QOVES — Organic Search Performance

Google Search Console · sc-domain:qoves.com · weekly since 2025-03-10

KPI window: 2026-06-16 → 2026-07-13 vs prev 28d

Snapshot 2026-07-13 · generated 2026-07-17

Organic clicks

28,780

▼ 2.7% vs prev 28d

Impressions

479,884

▲ 32.8% vs prev 28d

Organic CTR

6.00%

▼ 26.7% vs prev 28d

Avg position

4.7

• 0.0% lower is better

Non-brand clicks

767

▼ 24.9% named queries

Brand share

96.8%

▲ 0.9% of named clicks

Blog clicks

2,260

▲ 99.8% /insights/

Non-blog clicks

27,221

▼ 6.0% rest of site

Brand vs Non-brand clicks

Weekly organic clicks. Brand = query matching the qoves family (qoves & ~90 spelling variants). · audit: issue #20

Brand vs Non-brand impressions

Weekly impressions split by brand / non-brand.

Top-3 keyword rankings + non-brand clicks

Count of queries ranking in positions 1–3 each week (impression-weighted). Clicks on right axis.

Blog content performance (/insights/)

Blog vs non-blog organic clicks. Blog on right axis (smaller scale).

Timeline

Markers 1–9 on the charts · ● Google update · ● site change

1

Mar 2025 Core Update

13 Mar 2025
2

Jun 2025 Core Update

30 Jun 2025
3

Aug 2025 Spam Update

26 Aug 2025
4

Dec 2025 Core Update

11 Dec 2025
5

Feb 2026 Discover Update

5 Feb 2026
6

Migration to /insights/ (new site)

23 Mar 2026
7

Mar 2026 Core Update

27 Mar 2026
8

May 2026 Core Update

21 May 2026
9

Piotr starts (SEO engagement)

12 Jul 2026

Top organic queries

Query	Clicks	Impr.	CTR	Pos.
qovesBRAND	13,287	50,285	26.4%	1.1
quovesBRAND	1,747	8,949	19.5%	1.2
qoves facial analysisBRAND	1,114	2,539	43.9%	1.8
qoves studioBRAND	805	2,201	36.6%	1.7
qoves reviewBRAND	441	7,774	5.7%	3.4
qoves priceBRAND	344	2,678	12.8%	1.5
qoves costBRAND	250	2,077	12.0%	1.7
qoves face analysisBRAND	184	459	40.1%	2.9
qoves.conBRAND	182	264	68.9%	1.0
quoveBRAND	179	914	19.6%	1.3
qoves.comBRAND	158	499	31.7%	1.0
qoves face analyzerBRAND	156	236	66.1%	1.0
qouvesBRAND	155	532	29.1%	1.0
qoveBRAND	146	1,018	14.3%	1.3
qoves analysisBRAND	138	570	24.2%	3.9

Top landing pages (non-blog)

Page	Clicks	Impr.	CTR	Pos.
/	25,329	195,180	13.0%	3.3
/for-doctors	803	73,024	1.1%	2.0
/careers	319	48,606	0.7%	1.2
/legal/terms-and-conditions	189	44,550	0.4%	2.5
/contact-us	146	50,708	0.3%	1.3
/insights	109	58,662	0.2%	1.5
/welcome	105	19,250	0.5%	6.4
/legal/privacy-policy	61	7,428	0.8%	2.5
/research	48	37,476	0.1%	1.5
https://status.qoves.com/	29	177	16.4%	3.9
/news	15	4,097	0.4%	3.3
/research/llm-facial-attractiveness	13	384	3.4%	6.4
/legal	12	1,823	0.7%	2.8
https://qoves.com/	10	445	2.2%	2.1
/announcements	8	1,126	0.7%	4.8

Content traffic by page type (/insights/)

Weekly organic clicks stacked by page type: article, measurement, concern, procedure. Excludes anchor-link rows and the /insights index.

Pages by type

Published (in sitemap) vs live (had impressions in last 28d), and how many earned nothing. · ISR/sitemap errors: issue #16

Type	Pub	Live	0-imp	Clicks	CTR
article	8	9	0	1,503	0.75%
measurement	65	28	37	526	1.05%
concern	17	17	0	214	0.26%
procedure	13	13	0	14	0.71%
comparison	8	0	8	0	0.00%

Pub = URLs in sitemap.xml · Live/Clicks/CTR = last 28 days · 0-imp = published but no impressions.

