Nukebook
Market & Site Intelligence
Benchmark, August 2026

The state of AI answers in CPG snacks.

One category, measured properly: how often the answer exists, whom it names, whom it cites, and where the doors are.
If you searched "AI answers in CPG"

What we measured about it: 57 questions read twice: presence 15 to 17 of 40, small sites at 32% of citations, doors mapped.

Run on your market, a read shows: this exact benchmark, for your category, plus your real search demand under every phrasing, which AI answers in your market are open, partial or closed (with the evidence), what your competitors run in the public ad archives, and your break-even numbers computed from your own economics.

Book the 20-minute callor read the measured pieces first →

We read 57 questions category and occasion questions in the US snacks market and read every AI answer twice, 30+ minutes apart. An AI Overview appeared on 15 to 17 of 40 of the SERP-checked questions across the two pulls. measured

Whom the answers name and cite

The incumbent substitute (jerky) was named in 17 answers, while challenger categories went largely unnamed: the answer box defaults to the category the model already knows. Citations tell the openness story: small sites held 32% of citations in this market overall, and on individual questions the door swings wide: the "keto snacks" answer drew 60% of its citations from small sites. For contrast, the adjacent plant-based market showed small sites holding just 11%: same rules, different door.

The doors

Classified by fixed rules (open = no AI answer on two pulls; partial = the answer lacks decision figures or majority-strong citations; closed = complete and majority-held by strong sites), the CPG snack run produced 6 open, 12 partial, 3 closed doors. The pattern that matters for any brand in the category: the biggest-volume questions are closed (complete, authority-cited answers), while the occasion and diet questions one step out are open or figure-poor. The demand is not where the fight is loudest.

Method: full market run 2026-08-18, DataForSEO SERP endpoints, two-pull AIO protocol, citation classes from a fitted CPG taxonomy, door rules as stated above. Figures are market-level; no client data appears. Figure registry: /figures/state-of-ai-answers-cpg.figures.json.
Want this measured for your market?

One read: your demand, your winnable AI answers, every competitor's ads, and your site's catch rate. Price and scope on a short call, plainly.

Book a call or see what a read contains →