The Meal Kit & Ready-Meal Delivery AI Visibility Index 2026

Which meal-delivery brands the AI engines actually recommend — and which ones are cited from a category that no longer exists.

Executive Summary

5W tested 62 purchase-intent prompts across five AI platforms — ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. The finding: the brands winning AI visibility in meal delivery are increasingly not the brands that built the meal-kit category.

• Freshly is still recommended three years after being shut down. Nestlé closed the brand in early 2023. AI engines continue to surface Freshly on "ready-made meal delivery" queries in 2026 — a case study in how long a shuttered brand persists in the entity graph the models retrieve from.
• Counter is classified as GLP-1-relevant without positioning itself that way. The Macrofy retail brand — 30g of protein per bowl at under 400 calories, sold at Target, Kroger, Costco, and Lidl — does not market itself as a GLP-1 product. AI engines have decided it is one anyway, based on nutritional profile alone.
• Blue Apron holds significant citation share on queries the brand no longer serves. Wonder, the current owner, has repositioned Blue Apron as a prepared-meal service. AI engines continue to cite Blue Apron on meal-kit queries because a decade of editorial coverage sits in the training data.
• Nestlé and Conagra now compete on meal-delivery queries from the frozen aisle. Vital Pursuit (Nestlé) and Healthy Choice On Track (Conagra) — both marketed as GLP-1-friendly, both distributed nationally through frozen-food channels — now appear on "cheapest GLP-1 friendly meals" and adjacent prompts. Neither runs a subscription. Neither needs to.

Underneath those findings is a category story: meal kits, ready-to-heat prepared meals, and GLP-1 nutrition are no longer one business. AI engines are increasingly sorting them into separate answer sets. HelloFresh — the category's public-market proxy — posted 2025 revenue of €6.76 billion, a 9% constant-currency decline, and guided another 3–6% drop for 2026. Its U.S. growth engine, Factor, launched in 70+ Target stores across ten Midwest states in January 2026. Meal kits are contracting inside the very company that built the category.

One in eight U.S. adults is now on a GLP-1 medication. Every incumbent has bolted on a "GLP-1 Balance" filter. GLP-1-native brands — Clean Eatz Kitchen, Counter, MealPro, BistroMD — are competing directly for citation share on the highest-intent queries in the category.

5W's positioning: the brands building AI visibility today are competing for category leadership in 2028. Winning citation share requires the entity infrastructure, third-party editorial coverage, and answer-shaped content that AI engines actually retrieve — a discipline 5W calls AI Communications.

The Numbers

• -9% — HelloFresh 2025 revenue decline, constant currency
• -53% — HelloFresh one-year total shareholder return through Q1 2026
• €6.76B — HelloFresh 2025 revenue, down from €7.66B in 2024
• 1 in 8 — U.S. adults now on a GLP-1 medication (KFF, Nov 2025)
• 100+ — Weekly Factor menu options after Sept 2025 expansion, up from ~50
• 70+ — Target stores carrying Factor across ten Midwest states, Jan 2026
• 27 — Brands in the tracked universe across six sub-categories
• 62 — Consumer-intent prompts across five AI platforms in Q2 2026

The Leaderboard: AI Citation Share by Tier

Brands are grouped into tiers reflecting relative frequency of appearance in AI-generated answers to 62 consumer-intent prompts across ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews in Q2 2026. Tier bands are directional.

Tier I — Category Leaders (surfaced on 40%+ of prompts)

#1 HelloFresh (Cook-from-Scratch) — The category default in AI answers to generic "best meal kit" queries. A decade of editorial coverage sits in the models' training data. Publicly traded (XTRA:HFG); 2025 revenue €6.76B, down 9% constant currency. Citation share sits above the current commercial trajectory — a legacy asset that newer queries are already reallocating.

Tier II — Strong Category Presence (20–40%)

#2 Factor (Ready-to-Heat) — HelloFresh's U.S. ready-to-heat brand. The fastest-rising incumbent. Menu expanded from ~50 to 100+ weekly meals in September 2025. Launched in 70+ Target stores across ten Midwest states in January 2026. Cited across ready-to-heat, retail-available prepared meals, and GLP-1 queries — the widest cross-query footprint of any brand in the index.

#3 Blue Apron (Ready-to-Heat/Legacy Kit · Wonder) — Legacy citation share disproportionate to current commercial footprint. Repositioned by Wonder as a prepared-meal service; AI engines still cite it on meal-kit queries. Recognized by Consumer365 in Feb 2026 as "Best Prepared Meal Delivery Service." 600M+ meals delivered lifetime.