Published pages with no search impressions (last 28 days)

45 published pages had zero search impressions in the last 28 days — indexed/eligible but earning nothing (thin, new, or not ranking). Prime candidates for internal linking, consolidation, or content work.

measurement (37)

/insights/measurements/brow-position
/insights/measurements/chin-height
/insights/measurements/chin-width
/insights/measurements/columella-show
/insights/measurements/dorsal-axis-angle
/insights/measurements/double-break-angle
/insights/measurements/earlobe-size
/insights/measurements/eye-aperture-angle
/insights/measurements/eye-spacing-ratio
/insights/measurements/facial-depth
/insights/measurements/facial-fifths
/insights/measurements/forehead-width
/insights/measurements/intercanthal-width
/insights/measurements/interpupillary-distance
/insights/measurements/jaw-width
/insights/measurements/labiomental-angle
/insights/measurements/lower-anterior-face-height
/insights/measurements/lower-face-height
/insights/measurements/lower-gonial-angle
/insights/measurements/mandible-body-length
/insights/measurements/mandibular-plane-angle
/insights/measurements/mouth-width
/insights/measurements/nasal-aspect-ratio
/insights/measurements/nasal-bridge-width
/insights/measurements/nasal-height
/insights/measurements/nasal-length
/insights/measurements/nasal-root-width
/insights/measurements/nasal-tip-projection
/insights/measurements/naso-oral-ratio
/insights/measurements/nasofacial-angle
/insights/measurements/nasomaxillary-angle
/insights/measurements/neck-length
/insights/measurements/nostril-vertical-span
/insights/measurements/oral-commissure-tilt
/insights/measurements/smile-width
/insights/measurements/submental-angle
/insights/measurements/upper-gonial-angle

comparison (8)

/insights/qoves-vs-amazon-rekognition
/insights/qoves-vs-azure-face-api
/insights/qoves-vs-face-plus-plus
/insights/qoves-vs-haut-ai
/insights/qoves-vs-looksmax-ai
/insights/qoves-vs-perfect-corp
/insights/qoves-vs-umax
/insights/qoves-vs-youcam

Page	Clicks	Impr.	CTR	Pos.
/insights/facial-proportionality	505	59,735	0.8%	5.5
/insights/facial-dimorphism	221	23,910	0.9%	5.9
/insights/facial-adiposity	199	40,903	0.5%	5.9
/insights/facial-symmetry	193	24,722	0.8%	6.8
/insights/facial-neoteny	168	26,524	0.6%	5.4
/insights/facial-averageness	126	15,665	0.8%	5.5
/insights/facial-harmony	91	9,937	0.9%	5.9
/insights/how-a-glow-up-actually-happens	0	10	0.0%	7.4
/insights/attraction-science	0	66	0.0%	6.8

Core Web Vitals (PSI mobile)

Core Web Vitals not configured — no PageSpeed Insights API key. Set PSI_API_KEY and add a PSI pull to qoves_data.py to populate this card.

QOVES — SEO Audit (Keyword & Topic Audit, Sprint 1)

Google Search Console · sc-domain:qoves.com · issue #20

Snapshot 2026-07-13 · generated 2026-07-17

Non-brand: share of impressions vs. share of clicks

The headline finding in one chart. Non-brand visibility (impression share) is growing; non-brand clicks aren't following at the same rate — the gap between the two lines is what AI Overviews are absorbing before the click reaches QOVES.

Brand vs Non-brand by country (last 28d)

Share of clicks and impressions that are branded, per market (named queries only, from the query x country pull).

Clicks

Impressions

Questions from GSC (non-brand, last 28d)

Question-word match per country (local language + English word lists).

