Top sellers over-index on shopper friction: side effects in Women's Health (2.6x their peers), mixing in Energy & Performance (9.9x), value in Longevity & Cellular Health (20.5x) — and 0–2% of scanned titles answer any of them.
→
The next whitespace may not be a new ingredient. It may be the complaint, the search phrase or the delivery format the current shelf has not answered. In Women's Health the category's top sellers draw side-effect mentions at 2.6 times their peers' rate; in Energy & Performance, mixing complaints run 9.9 times; in Longevity & Cellular Health, value complaints run 20.5 times — and 0–2% of scanned titles answer any of them. Shoppers, meanwhile, are searching for pairings and delivery formats that few titles name, while the shelf keeps adding claims that consumer attention is not following. The leads are Amazon-led — listings, Amazon's own review summaries and modelled Amazon search estimates — and each was then checked against social listening, trade press, research and four LA grocery shelves.
→
→
→
The category's top sellers draw side-effect mentions at 8.5% of their review-aspect mentions; the rest of the scanned products draw them at 3.3%. No scanned product title answers the complaint. Across all products in the category, 'no effect' accounts for 14.3% of all aspect mentions with no title answer, and 'side effects' accounts for 7.2% category-wide. Tolerability and gentleness are therefore messaging territories worth testing — not confirmed gaps.
→ A founder entering Women's Health has a tolerability hypothesis worth testing: first in listing language, then on the front panel if the product experience substantiates it.
The category's top sellers draw texture/mixing mentions at 3.5% of their review-aspect mentions — 9.9 times the peer rate — and quality/authenticity at 2.8%, 18.8 times the peer rate. Neither complaint is answered by any scanned product title. The unanswered category-wide pain is 'no effect' at 5.9% of all mentions with no title coverage. Beyond Amazon — adjacent evidence: Reddit fitness communities show shoppers seeking creatine as a flavoured drink mix rather than plain powder (Jun 2026). The quality complaint has no outside corroboration.
→ A founder in Energy & Performance should verify the packaging, then test whether mixability or quality-verification language answers the complaints the category's top sellers are drawing.
The category's top sellers draw value/price mentions at 5.7% of their review-aspect mentions against 0.3% for the rest of the scanned products — 20.5 times the peer rate. Only 2% of scanned titles answer it. 'No effect' draws 9% of top-seller review-aspect mentions at 5.1 times the peer rate, with only 2% title coverage. The category's largest complaint, 'no effect' at 3.8% category-wide, is also minimally answered. Beyond Amazon — no outside corroboration of the value or efficacy complaints this cycle.
→ A founder in Longevity & Cellular Health with a verifiable efficacy story and transparent pricing rationale has two unanswered complaints to position against — verify whether either belongs in the title before testing.
'Magnesium ashwagandha gummies' drew 6,389 estimated monthly searches, up 83% in one month, with no scanned product title using that phrase. 'Magnesium ashwagandha l theanine' drew 5,846 searches, up 68%, also with zero title coverage. Search interest clusters around the pairing — magnesium with ashwagandha, often in a gummy — and no scanned title names it. One month of change is directional, not structural. Beyond Amazon — adjacent evidence: Instagram and TikTok posts promote 'ashwagandha + magnesium glycinate' as a pairing (Jun 2026) — the pairing circulates, though that does not measure the rising Amazon searches.
→ A founder in Adaptogens & Stress Support should verify the live shelf and test whether naming the combination in the title, front panel, or listing copy closes the gap.
'NAD patches' drew 14,312 estimated monthly searches, up 19%, with 4% of scanned titles using the phrase. 'NAD+ patches' drew 5,564 searches, up 54%, with 14% title coverage. 'NAD cream' drew 8,895 searches, up 18%, with only 2% title coverage. The format pattern — topical or transdermal NAD — recurs across all three searches. This is one month of change, so seasonal or event effects cannot be ruled out. Beyond Amazon — no outside corroboration: no social, trade, research or grocery-shelf signal mentions NAD patches or creams.
→ A founder in Longevity & Cellular Health should test whether a patch or cream delivery format belongs in the title and verify the regulatory path before committing.
