TL;DR
New industry polling shows a two‑speed AI path in pet food: marketing and sales get funded first, while production and formulation lag due to skills and data prerequisites. That’s not caution—it’s sequencing.
The state of play: early adoption, clear priorities
In a late‑August poll of 139 industry professionals, 36% reported using AI (about 16% across multiple functions; 21% in one or two areas). 26% said they’re not using AI, 26% were unsure, and 12% are still evaluating—classic early‑market distribution. Where leaders see value first: marketing and sales (35%)—chiefly customer targeting and campaign optimization—followed by market insights (21%), production efficiency (18%), and formulation and regulatory at 13% each. Fieldwork: Aug 7–21, 2025; published Aug 22, 2025. PetfoodIndustry
What’s in the way? The top barrier is expertise (nearly a quarter cite a skills gap). Legacy systems (15%), “no perceived need” (17%), and data quality issues (~10%) follow—while cost/unclear ROI together are ~7%, suggesting capital isn’t the primary constraint. PetfoodIndustry
Why marketing leads (and should)
Data readiness. Marketers already sit on consented first‑party and retail media signals that models can exploit with minimal re‑plumbing.
Modularity. You can pilot an AI audience modeler, copy assistant, or MMM uplift tool without rewriting your ERP.
Fast feedback loops. A/B tests give tight confidence intervals fast—fueling budget decisions without multi‑quarter lag.
These dynamics explain the poll’s pecking order: teams pursue high‑certainty, near‑term lift before heavier stack changes.
Don’t confuse “marketing first” with “marketing only”
The same data foundation that powers personalization (clean IDs, taxonomies, features) is reusable for demand forecasting, new‑product sizing, and channel mix. Leaders who design for data re‑use—shared feature stores, consistent claims libraries, and cross‑functional governance—convert early marketing wins into an enterprise AI flywheel.
A realistic two‑speed roadmap
- Front‑of‑house (0–6 months): audience modeling, creative/offer variant testing, and granular incrementality measurement.
- Back‑of‑house (6–18 months): production anomaly detection, predictive maintenance, yield optimization, and AI‑assisted formulation—after you’ve stabilized data pipelines and model governance.
Bottom line: The “split” isn’t indecision; it’s sequencing. Use marketing to prove value and harden guardrails, then port that muscle to ops.
Key stat box
36% using AI; 26% not using; 26% unsure; 12% evaluating. Top priority: marketing/sales (35%), then market insights (21%), production (18%), formulation (13%), regulatory (13%). Barriers led by expertise gaps. PetfoodIndustry
Source
Lisa Cleaver, Pet food industry split on AI adoption, marketing emerges as top priority (Aug 22, 2025; poll fielded Aug 7–21, 2025; n=139). PetfoodIndustry