Query	Clicks	Impr.	CTR	Pos.
what are sanpaku eyes	1	436	0.2%	6.6
what is sanpaku eyes	2	259	0.8%	7.6
what is dimorphism in face	4	242	1.7%	2.1
how to measure facial thirds	3	151	2.0%	6.2
what is symmetry face	0	119	0.0%	1.1
what is sanpaku	0	94	0.0%	9.9
what is high dimorphism	0	83	0.0%	4.6
what causes sanpaku eyes	0	74	0.0%	7.3
what is a sanpaku	0	62	0.0%	8.3
what does sanpaku eyes mean	0	53	0.0%	8.4
what is facial dimorphism	0	51	0.0%	1.9
what is facial symmetry	0	48	0.0%	9.5
what are facial thirds	2	47	4.3%	5.3
how to know if you have sanpaku eyes	0	45	0.0%	6.4
sanpaku eyes vs normal eyes	0	40	0.0%	8.8

586 question queries · 25 clicks · 3,681 impressions (28d)

1,151 non-branded question queries in 28d (8,347 impressions, 79 clicks, CTR 0.9% vs 0.7% non-brand average). Mostly definitional (what is X) shapes that AI Overviews answer inline — the win here is being the source the AI answer cites, not the blue link.

Long-tail queries (10+ words, non-brand, last 28d)

Regex ^(?:\S+\s+){9,}\S+$ — conversational / AI-shaped demand.

Query	Impr.	CTR	Pos.
canthal tilt definition outer corner lower than inner corner source ophthalmology	74	0.0%	6.0
canthal tilt definition outer corner lower than inner corner ophthalmology	40	0.0%	7.0
eyes where you can see the white at the bottom	33	0.0%	7.4
canthal tilt definition outer corner lower than inner corner source	31	0.0%	8.3
what's it called when you can see the white under your eye	29	0.0%	9.0
negative canthal tilt definition outer corner lower than inner corner source ophthalmology	21	0.0%	6.0
what is it called when you can see the white under your eyes	20	0.0%	6.5
facial beauty analysis metrics symmetry proportions golden ratio attractiveness scores	19	0.0%	9.1
what does it mean when you can see the white under your eyes	19	0.0%	6.3
at what body fat percentage do you get hollow cheeks	18	0.0%	8.4
what is it called when the whites of your eyes show	17	0.0%	5.1
at what body fat percentage does your face look best	17	0.0%	6.9
what is it called when you can see white under your eyes	15	0.0%	7.9
what is it called when you have white under your eye	14	0.0%	6.6
what is it called when you can see the bottom white of your eye	12	0.0%	8.6

89 non-branded queries of 10+ words in 28d (625 impressions, 1 clicks, CTR 0.2%). Several read as verbatim AI-agent retrieval prompts (e.g. “canthal tilt definition outer corner lower than inner corner source ophthalmology”) — query fan-out arriving in GSC directly. 53 of them are also questions.

Keyword audit snapshot (non-brand fan-out clusters, last 28d) · issue #20

Impressions by topic cluster with CTR — tool-intent queries convert at ~25x the informational clusters sitting under AI Overviews.

Branded click share (28d)

96.8%

Non-brand share of impressions

45.9%

was 24% in the prior 28d

AI Overview absorption

~3x

Ahrefs-modeled non-brand traffic vs GSC actual (2026-07-16 check)

Best non-brand CTR

32%

"glow up without surgery" (90d, 2026-07-16 check)

What the audit found

QOVES's click profile is overwhelmingly branded (96.8% of named clicks in the last 28 days, ~98.5% all-time), but the impression footprint tells the opposite story: non-brand impressions roughly doubled period-over-period (24% → ~46% of the footprint) while non-brand clicks fell. Visibility is growing fast on informational SERPs and the clicks are not following. The reason is structural, not positional: QOVES holds position 1 for terms like sanpaku eyes meaning (8,400 searches/mo) and still earns almost nothing, because nearly every one of those SERPs carries an AI Overview that answers the question inline. Ahrefs's position-based traffic model expects ~10% of QOVES traffic to be non-branded; GSC reality is ~3%. That ~3x shortfall is the AI answer layer's absorption, measured.

