Brilliant Commerce29 Loka 2025

Belkin's Jyoti Malik on Agentic Commerce: Audit Data Quality Before Chasing AI Agents

Belkin's Senior Director of E-commerce tested AI chatbots for customer service beyond basic order tracking, then deliberately pulled back to human agents for relationship-critical moments. Jyoti Malik explains why most omnichannel strategies fail (teams treat digital as a retail add-on instead of rethinking the entire operating model), shares her phased approach to channel expansion, and reveals the three-part assessment she uses before investing in first party data infrastructure.

The practical takeaway: before chasing martech solutions or agentic commerce readiness, map your business outcomes two years forward and audit whether you can actually act on the data you collect. Jyoti's team experiments with AI for content creation (a non-creative team member now generates banner assets in ChatGPT), but she's cautious about where automation creates value versus where it damages customer relationships.

Topics discussed:

Phased omnichannel expansion starting with shared business pain points and pilot programs
Why digital commerce as a "retail bolt-on" operating model fails in mature organizations
Three-part first party data assessment before infrastructure investment: quality, quantity, and utility in your business context
Preventing martech tech debt by defining business needs before evaluating solutions
Testing AI chatbots beyond order status and why capabilities fall short for complex customer service
Agentic commerce readiness through data quality audits and friction reduction strategies
AI content tools for non-creative team members generating campaign assets
Decision framework for when to automate versus when to keep humans in customer experience

Tämä jakso on lisätty Podme-palveluun avoimen RSS-syötteen kautta eikä se ole Podmen omaa tuotantoa. Siksi jakso saattaa sisältää mainontaa.

Jaksot(34)

Why community and education became Evvy's entire growth engine

Women weren't required in US clinical research until 1993. Priyanka Jain built a company to close that gap. Evvy now holds the world's largest real world data set on the vaginal microbiome.Priyanka te...

7 Loka 24min

Saying no to business: The hard call behind Centra's fashion-only focus

What happens when an agency spins its internal tooling into a software company built for one vertical only?Jenna Holcombe tells Bryan how a hard call to focus exclusively on fashion, made around 2019,...

24 Syys 41min

Why the AI moment of discovery is retail's next Google ZMOT

What if the biggest AI opportunity for retailers is knowing where not to automate? Chris Parsons, VP of Partner Growth and Marketing at Hale, spent 25 years inside retail at Walmart Canada, Ren's Pets...

9 Syys 33min

How ad platforms shifted from buying audiences to buying outcomes

Meta's algorithm started rewarding new creative volume, and the industry treated it as a race to publish more. Vincent Bourzeix, Founder and CEO of Superbolt, saw the problem coming.Vincent walks Brya...

26 Elo 45min

Why over-personalizing your site dilutes your brand story

With nearly two decades in e-commerce and stints at Victoria's Secret, Bed Bath & Beyond, The Children's Place, and Foot Locker, Laurie Falletta brought something most e-commerce leaders don't have wh...

12 Elo 38min

Why unit economics decide how fast Bloom & Wild can expand into new gifts

Charlotte Langley built Bloom & Wild's retention engine by paying close attention to complaints too small to register as data, like customers who wanted out of Mother's Day marketing entirely.She tell...

29 Heinä 41min

From 1% to 11% brand awareness: how WeCook built a brand worth caring about

What does it actually take to build a brand that earns permission to expand into retail, land cultural partnerships, and grow aided awareness from 1% to 11%? Matthew Morein, Senior Director of Brand a...

13 Heinä 40min

How Summersalt built a swimwear brand from 10,000 body scans

Summersalt launched from a data set most brands would never attempt to build. The team scanned 10,000 women, made a swimsuit for each, and collected detailed feedback on fit. That pattern library stil...

1 Heinä 38min