Episode · 2026-04-15
2026 DTC Software Rankings
2026 DTC Software Rankings | Lululemon Investigation | Michael Brandt x Adam Kitain x Tyler Simmons
Rundown
Chris Hall and Colin run through new attribution-platform market share data before hosting three guests: Adam Kitain of Intelligems on AI-driven offer/price testing, Tyler Simmons of Bucket Golf on retail expansion after Shark Tank, and Michael Brandt of Ketone IQ on category creation and consumer education. The show blends software rankings, DTC-to-retail strategy, and brand-building tactics for founders.
Takes from this episode
Triplewhale has unexpectedly pulled ahead of legacy players like Northbeam and Rockerbox in both SMB and mid-market attribution adoption.
Number one, Triplewhale, 14% mid-market adoption. Number two and number three, House and Northbeam tied at 9% mid-market adoption. I did not see this coming.
AI agents like the Claude SDK will soon remove the need for dedicated developer hours to run A/B and offer tests, lowering the skill barrier to testing tools.Prediction
What if you didn't need ten developer hours? What if it was just the Claude SDK running in the background?
Dynamic pricing on identical retail goods invites consumer backlash, whereas dynamic offers (or pricing in categories where variance is already expected, like flights) are viable.
I love the idea of looking at it from dynamic offers and not dynamic pricing because you will get a ton of blowback from dynamic pricing.
Bucket Golf's organic-social-first following translated directly into retail sell-through, with stores like Dick's Sporting Goods unable to keep the product stocked.
we're in every single Dick's Sporting Goods store now and they can't really keep us on the shelves... heat online, heat on the internet, it drives purchases
True category-creating products need long-form educational content (podcasts, extended explainers) to land, while incremental 'better version' products can win purely on a quick shelf impression.
if you're creating something really inventive and new then you need to think about surface areas where you can download a larger packet of information into people's brains... that 20 or 30 second info packet moves mountains