How to read the clusters

The chart is the strategy in one picture. The big informational clusters (Sanpaku/eyes at ~51K impressions, 0.2% CTR; proportions; dimorphism) no longer pay in clicks — their job is fan-out coverage: being retrievable and cited when AI systems decompose a question into sub-queries (the long-tail table above shows those sub-queries arriving in GSC verbatim). Click capture happens on Analyzer/tool intent (5.3% CTR, ~25x the informational rate) — query shapes an AI answer cannot satisfy because the user wants their own face measured. The two proven wedges: tool-intent surfaces ("face analysis app", "facial assessment tool"), and the analysis-first non-surgical angle — "glow up without surgery" converts at 32% CTR, QOVES's best non-brand query anywhere. The older-audience cluster (look younger at 40, facial rejuvenation, face exercises for jowls) is a true content gap: zero current footprint, keyword difficulty mostly 0–8. Full working in issue #20 (2026-07-16 refresh; prose is point-in-time, hand-maintained).

Non-brand CTR & position by country (last 28d)

Is the gap a ranking problem or an AI Overview absorption problem? A market with a decent average position but a bad CTR isn't explained by rank — something is eating the click before it reaches QOVES. Countries in red rank at-or-better than the median position here but still lag the median CTR — computed from this chart's own 28d data, not assumed.

Non-brand CTR

Non-brand avg. position (lower is better)

Open question, not yet scoped

A separate, wider 90d GSC pull (2026-07-17, see issue #20, a different window than the 28d chart above) found India and the Philippines with a decent average position (8.1, 7.5 — not an AI-Overview-suppressed shape) but the second- and third-worst non-brand CTR of any market (1.77%, 0.44%). That's not explained by rank and not explained by language (both are large English-competent markets). Most likely a purchase-intent or pricing mismatch rather than an SEO problem — flagged on #20 as an open question, deliberately not scoped as a task yet.

QOVES — Checkout Performance Audit

GA4 Funnel Exploration · all traffic, global · membership checkout (/welcome/checkout)

Funnel window: 3–5 Jul 2026

Live re-check window: 19 Jun – 16 Jul 2026 vs prev 28d

Reach checkout

31.2%

of all sessions

Checkout → purchase

2.6%

of checkout visitors

Cart abandonment

97.4%

vs 70% industry avg

Desktop conversion

5.4%

▲vs mobile

Mobile conversion

1.4%

most traffic is mobile

India checkout conv.

0.09%

22.6% of sessions

US checkout conv.

7.5%

best-performing market

Projected upside

+$600K

/mo at 10% target conv.

The funnel

Session start → page view → checkout page → purchase. Direct labels show % of Step 1 and the drop from the prior step.

Checkout → purchase by country

India drives 22.6% of sessions but converts near zero, worth confirming traffic quality before investing further.

Cart abandonment vs. industry benchmarks

Bands show typical ranges; QOVES bars are single-point, both sit well above every comparable category.

Revenue scenarios (English-speaking markets)

Monthly revenue at $220 AOV if checkout→purchase moves from today's 6.3% to a realistic (3mo) or stretch (6mo) target.

Desktop vs. mobile conversion

Most checkout traffic is mobile, so this gap is the single highest-leverage fix available.

Questions the team asked

"Is there a big drop in organic performance this month? Is GA4 showing the same?"

Re-verified live 2026-07-16 with a fresh GA4 pull, last 28 days (19 Jun–16 Jul) vs. the prior 28 days. Organic Search: sessions −14.9% (35,154 vs 41,317), transactions −1.4% (748 vs 759), revenue −4.1% ($151,950 vs $158,375). Verdict: yes, real, and it's grown since the last check on this page five days earlier (which had sessions at −7.4%). Traffic is declining faster than transactions or revenue, so it hasn't hit the business yet, but the trend is worsening, not stabilizing.

"Checkout abandonment, is that correct?"

Recomputed row-by-row against the same live 28-day window (not just the original 3-day Notion snapshot): US abandonment is now 94.6% (was 92.5%), UK 97.3% (was 96.3%), India still effectively total (>99.9% of 87K checkout sessions converting to under 93 purchases). Verdict: still directionally correct, and slightly worse than the audit reported. Same caveat stands: /welcome/checkout still conflates the email-signup step with the payment step, so this number isn't yet diagnostic of where in that page people actually leave.