#4 CookUnity (Ready-to-Heat) — The chef-driven premium ready-to-heat leader. 50+ contracted chefs, $10.99–$13.99 per meal. Dominates "chef-made," "restaurant-quality," and premium-tier prepared-meal queries. Canadian expansion May 2025. Heavy citation by NBC Select, MealFan.

#5 Home Chef (Cook-from-Scratch/Hybrid · Kroger) — Owned by Kroger. The widest cross-category footprint outside Factor — cited on meal-kit, oven-ready, family-plan, and grocery-adjacent queries. Kroger integration remains under-leveraged in the entity signals AI engines pull from.

Tier III — Specialist Presence (10–20%)

#6 Green Chef (Diet-Specific) — HelloFresh-owned. Dominates keto, paleo, Mediterranean, and mostly-organic queries. Growing citation share on GLP-1-adjacent "high-protein meal kit" prompts.

#7 Sunbasket (Diet-Specific) — The Mediterranean and diabetes-friendly leader. Widest diet-filter set in the category. Cited on health-condition-specific queries where mass-market brands don't surface.

#8 EveryPlate (Family & Value) — HelloFresh's budget brand. Owns "cheapest meal kit" and "meal kit under $5" queries. AI engines correctly identify it as value-tier — rarely surfaces on quality or health queries.

#9 Purple Carrot (Diet-Specific/Plant-Based) — The vegan-native brand. Dominates "plant-based meal delivery" and "vegan meal kit" queries. A specialist that AI engines slot into a specialist lane.

#10 Dinnerly (Family & Value) — Marley Spoon's budget brand at $4.99–$5.99 per serving. Frequently paired with EveryPlate in "cheapest meal delivery" answers.

Tier IV — Niche & Emerging Presence (under 10%)

#11 Trifecta — Athlete and macro-tracking specialist; cited on high-protein and GLP-1 protein-preservation queries.
#12 Freshly — Legacy/zombie citation. Shut down by Nestlé in early 2023; AI engines still surface it on "ready-made meal delivery" queries with no commercial owner.
#13 Marley Spoon — ASX-listed, Berlin-based; Martha Stewart partnership drives most consumer-facing citation share.
#14 Hungryroot — Grocery-plus-recipes hybrid; cited on "AI-powered grocery service" queries.
#15 Gobble — The 15-minute meal-kit brand; owns a narrow but defensible "quick meal kit" query surface.
#16 Territory Foods — Local-chef-driven ready-to-heat network; high citation-per-market density in seed markets.
#17 Tempo — Growing citation share on GLP-1 and high-protein queries.
#18 Fresh N Lean — Organic, gluten-free, plant-based, keto, paleo; cited on "organic prepared meals."
#19 Daily Harvest — Smoothies, harvest bowls, frozen; residual reputational drag from 2022 recall coverage still surfacing in the entity graph.
#20 Thistle — Plant-forward, West Coast-native; cited on "plant-based prepared meal delivery."
#21 Mosaic Foods — Frozen plant-based bowls and pies; defensible niche on "frozen vegan meals."
#22 Clean Eatz Kitchen — Subscription-free GLP-1-native brand, $8.99/meal starting price; fast-rising citation share on GLP-1 queries.
#23 Counter (Macrofy) — 30g protein per bowl under 400 calories, distributed through Target, Kroger, Costco, Lidl; classified as GLP-1-relevant by AI engines without marketing itself that way.
#24 BistroMD — Physician-designed weight-loss meal delivery; cited on "doctor-designed meal delivery" queries.
#25 MealPro — Fitness-and-macros native; cited on "macro-tracked meal delivery" and GLP-1 protein queries.
#26 Vital Pursuit (Nestlé) — Nestlé's dedicated GLP-1 frozen line, launched May 2024; cited on retail-adjacent GLP-1 queries.
#27 Healthy Choice On Track (Conagra) — Conagra added "GLP-1 Friendly" labels to 26 existing meals with six more planned; surfaces on "cheapest GLP-1 friendly meals" queries.

The Analysis: Five Key Findings

1. Zombie citation: Freshly is still being recommended three years after being shut down

Nestlé shut down Freshly in early 2023. AI engines continue to surface Freshly on "ready-made meal delivery" and "prepared meals delivered" queries in Q2 2026. The brand still exists in the entity graph the models retrieve from — a decade of editorial coverage, product reviews, and consumer discussion did not disappear when Nestlé stopped shipping product. The buyer types the query, gets Freshly in the answer, clicks the link, and finds nothing.