'Curcumin boswellia' drew 5,515 estimated monthly searches, up 127% in one month, with no scanned title using the phrase. 'Turmeric boswellia supplement' drew 5,964 searches, up 117%; 10% of titles use the term, but the core phrase 'turmeric boswellia' appears in only 12% of titles. 'Curcumin + boswellia joint support' drew 6,551 searches, up 36%, with zero title coverage. Search interest clusters around the two-ingredient combination, not either ingredient alone. One month of change; directional only.
→ A founder in Anti-Inflammatory & Metabolic should verify the live shelf and test whether naming the boswellia combination explicitly in the title or listing copy captures search that current titles are missing.
Share moved from 23.8% to 34.7% between the Sep 7 and Sep 29 scans. Mentions per product rose 10%, and at 24 points clear of the cohort median on mentions per product, the move distinguishes itself from peer claims. Most of the share shift reflects other top claims retreating. Consumer attention sits at 3.05% of recent signals against a prior baseline of 4.69% — under-indexing against its own history. Research base: deep — 47 independent signals, plus 39 rows from our own shelf scrape set aside separately. Beyond Amazon — direct corroboration: SupplySide coverage (May 2026) reported retail buyers seeking dual-function mushroom blends, and mushroom-blend products sit on grocery shelves scanned Sep 29, including a mass-banner private label.
→ Brands are stacking the 'Mushroom Complex' claim onto the shelf while recent consumer attention under-indexes its prior baseline — a crowding warning before committing to that language.
Mentions per product rose just 2% — 16 points clear of the cohort median on that measure, so the move is distinguishable. Share of the top five claims moved from 25.7% to 37.1%, but most of that shift reflects other top claims retreating rather than the shelf adding this one. Consumer attention is flat: 5.84% of recent signals against a prior baseline of 5.91% — tracking its own history rather than rising. Research base: 26 independent signals (91 derived shelf rows, which are counted separately and already excluded). This is relative crowding rather than strong absolute claim growth.
→ The shelf is labelling into a claim with flat consumer attention — a founder entering this space should test whether the claim is table stakes or clutter.
Share moved from 16% to 21%. Mentions per product rose 13%, at 27 points clear of the cohort median on mentions per product. Consumer attention sits at 10.15% of recent signals against a prior baseline of 13.19% — under-indexing against its own history. Research base: deep — 44 independent signals, plus 16 shelf rows set aside. Beyond Amazon — adjacent evidence: cortisol was loud in spring social (TikTok 'cortisol-lowering' trends, May 2026), the high baseline attention now under-indexes against.
→ The claim is appearing on more products while recent consumer attention under-indexes its prior baseline — verify whether it belongs in the title before investing in it.
GLP-1 sits at the top third (99th percentile) on the shelf and top third (92nd percentile) on attention, but buying is directional at the bottom third (27th percentile) against category peers, based on modelled Amazon keyword-sales estimates. Search is in the middle third (65th percentile). The shape is a crowded shelf with buying that under-indexes: a claim printed widely without matching buying behind it. Beyond Amazon — direct corroboration of the crowded shelf: GLP-1-positioned hydration, electrolyte and probiotic launches in trade and research coverage, including a GLP-1 'booster' capsule on Sep 7. The buying half is Amazon-only.
→ A founder considering GLP-1 language in Longevity & Cellular Health faces a shelf already saturated with the term and buying that under-indexes — verify whether the claim belongs in the title or only in supporting copy.
The retail shelf scan for Women's Health shows 82% capsules, with liquid under 10% of the shelf and powder, gummy, beverage, and softgel absent from scanned titles. Gummy-related signals sit 4.02 points above the all-category baseline; beverage-related signals sit 3.67 points above baseline — both elevated. This is one retailer's shelf and one consumer-signal pool; it is a hypothesis about a gap, not confirmed whitespace. Beyond Amazon — adjacent supply evidence: a women's-health gummy brand raised a $20M Series B (Mar 2026), a women's hydration brand relaunched its platform (Jun 2026), and a botanicals brand released a women's liquid shot (May 2026) — supply-side format activity, not a measure of the consumer-format signals above.
