing business discipline, not a one-time migration checkpoint. |
Preparing for what's next
A modernisation programme is one of the few moments when a retailer's product data, rules and workflows are all open to change at once. That makes it the right moment to decide what must be protected. Retailers that get this right go into AI adoption with product data their teams already understand and trust.
More product data will not, by itself, make retail AI work. What matters is whether that data still means what the business intends after the systems change. TCS helps retailers protect that meaning through modernisation, giving them a stronger base for customer experience, day-to-day operations and AI adoption. |