The operator lesson: a brand's presence in AI answers persists long after its commercial presence does. Retiring a brand from the AI graph requires the same discipline as building one — press coverage of the shutdown, updates to Wikipedia and Wikidata, structured signals from the parent company, and often years of natural decay. Every operator with a legacy brand or planned shutdown needs a de-listing plan, not just a de-marketing plan.

2. Reverse-classification: Counter is being ranked as a GLP-1 brand without positioning itself that way

Counter, made by Macrofy, does not market itself as a GLP-1 product. The brand emphasizes protein density (30 grams per bowl), retail distribution (Target, Kroger, Costco, Lidl), and calorie profile (under 400 calories per meal). AI engines are surfacing Counter in answers to "best frozen meals for Ozempic," "high-protein GLP-1 meals," and adjacent queries anyway — because the nutritional profile matches the query intent better than most brands that do buy the label.

The operator lesson: the AI answer layer classifies brands by the intent-matching content and coverage the models retrieve, not by the brand's own positioning. Owning a category label buys less than earning coverage that matches the queries in that category. The goal is to be classified by the machine as the answer to a specific question.

3. Legacy citation share masking commercial decline: Blue Apron and HelloFresh core

Two of the highest-ranked brands in this index — HelloFresh (Tier I) and Blue Apron (Tier II) — hold citation share that reflects a decade of editorial coverage rather than current commercial trajectory. HelloFresh revenue fell 9% in 2025 and is guided to fall another 3–6% in 2026. Blue Apron has been acquired and repositioned as a prepared-meal service by Wonder, meaning much of its citation share routes buyers to a category the brand no longer defines.

The operator lesson: AI citation share reflects accumulated coverage plus current retrieval. When the two disagree — a shrinking commercial footprint against an intact editorial legacy — the citation share becomes a lagging asset that decays as newer coverage displaces it. Legacy leaders in any category should read this as a warning: entity strength has a half-life without ongoing investment.

4. The frozen aisle is competing on delivery queries

Vital Pursuit (Nestlé) and Healthy Choice On Track (Conagra) are marketed as GLP-1-friendly, distributed nationally through frozen-food channels, and priced far below subscription delivery. Both now surface in AI answers to "cheapest GLP-1 friendly meals," "GLP-1 frozen meals," and adjacent queries. Neither runs a subscription. Neither needs to.

The operator lesson: the category boundary the delivery brands assumed — subscription delivery on one side, frozen retail on the other — no longer holds inside AI answers. The frozen aisle has more national distribution, lower prices, and increasingly the labeled positioning to compete for the same query intent. Delivery brands treating retail as a separate category miss the actual competitive set the machine is showing consumers.

5. Editorial and review outlets are the primary citation-source layer

Across all five platforms, the citation sources the models pulled from were dominated by editorial and review outlets: Consumer Reports, Wirecutter, NBC Select, Good Housekeeping, MealFan, CNET, Reviews.com, Eatingwell. Brands cited most frequently by these outlets — Factor, CookUnity, HelloFresh, Home Chef — sit at the top of the leaderboard. Brands with thinner editorial footprints appear less often regardless of product quality or marketing spend.

The operator lesson: earned coverage in the outlets the models trust, at volume and sustained, is the mechanical lever that moves AI citation share. This is the discipline of AI Communications — building the citation footprint that becomes the retrieval layer.

The Board: Winners and Losers

Winners

Factor — the incumbent that pivoted early. The reason HelloFresh still has a growth story. Menu expansion from ~50 to 100+ weekly meals in September 2025. Retail expansion into 70+ Target stores across ten Midwest states in January 2026. Cited across three separate query buckets — ready-to-heat, retail-available prepared meals, and GLP-1 nutrition. Widest cross-query footprint in the index.

CookUnity — the premium ready-to-heat category-definer. Owns "chef-made" as a query category in AI answers. 50+ contracted chefs. Consistently cited by NBC Select, MealFan, and category publications the engines pull from. Canadian expansion in May 2025 extended the citation surface internationally.