→ A founder considering a gummy or beverage format in Women's Health should verify the live shelf across channels and test whether format language belongs in the title, front panel, or listing copy before committing.
The retail shelf scan for Skin & Beauty from Within shows 90% capsules with powder present at under 10% of the shelf; gummy, beverage, liquid, and softgel are absent from scanned titles. Beverage-related signals sit 8.25 points above the all-category baseline — the highest positive elevation in the format data. Gummy signals sit 2.30 points above baseline. This is one retailer's shelf only. Beyond Amazon — direct corroboration: a collagen drink mix and a collagen sparkling water sit on grocery banners scanned Sep 29, and a major collagen brand launched ready-to-drink collagen beverages.
→ A founder in Skin & Beauty from Within should test whether a beverage or gummy format is feasible for the category's active ingredients before reading this as a confirmed opportunity.
→
→
→
→
→
260 product listings across 13 categories (Sep 29), Amazon's own review summaries on 219 of them, and modelled Amazon search estimates across about 134,000 search terms in 17 search sets.
→
SOURCE: Aterra Signal Library
Instagram and TikTok (2,491 signals), Reddit (101 posts and 1,625 brand-thread signals), YouTube creator analyses (287) and agent social-listening reports (745). Amazon review imports are excluded from this lane.
→
SOURCE: Aterra Signal Library
2,166 signals from our daily CPG agent — trade coverage including Food Dive, FoodNavigator, NutraIngredients, BevNET and SupplySide, launch and funding announcements, podcasts and LinkedIn.
→
SOURCE: Aterra Signal Library
13,088 research entries and an 88-study research library.
→
SOURCE: Aterra Signal Library
821 products across four LA grocery banners spanning innovation, natural, premium and mass tiers, and a 246-product supplement e-tailer scan, both Sep 29, 2026.
→
SOURCE: Aterra Signal Library
Amazon-led window: complaints, search and claims are Amazon data (listings, review summaries, modelled search and sales estimates); formats are single-retailer shelf scans. Each finding was then checked against social listening, trade press, research and four LA grocery shelves — each item's 'Beyond Amazon' line says what those sources show, including where they are silent. Labels: direct corroboration means an outside source shows the same thing; adjacent evidence is related but not the same measure; no outside corroboration means the other sources are silent. Discussion is attention, search is intent, buying is choice, shelf organisation is structure — none stands in for another.
→
Two Amazon shelf scans, Sep 7 and Sep 29, 2026, 60 products per category. Share is a claim's portion of its category's top five claims; mentions per product normalises for sample size but not for which products are in the frame. A move is reported only if it beats the 26-claim cohort. Consumer attention compares signals since Sep 7 with an unbounded prior baseline — comparative, not a trend.
→
Modelled estimates of 30-day Amazon search volume, one month of change. Title density counts titles containing the exact phrase, so every gap is also tested against its core phrase (filler, doses and word order removed); 85 candidate gaps failed that test this cycle and were dropped. One month cannot separate a trend from a season.
→
Amazon's own review summaries ('Customers say'), Sep 29, 2026. Percentages are a complaint's negative mentions as a share of all aspect mentions, positive and negative. Top sellers are the five highest by Amazon's stated purchase volume; a blind spot needs them to draw the complaint well above peers and fewer than 10% of scanned titles to answer it. Titles are not full labels.
→
Format rows are one retailer's shelf per category (16–20 products), never compared across retailers. Ladder stages are percentiles against the same category's other themes; buying uses modelled Amazon keyword-sales estimates and is directional.
→
"For the multi-ingredient gummy combinations surfacing in Adaptogens & Stress Support — magnesium, ashwagandha, and l-theanine together — what does the shelf look like at other retailers, and are any emerging brands already naming that combination in their titles?"
The Gravity Letter — biweekly CPG intelligence from 20,000+ classified signals. Free.
BUILT FROM THE ATERRA SIGNAL LIBRARY — 20,000+ CLASSIFIED SIGNALS, ~3,000 ADDED MONTHLY. EVERY ITEM ABOVE CARRIES ITS SOURCE.