How AI Search Is Recommending Sustainable Pet Food and Alternative Protein: Monthly Trends

Executive Summary

Open Farm remains the category leader in September 2026 with a 66.8% valid recommendation coverage rate, up 1.0 point from 65.8% in August 2026, extending its lead atop the category since the baseline month. The gap to second-place Freshpet widened to 42.3 points, as Freshpet's coverage moved down 1.9 points to 24.5% from 26.4% in August. The category retains a clear two-tier structure: Open Farm and Freshpet hold the majority of recommendation share, while the remaining eight tracked brands cluster between 0.2% and 13.1%.

V-Dog posted the largest point gain this month, moving up 2.1 points to 13.1% coverage, keeping it in third place ahead of Wild Earth (12.3%, up 0.3 points) and Petaluma (5.5%, up 0.6 points). Petaluma and Earth Animal also gained modestly, while Jiminy's (down 0.5 points to 0.7%), PawCo Foods (down 0.6 points to 4.3%), and Freshpet moved down.

September was a quiet month by the benchmark's own measurement standard: none of the ten tracked brands moved beyond normal month-to-month variation on the primary coverage metric. The one placement-level shift worth watching is Open Farm's rank-one recommendation rate, which moved from 14.9% in August to 8.6% in September — the largest point change among the placement metrics tracked this month.

Each monthly run begins with 800 prompt-surface observations (580 unique questions in August; 510 in September) across the benchmark's defined AI/search surface universe. Of those, 800 mentioned a tracked brand or competitor in August and 799 in September; 600 were relevant and 200 were irrelevant in August, versus 615 relevant and 184 irrelevant in September. The public metrics use the 409 qualified observations (August) and 421 qualified observations (September) that survive both qualification stages, spanning all six AI surface families: ChatGPT, Copilot, Gemini, Perplexity, AI Overviews, and AI Mode.

Key Findings

Benchmark Context

The report separates the raw collection universe from the qualified analysis set. Brand-level recommendation percentages are calculated within the qualified benchmark set, which in September 2026 was 421 observations from a collection of 800 prompt-surface runs.

Qualified surface breadth held at 6 AI surface families in both months, meaning the funnel narrowed the observation count without narrowing the range of surfaces measured.

Benchmark-Level Metrics

The share of responses taking a recommendation-shaped form rose from 48.4% in August to 52.7% in September, meaning AI systems were more likely to produce recommendation-oriented output this month. The valid shortlist share stayed effectively flat at 77.2%, indicating the qualification structure remained consistent.

AI Recommendation Trend

Recommendation Coverage by Brand

No brand exceeded normal month-to-month variation on the primary coverage metric in September. The month's movement came instead from the combination of several smaller shifts: V-Dog's gain of 2.1 points, Freshpet's decline of 1.9 points, and smaller movements among the mid-tier brands. Open Farm's 6.3-point move in rank-one placement did not translate into a coverage decline because the brand remained widely present in recommendation shortlists.

What Changed This Month

Open Farm: Leadership Steady, Rank-One Placement Rate Moves Lower

Open Farm's valid recommendation coverage rose 1.0 point from 65.8% in August to 66.8% in September, keeping its lead atop the category. Raw mention presence also increased from 75.3% to 78.9%, meaning the brand appeared in some form in nearly four out of five qualified observations.

The more notable shift was in placement quality. Open Farm's rank-one recommendation rate moved from 14.9% in August to 8.6% in September, a 6.3-point difference and the largest change among the placement metrics tracked this month. Top-three rate also eased from 29.8% to 24.9%. The brand's average recommended rank moved from 3.0 to 3.3.

The distinction to notice is between presence and primacy. Open Farm is still the most recommended brand overall, but the current data records it appearing less often in the rank-one position when it is recommended. The benchmark cannot establish why that placement pattern shifted.

Highest-priority diagnostic: Which prompt types or surfaces are associated with the rank-one placement change, and which brand or brands occupy the top spot when Open Farm does not?

Freshpet: Coverage Moves Lower as Rank-One Rate Improves

Freshpet's valid recommendation coverage moved down 1.9 points from 26.4% in August to 24.5% in September, the sharpest decline among tracked brands this month. Its raw presence also moved down from 36.7% to 34.7%, a 2.0-point difference. Valid recommendation count decreased from 108 to 103 qualified observations.

A countervailing signal appeared in placement: Freshpet's rank-one rate moved up from 2.9% to 3.8%, and its rank-one count rose from 12 to 16. The brand's average recommended rank held close to its prior level at 3.3.