Clean Eatz Kitchen — the GLP-1 native that outflanked the incumbents. Every other GLP-1 delivery option requires weekly auto-ship — a friction GLP-1 patients cannot tolerate. Clean Eatz ships flash-frozen with no subscription, starting at $8.99/meal. That single differentiator earned it citation share on "best meal delivery for Ozempic" while the incumbents were still bolting a filter onto existing plans.

Counter — retail-native winning on protein density. 30g of protein per bowl under 400 calories, distributed through Target, Kroger, Costco, and Lidl. Does not market itself as GLP-1 — AI engines have classified it as one anyway based on nutritional profile alone. When the entity graph decides your category for you, positioning is what you have already been indexed as.

Losers

HelloFresh (parent brand, not Factor). Legacy citation share against deteriorating fundamentals. Revenue down 9% in 2025, guidance down another 3–6% in 2026, exits from Spain and Italy, share price down 94% over five years. Newer queries — GLP-1, ready-to-heat, chef-driven — cite Factor and CookUnity, not the parent brand.

Blue Apron (as a meal-kit brand). The clearest example of citation share disconnected from current business. Wonder has repositioned Blue Apron as a prepared-meal service — but AI engines still cite it on meal-kit queries. The entity in the graph needs to be actively rebuilt, or the citation asset keeps routing buyers to a category the brand no longer serves.

Sunbasket — outflanked in its own niche. Broadest diet-filter set in the category, losing citation share to specialists. Purple Carrot outcites it on vegan. Trifecta outcites it on performance. Green Chef outcites it on organic keto. Clean Eatz outcites it on GLP-1. Breadth is not a defensible entity property in AI search; specificity is.

EveryPlate and Dinnerly — the value trap. Both surface on "cheapest meal kit" queries and virtually nowhere else. AI engines correctly identify them as value-tier, which caps the query surface they can compete on. No path to premium repositioning inside the AI graph without a new brand.

The Playbook: Ten Actions

1. Audit citation share across ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews on the top 50 consumer-intent queries in your sub-category. Baseline before you optimize.
2. Rebuild the entity graph. Ensure every AI engine can identify what you sell, who it is for, and where you fit — through structured data, Wikipedia, Wikidata, and the third-party outlets the models retrieve from.
3. Prioritize earned coverage in the outlets that populate the AI retrieval layer — Consumer Reports, Wirecutter, NBC Select, Good Housekeeping, MealFan, Reviews.com, CNET, Eatingwell.
4. Own one sub-category cleanly before competing in another. Specialists outcite generalists in every sub-category measured in this index. Positioning is a citation strategy.
5. Publish answer-shaped content — FAQ schema, entity-rich comparison pages, primary-source data — on the exact questions buyers type into AI engines. The engines retrieve answers, not brochures.
6. For GLP-1 targeting, redesign the buying experience, not just the menu filter. Subscription flexibility, protein density, portion transparency, and no-waste positioning are the citation-driving differentiators.
7. Track competitor citation share monthly. Category dynamics in AI search move faster than in traditional search — a product launch or partnership can shift the leaderboard in weeks.
8. For legacy brands, treat the entity graph as a communications asset. Blue Apron, HelloFresh, and Freshly illustrate what happens when the brand's AI representation lags the commercial reality.
9. For emerging brands, pursue citation share aggressively in queries the incumbents have not yet claimed. Category-native positioning plus earned coverage volume plus entity clarity displaces incumbents faster than paid media.
10. Ship every product launch, retail partnership, and executive move as a distribution kit built to be cited — press release, byline, third-party editorial pitch, structured data update, primary-source data asset.

FAQ

What is AI citation share, and how is it different from market share?
AI citation share is the frequency with which a brand is named in AI-generated answers to consumer-intent queries, relative to competitors. Market share measures revenue or units sold. Citation share measures visibility inside the answer engines that consumers now use for product research — ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. The two can move independently: HelloFresh leads this index while its revenue and share price decline.

Which AI platforms were tested?
Five platforms in the same fixed time window (Q2 2026): ChatGPT (OpenAI), Claude (Anthropic), Perplexity, Gemini (Google), and Google AI Overviews. All were queried using the same 62-prompt consumer-intent query set spanning six sub-categories.

Why is HelloFresh ranked highest if its revenue is declining?
Citation share reflects the accumulated editorial coverage AI engines have retrieved from the open web. HelloFresh has a decade of coverage in Consumer Reports, Wirecutter, NBC Select, and dozens of other outlets the engines pull from — a legacy citation asset that outweighs current commercial trajectory on generic queries. Newer, higher-intent queries (GLP-1, ready-to-heat, chef-driven) cite Factor and CookUnity, not HelloFresh.

