I’ve been skeptical about some of the grandiose claims surrounding AI implementation in the product world.

(I mean, aside from the stuff we see daily, like new ChatGPT competitors, and ways to create attractive but conceptually flawed ads or terrifyingly realistic video fakes.)

But then I got into a case study about Reckitt (maker of brands like Lysol, Mucinex, and Finish) that offers something inspiringly different:

Actual, measurable results from their generative AI pilots.

No vague promises of “transformation.” No fuzzy talk about “future potential.”

Just hard, valuable numbers from real-world implementation:

• Product development time reduced by 60% while improving quality
• Ad localization costs cut by 30% with increased asset quality
• Post-campaign media analysis time slashed by 90% with 2x quality improvement
• Emissions data collection for 25,000 products with 75x greater accuracy

Whoa.

What makes this case study super valuable is that Reckitt didn’t approach AI as one SINGLE solution, but rather identified specific operational pain points where the technology could deliver immediate, sought-after value.

In I Need That, I write about how true innovation often comes not from chasing technology trends, but from systematically identifying friction points in your processes and addressing them with appropriate tools.

Reckitt’s approach mirrors this perfectly. They didn’t ask “How can we use AI to say we did?”

Instead they asked “WHERE are our worst bottlenecks, and HOW could AI help remove them?”

Their pilot focused on four distinct areas:

Product Development: Using AI to accelerate concept development without sacrificing quality.

Creative Production: Adapting advertising assets for multiple channels and markets faster.

Marketing Analysis: Automating repetitive post-campaign analysis to free up human strategists.

Sustainability Tracking: Processing massive datasets to gain unprecedented visibility into emissions.

For me (and I think for you), the key insight here is not that AI is magical. Although it kinda is.

It’s that Reckitt made a STRUCTURED plan to use AI everywhere.

They identified specific, measurable pain points that were costing time, money, or quality, then applied the right tools to those exact problems.

This is not using AI just ’cause everyone is.

It’s entirely about carving out REAL competitive advantages.

In ways you can sell to the CFO, and measure in profits.

Product Payoff: Starbucks took a similar approach with their Deep Brew AI initiative, leaning away from flashy customer-facing features, and into predictive inventory management. By precisely forecasting demand at individual store levels, they reduced food waste by 15% and improved barista scheduling efficiency by 30%, delivering over $500 million in operational savings annually — all while using AI in ways invisible to the customer.

Action for today: Identify the three most time-consuming, repetitive tasks in YOUR product development or marketing processes. These “operational bottlenecks” are often the perfect places for AI augmentation, offering faster ROI than customer-facing applications.

Want to explore targeted AI implementations for your product operations? Tap that reply arrow and let’s discuss identifying your highest-value AI opportunities. Or reach out to my team of product strategy specialists at Graphos Product.