How Pentland Brands Uses AI to Replace Product Photo Shoots (2026)

The death of the physical product shoot is not happening in a high-fashion studio. It’s happening in the quiet, operational back-rooms of the global wholesale market. By pioneering the transition toward Pentland Brands AI photo shoots, the multi-billion-pound group behind Speedo and Berghaus has quietly rewritten the rules of go-to-market speed. This is no longer a creative experiment; it’s a margin-protecting necessity. With retailers demanding marketing collateral half a year before manufacturing concludes, Pentland is entirely bypassing physical delays. They are generating highly realistic product imagery months before a tangible sample ever exists.

The macroeconomic imperative driving this shift is severe. Global supply chains have barely recovered from pandemic-era volatility, and capital costs remain historically elevated. According to recent Bank of England monetary policy reports, persistent inflationary pressures have forced consumer goods companies to aggressively defend profit margins through operational efficiency rather than price hikes. Waiting for physical samples to arrive from overseas factories before beginning wholesale negotiations is a luxury that modern fashion groups can no longer afford. Similarly, data from the Office for National Statistics (ONS) demonstrates that import frictions continue to add weeks to traditional retail timelines. Pentland recognised that their structural bottlenecks were not creative; they were chronological. By decoupling the sales cycle from the physical manufacturing cycle, they have insulated their B2B revenue streams from factory delays and shipping disruptions.

The Core Development: Erasing the Timeline

The most striking aspect of Pentland’s strategy, detailed at Shoptalk Europe 2026, is that it did not originate as a top-down technological mandate. The company launched an AI entrepreneurs programme, actively encouraging employees across all departments to test artificial intelligence against their daily frictions. The mandate was clear: let the technology infect the business from the bottom up. For the operations and creative teams, the most pressing friction was the wholesale selling packs.

Partners like Decathlon and JD Sports require comprehensive visual assets up to six months prior to consumer launch. Historically, if a factory in Asia was two weeks late producing a physical sample, the entire global sales schedule slipped. Today, Pentland has stopped physical shoots for these preliminary packs entirely. By partnering with specialist firms like Grasswold AI, they generate photorealistic imagery directly from digital design files. This compresses operational timelines from weeks into mere days.

The threshold for success was extraordinarily high. As Anna, a senior leader at Pentland, noted, early generic outputs simply lacked the fidelity required for technical apparel. For a brand like Speedo, the imagery cannot just show a garment; it must place models in specific, highly realistic pool environments to evoke genuine buyer confidence. Achieving this required moving past superficial prompt engineering and integrating proprietary brand assets with advanced generative models. Furthermore, the operational gains extend beyond photography. Pentland’s teams are now deploying real-time AI negotiation tools in the room with suppliers, modelling SKU combinations and pricing dynamically rather than retreating to spreadsheets.

Analytical Layer: Beyond Creative Novelty

The broader industry has fundamentally misunderstood the primary utility of generative AI in fashion. The dominant narrative focuses on consumer-facing marketing and creative campaign generation. Yet, the true economic value lies in backend timeline compression and structural coordination.

How is AI replacing product photography in fashion?

AI is replacing product photography in fashion by using digital CAD files to generate photorealistic imagery months before physical manufacturing begins. This eliminates the need for physical samples, bypasses supply chain delays, and allows brands to finalize wholesale selling packs and secure orders much earlier in the retail cycle.

This shift from physical to digital assets represents a profound restructuring of the go-to-market calendar. By adopting AI-generated product imagery, brands are effectively digitizing their supply chain’s most vulnerable touchpoints. However, scaling these isolated successes into enterprise-wide capabilities reveals significant foundational cracks. Pentland leadership has been unusually transparent about their current limitations, particularly regarding data architecture.

Despite managing both wholesale and direct-to-consumer (DTC) channels, the business currently lacks a unified semantic data layer. This means definitions of margin, revenue, and product metadata remain siloed across different brand divisions. You cannot fully automate operations if the underlying data speaks different languages. The second major hurdle is quantifying the exact return on investment (ROI). While executives can clearly see the anecdotal time saved by skipping physical shoots, translating distributed, qualitative efficiency gains into a hard financial metric for the board remains an unresolved challenge.

Implications and Second-Order Effects

The downstream consequences of this operational shift are vast, particularly for mid-market suppliers and competing retail groups. If a major player like Pentland can finalize their wholesale orders weeks faster than a competitor waiting on physical samples, they lock in retailer budgets before rivals even enter the room. This creates an arms race in B2B fashion wholesale.

• Supplier Dynamics: Manufacturers will increasingly be required to provide high-fidelity 3D assets alongside, or even before, physical prototypes.
• Retailer Expectations: Buyers at major sporting goods chains will come to expect fully visualized seasonal ranges earlier in the year, penalizing brands that still rely on mood boards or flat sketches.
• Talent Restructuring: The composition of fashion operations teams is changing. The demand for physical logistics coordinators is softening, replaced by a need for technical artists and systems integrators.

The Financial Times recently reported on the widening productivity gap between firms that treat AI as a workflow automation tool versus those treating it as a novelty. Pentland’s approach places them firmly in the former category. By building a repeatable framework to evaluate and scale employee-led AI experiments, they are constructing a resilient operational moat. They are preparing to extend this capability directly to DTC consumers, eventually bypassing the sample-and-shoot process for e-commerce portfolio pages as well.

Not everyone views the elimination of physical photography as a pure operational triumph. Dissenting voices within retail strategy argue that the over-reliance on synthesized imagery risks eroding brand equity. According to research from the OECD on digital consumer trust, hyper-optimized, digitally generated environments can trigger a “synthesization fatigue” among buyers, where the lack of physical imperfection translates to a lack of authenticity.

Critics argue that physical photo shoots, despite their logistical nightmare, force a serendipitous interaction between the product, the model, and the environment. Fabric drapes unpredictably. Light catches technical materials in ways that algorithms, currently, only approximate based on historical training data. If every brand in the wholesale catalog uses the same foundational AI models to generate their selling packs, the visual language of the industry risks collapsing into a sterile, predictable uniformity. That said, in a B2B environment where speed and inventory certainty dictate the survival of a brand, aesthetic serendipity is a premium that operations directors are increasingly willing to sacrifice.

The transition occurring within Pentland Brands is a masterclass in pragmatic technology adoption. They did not wait for a perfect, centralized corporate strategy to mandate artificial intelligence. Instead, they weaponized the frictions of their own employees, turning a chronic supply chain headache into a competitive advantage. The gaps they face regarding clean data architecture and ROI measurement are not unique; they are the standard growing pains of a legacy business rewiring itself for a new era.

The industry’s obsession with generative AI’s creative potential has blinded many to its true industrial application. AI is not just a cheaper camera; it is a time machine that pulls future revenue into the present by erasing physical delays. The brands that win the next decade will not be the ones with the most advanced algorithms, but the ones with the operational courage to stop waiting for the sample.