My guest on this episode of the podcast is <a href="https://www.linkedin.com/in/lfiaschi/">Luca Fiaschi</a>, a machine learning expert who previously held executive data science roles at MistPlay, StitchFix, and HelloFresh. Luca is now a Partner for the Generative AI vertical at <a href="https://www.pymc-labs.com/">PyMC Labs</a>, a consultancy that specializes in the application of Bayesian methods to business problems and which maintains the open source <a href="https://github.com/pymc-devs/pymc">PyMC library</a> for Bayesian statistical modeling as well as the open source <a href="https://github.com/pymc-labs/pymc-marketing">PyMC Marketing</a> media mix modeling library. The subject of my discussion with Luca is PyMC Labs&#39; recent paper, <a href="https://arxiv.org/html/2510.08338v1">LLMs Reproduce Human Purchase Intent via Semantic Similarity Elicitation of Likert Ratings</a>. I found the paper fascinating and wrote an overview on <a href="https://www.linkedin.com/posts/ericseufert_yesterday-i-wrote-about-why-marketing-teams-activity-7384593600791085058-93gj/">LinkedIn</a>; the paper&#39;s authors (from PyMC Labs as well as Colgate) use an LLM to score product concepts, finding that the distribution of LLM-produced scores can be comparable to that produced by human panels.Among other things, the podcast episode covers:<ul><li>Background on the paper, and how the partnership with Colgate-Palmolive came about.</li><li>An overview of the LLM querying methodologies used in the paper and how they produced score distributions.</li><li>The approaches used in the paper for calculating similarities between the LLM-produced and human-produced scores.</li><li>The conclusions / findings of the paper.</li><li>What the implications of the paper are on marketing creative ideation and the use of LLMs for evaluating product and advertising concepts.</li></ul>Thanks to the sponsors of this week’s episode of the Mobile Dev Memo podcast:<ul><li><a href="https://webshop.xsolla.com/home?utm_source=content-syndication&utm_medium=infuencer&utm_campaign=webshop2025&utm_content=MDMPod">Xsolla</a>. With the Xsolla Web Shop, you can create a direct storefront, cut fees down to as low as 5%, and keep players engaged with bundles, rewards, and analytics.</li><li><a href="https://www.incrmntal.com/">INCRMNTAL⁠⁠</a>. True attribution measures incrementality, always on.</li><li><a href="https://www.universalads.com/">Universal Ads</a> is Comcast’s self-serve TV ads platform that lets you launch campaigns in minutes across premium inventory from NBC, Paramount, Warner Bros. Discovery, Roku, and more.</li></ul>Interested in sponsoring the Mobile Dev Memo podcast? Contact <a href="mailto:ari@marketecture.tv">⁠Marketecture⁠</a>.

Podurama

Home

Trending

Personal

Your Podcasts

New Episodes

My Files

Favourite Episodes

Tags (Folders)

Recently Played

Bookmarks

Followed Playlists

Season 6, Episode 16: Can an LLM evaluate ad creative? (with Luca Fiaschi)

Mobile Dev Memo Podcast

AI Chapters

AI Insights

Reviews