How can a meal-delivery brand improve AI citation share?
Four levers: (1) earned coverage in the outlets AI engines retrieve from; (2) entity infrastructure — Wikipedia, Wikidata, structured data, category-clean positioning; (3) answer-shaped content — FAQ schema, entity-rich comparison pages, primary-source data; (4) category specificity — own one sub-category cleanly before competing across the whole space.

Methodology

The 5W Meal Kit & Ready-Meal Delivery AI Visibility Index 2026 measures AI citation share — the frequency with which a brand is named in AI-generated answers to consumer-intent queries, relative to competitors — across five platforms in Q2 2026.

Brand universe: 27 brands representing the tracked category — HelloFresh, Factor, Blue Apron (Wonder), Home Chef, EveryPlate, Green Chef, CookUnity, Territory Foods, Tempo, Trifecta, Sunbasket, Purple Carrot, Marley Spoon, Dinnerly, Gobble, Hungryroot, Daily Harvest, Thistle, Mosaic Foods, Fresh N Lean, Clean Eatz Kitchen, BistroMD, MealPro, Counter (Macrofy), Vital Pursuit (Nestlé), Healthy Choice On Track (Conagra), and Freshly (retained in the universe as a study case for post-shutdown citation persistence).

Sub-categories: Six 5W classifications for analytical purposes — (1) Meal Kits, Cook-from-Scratch; (2) Ready-to-Heat Prepared Meals; (3) GLP-1 & Metabolic Health; (4) Family & Value; (5) Diet-Specific & Performance; (6) Grocery-Adjacent & Hybrid.

Query set: 62 consumer-intent prompts distributed across the six sub-categories, drawn from real buyer search language, including "best meal kit for beginners," "HelloFresh vs Home Chef vs Blue Apron," "best ready-made meal delivery," "Factor vs CookUnity," "best meal delivery for Ozempic," "high-protein meals on Wegovy," "GLP-1 friendly frozen meals," "cheapest meal delivery for families," "keto meal delivery," "vegan prepared meals," "chef-made meal delivery," "AI-powered grocery service," and "healthy frozen meals delivered."

Testing protocol: Each prompt was submitted in clean, logged-out sessions in U.S. English, from U.S. IP addresses, with no personalization signals. Prompts were run as single-pass queries within the Q2 2026 window. Brand mentions within a single AI response were deduplicated. Ranking position within a response was not weighted.

AI Overviews coverage: Google AI Overviews did not surface for every query type. Where AI Overviews did not appear, the response was recorded as unavailable rather than empty, and citation share for that prompt was calculated across the remaining four platforms.

Citation share calculation: A brand's share of total brand-mentions across all tracked prompts and platforms, expressed as a tier band. Brands were sorted into four tiers reflecting frequency of appearance: Tier I (40%+), Tier II (20–40%), Tier III (10–20%), Tier IV (under 10%). Ordinal ranking within each tier reflects observed frequency in the run. Precise per-brand percentages were not published because prompt-level variation across platforms and the single-pass protocol produce meaningful directional signal but not measurement-grade precision.

Citation source layer: Aggregated across all five platforms, the primary citation sources for the category were editorial and review outlets (Consumer Reports, Wirecutter, NBC Select, Good Housekeeping, CNET, Reviews.com, MealFan, Eatingwell); user communities (Reddit meal-kit and GLP-1 subreddits); nutrition and clinical outlets (NPR, KFF, published research); business press and SEC filings for public brands (Bloomberg, Reuters, Investing.com, Simply Wall Street, Yahoo Finance); and brand-owned publications and product pages. The source mix varied by sub-category — GLP-1 queries pulled more heavily from clinical and nutrition sources; meal-kit queries pulled more heavily from editorial reviews.

Important framing: 5W estimates in this index are based on frequency of brand mention across the tracked prompt set and platforms in the stated time window. These are directional measures of relative visibility in AI-mediated consumer research, not precise market measurements. The index does not measure meal quality, safety, nutritional appropriateness, customer outcomes, or fit for any individual buyer. Consumers should consult healthcare providers before making decisions related to GLP-1 medications or dietary changes. This index is published for communications strategy purposes only.

Seventh installment in the 5W AI Visibility Index series, following the Dating App, Legal Tech, Real Estate, Fintech, Weight Loss & Metabolic Health, and Pet Industry indexes.