Freshpet is appearing in fewer recommendations overall this month, while placements that do occur are somewhat more likely to land first. The brand remains solidly in second place by coverage, 11.4 points ahead of third-place V-Dog.

Highest-priority diagnostic: Which question contexts reduced Freshpet's recommendation frequency, and were the additional rank-one placements concentrated on specific AI surfaces?

V-Dog: Largest Coverage Gain Among Tracked Brands

V-Dog posted the largest point gain in September, moving from 11.0% coverage in August to 13.1% (up 2.1 points). Its valid recommendation count increased from 45 to 55 qualified observations, the largest absolute gain among all tracked brands after Open Farm.

Raw presence also rose from 13.2% to 14.0%, while top-three rate eased slightly from 8.6% to 8.3%. The brand's average recommended rank moved from 1.8 to 1.9, staying close to the top of the ranking scale. Rank-one count moved from 20 to 18.

V-Dog holds the clear third-place position in the current benchmark, 0.8 points ahead of Wild Earth and well clear of the rest of the mid-tier field.

Highest-priority diagnostic: Which prompt contexts or surfaces added the extra valid recommendations, and is the gain supported by evidence sources that can be reinforced?

Wild Earth and Petaluma: Modest Gains in the Mid-Tier

Wild Earth moved up 0.3 points to 12.3% coverage in September from 12.0% in August, with valid recommendation count rising from 49 to 52. Its raw presence moved down from 15.2% to 14.2%, and rank-one rate moved down from 2.7% to 1.4%.

Petaluma moved up 0.6 points to 5.5% from 4.9%, with valid recommendation count up from 20 to 23 and rank-one count increasing from 2 to 3. Raw presence rose from 5.4% to 6.2%.

Both brands remain well behind the Freshpet/V-Dog tier. Neither movement was large enough to break from the category's stable pattern this month, but both represent a continued small upward drift for mid-tier players.

Highest-priority diagnostic: Are Wild Earth's and Petaluma's gains concentrated in the same AI surfaces, or are they arriving through different recommendation pathways?

Buyer-Intent Interpretation

In September 2026, all 421 qualified observations fell into the brand recommendation cluster. The benchmark did not capture any qualified observations that asked for pricing analysis, value comparisons, or multi-brand head-to-head evaluation, despite 3 pricing analysis responses and 24 comparison analysis responses appearing in the raw response-type distribution.

This concentration means the public benchmark can currently answer which brand AI systems recommend in general terms. It cannot yet speak to price positioning, value perception, or comparative win rates between specific brands, because the qualified observation set did not include those commercial question types.

Brand Opportunity Summary

The benchmark identifies where attention is warranted; a company-level analysis is needed to explain why those patterns exist.

Evidence Behind the Benchmark

The aggregate metrics are built from prompt-level observations that capture the query, the AI surface that answered, whether a brand was recommended, the rank of that recommendation, the sentiment expressed, and any citations the system exposed. Company-level analysis can go deeper into the specific prompts where a brand wins or loses, which competitor takes the recommendation when a brand is absent, and which external sources are shaping the AI system's answer. Source presence in an AI response is not automatically treated as proof of causation.

About This Benchmark

This report is part of the CiteWorks Studio AI Industry Market Discovery research program.

• AI Industry Market Discovery
• AI Industry Market Discovery Methodology
• AI Industry Market Discovery Metrics
• AI Industry Market Discovery Standards

Report-Specific Interpretation Notes

• Brand-level percentages in this report are calculated against the qualified benchmark set of 421 observations, not the raw collection universe of 800 prompts. Small observation counts at the individual brand level (for example, 1 valid recommendation for Yora) mean percentage movement for those brands should be interpreted with caution.
• Open Farm's rank-one rate decline is reported as a placement-level change worth watching. It does not by itself establish a cause for that change.
• The benchmark has used six AI surface families throughout this series, and qualified surface breadth held at 6 in both August and September 2026.

Next Step

The Public Benchmark Shows Where a Brand Is Winning or Losing. A Company-Level Audit Shows Why.

Beneath the aggregate coverage percentage sit the questions that matter commercially: which high-intent prompts is a brand winning, which competitor takes the recommendation when a brand loses, what attributes do AI systems associate with each option, and which external sources are shaping those answers. A brand can hold steady overall coverage while its rank-one placement rate moves lower, as the September data shows for the category leader.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. It investigates why the placement pattern moved and what can be done about it, turning the benchmark signal into an action plan.

Request an AI visibility audit