September’s follow‑up poll (fielded September 4–18, 2025; n=119) zooms in on marketing and reveals a measured, experimental stance. Most teams aren’t hands‑on yet: 56% say they’re not currently using AI in their campaigns. Among adopters, the lion’s share of usage sits with customer insights and data analysis (21.21%), while content creation (11.36%), email personalization (6.06%), and visual content/design (5.30%) follow. Notably, the most cited measurable benefit so far is faster content production (15.38%), but two‑thirds (66.67%) of AI users haven’t quantified results at all. That measurement gap explains why many teams remain “in planning/research” (56%) or confine tests to one or two people (25%). Published September 22, 2025, the analysis also notes that 63.33% of respondents are satisfied with their current tools, signaling that AI must prove incremental value—not just novelty—to displace incumbents. PetfoodIndustry
What this means in practice: pet food marketers are rightly leading with low‑risk, high‑learning use cases—audience discovery, insight generation, and content acceleration—before ceding high‑stakes tasks like pricing, offer eligibility, or claim language to automation. That order of operations is healthy. But without clear KPIs and guardrails, teams risk running “toy pilots” that don’t survive procurement scrutiny. The goal of the next 90 days is to turn curiosity into confidence.
A 90‑Day AI Marketing Plan (built for pet food)
Days 1–15: Foundation & guardrails
- Define one business question per pilot. Examples: “Can we lift repeat purchase among sensitive‑stomach dog owners by 10%?” or “Can we cut cost‑per‑add‑to‑cart by 12% in cat treats?” Tie each question to an owned KPI and a decision (what will we do differently if it works?).
- Data hygiene sprint. Ensure key consented first‑party signals (species, lifestage, diet attributes, basket size, retailer channel) are captured, deduped, and mapped to stable IDs. Bad inputs ≈ bad models—the September and August polls both point to data quality and expertise as material hurdles. PetfoodIndustry+1
- Risk & review. Create a lightweight “human‑in‑the‑loop” rubric for brand and regulatory claims (e.g., nutrition benefits) and for image usage. Keep a log of AI‑assisted outputs and the approving human—fast to implement, invaluable when questions arise.
Days 16–45: Pilot two complementary use cases
- Insight accelerator: use an AI clustering/modeling tool to surface high‑value segments (e.g., households buying digestive‑care SKUs + wet toppers + purchasing monthly). Success = new segment definitions with actionability (size, expected LTV, channel reach).
- Content velocity + personalization light: generate 10–20 creative variants (copy and static visuals) tailored to 2–3 segments. Keep claims conservative and on‑label. Pair with rules‑based personalization (offer type, hero benefit) rather than fully generative decisioning.
Days 46–90: Test, measure, decide
- Design clean experiments. Hold out geo‑regions or audiences. Pre‑register success thresholds (e.g., +8% CTR, +5% conversion, +10% repeat within 60 days) so you can shut down under‑performers without politics.
- Instrument everything. Even September’s early adopters struggled with measurement: 66.67% hadn’t quantified results. Don’t add to that statistic—wire up content‑production cycle time, review rejection rate, and per‑asset cost alongside media metrics. PetfoodIndustry
- Scale or shelve. If the pilot clears thresholds, turn it into a playbook (briefing checklist, prompt library, segment taxonomy). If not, document findings and move on; sunk‑cost fallacy is the enemy.
Where to focus first (given the polling)
- Data over dazzle. The September breakdown shows the industry is using AI most for insight work (21.21%). Double down there: use modeling to find “look‑alike loyalists,” predict trial‑to‑repeat, or identify churn precursors for autoship. PetfoodIndustry
- Content speed with control. Since content production speed is the most commonly measured benefit (15.38%), pair generative tools with strict brand templates and nutrition‑claim libraries to reduce rework. Track “time‑to‑first‑draft” and “approval‑to‑publish latency” as operational KPIs. PetfoodIndustry
- Sequenced personalization. Move from rules‑based (if cat‑indoor → emphasize hairball control) to model‑scored (probability of repeat within 30 days) only after you can evidence uplift and manage explainability. This staged approach mirrors the industry’s measured stance in both the August and September polls. PetfoodIndustry+1
Common pitfalls—and how to avoid them
- Starting with the hardest problems. Teams sometimes jump straight to dynamic pricing or real‑time offer orchestration. The polling suggests the leaders who are getting traction begin with analytics and content acceleration, not fully automated decisioning. PetfoodIndustry
- Under‑investing in talent. The August survey cites expertise gaps as the No. 1 barrier. Budget for a hybrid profile—data‑savvy marketers or marketing‑savvy analysts—before you add more tools. PetfoodIndustry
- Neglecting data quality. If ingredient, benefit, and lifestage taxonomies aren’t consistent across ecommerce, PIM, and CRM, personalization will misfire. The August data flags data quality issues for ~10% of companies; treat taxonomy clean‑up as part of your AI budget. PetfoodIndustry
A note on expectations
The September poll shows 63.33% of marketers are content with current tools; AI has to earn its place. That doesn’t require moonshots. A sustained few points of lift in repeat purchase for a strategic sub‑segment, or a 20–30% cycle‑time reduction in content production, will more than justify continued investment—provided you can measure it and keep brand safety tight. PetfoodIndustry
Bottom line: The pet food sector is behaving rationally. It’s testing AI where the economics are clearest (marketing) and the surface area of risk is manageable, while it builds the skills and data foundations needed for deeper transformation. Treat the next 90 days as a proof‑building sprint: one insight‑led pilot, one content‑velocity pilot, rigorous measurement, and a decision. Do that, and you move from cautious to confident—without skipping the steps that keep your brand and consumers safe. PetfoodIndustry+1