Still open, per the audit

Is the Indian traffic organic or paid? At 0.09% checkout conversion it isn't recoverable with optimisation alone.
Why does the UK convert at half the US rate (3.7% vs 7.5%), pricing, payment methods, or something else?
What does the mobile checkout experience actually look like? Most traffic is mobile and it converts at a quarter of desktop's rate.
Has a one-time payment option been tested against the annual membership framing?

Recommendations

Immediate (this week)

Replace the money-back-guarantee claim near the Join Now button with a neutral link to the actual refund policy (no guarantee claim; keep the no-refund-after-photo-submission detail on the dedicated policy page, not as visible copy beside the CTA)
Expand the membership value breakdown on the checkout page
Separate the email-capture URL, or fire an "email_submitted" event (issue #17)

Short-term (next 2 weeks)

Test reframing checkout copy away from "subscription" language
Add PayPal as a payment option

Medium-term (next month)

Investigate the mobile checkout experience specifically
Test removing the separate email step entirely
Build an abandoned-checkout email sequence

Tracking improvements needed tracked in issue #17

Task	Owner	Priority
Implement begin_checkout event (fires on payment page load)	Dev	HIGH
Implement add_to_cart event (fires when plan is selected)	Dev	HIGH
Separate URL for email capture, or fire "email_submitted" event	Dev	HIGH
Both events need value, currency, items parameters	Dev	HIGH
Enable PayPal in Stripe checkout	Josep	MEDIUM

QOVES — AI Assistant Referral Traffic

GA4 sessionSource, regex-matched against known assistant domains · all traffic, global

Window: 17 Apr – 16 Jul 2026 (90d)

Live pull 2026-07-16 · property 533102069

Part 1 — clickthrough traffic. Tracks visitors arriving via a link clicked inside an AI assistant (ChatGPT, Perplexity, Claude, Gemini, Copilot, You.com, Phind, HuggingChat, …). This is GA4-native, live today. Part 2 below (LLM brand visibility) measures a different thing — whether those assistants mention QOVES when prompted, independent of clickthroughs. Full spec in issue #28.

Regex-matched sessions

~10,500

last 90 days

GA4's "AI Assistant" bucket

4,084

sessionDefaultChannelGroup, same window

Missed by GA4's bucket

~6,400

61%of real AI-referral traffic uncounted

Share of all sessions

0.26%

vs 0.10% trusting GA4's bucket alone

Daily sessions from AI assistant referrers

Near-zero through April, a step-change starting 7 May (2–18/day → 84/day overnight), peaking at 266/day on 27 May, then a slow decline to ~50–70/day by mid-July — the shape of a single event fading, not steady organic growth.

Where each source's sessions actually land in GA4's channel grouping

Indigo is the slice GA4 correctly labels "AI Assistant"; rose is real assistant traffic hiding under Referral, Unassigned, or Organic Search. chatgpt.com sessions with a bare/ambiguous referrer (54% of its traffic) fall to Unassigned; a bare token like perplexity (no TLD, likely an in-app browser stripping the referrer) is 100% missed.

Source breakdown, 90 days

sessionSource	Sessions	GA4 "AI Assistant"	Missed	Worst bucket
chatgpt.com	10,338	4,009	6,329	Unassigned (5,578 · 54%)
gemini.google.com	91	51	40	Referral (40)
perplexity (bare, no TLD)	33	0	33	Unassigned (33 · 100%)
claude.ai	30	12	18	Referral (18)
perplexity.ai	17	6	11	Referral (11)
openai	1	0	1	Organic Search (1)

Part 2 — LLM brand visibility (LLMTracker)

Does ChatGPT/Claude/Gemini mention or recommend QOVES when prompted — independent of Part 1's clickthrough traffic.

Brand visibility by model (1 weekly cycle, $0.74 spent)

Model	Appearances	Avg rank	Avg SoV	Cost
Claude Opus 4.7	2/2	1.0	96.9%	$0.33
GPT 5.5	2/2	1.0	95.7%	$0.28
Gemini 3.1 Pro	2/2	1.0	100.0%	$0.13

ChatGPT query probe (2 capture cycles committed (latest: 2026-06-20))

Real chatgpt.com sessions captured via the browser extension (real web-search + memory behavior, not the raw API) — see LLMTracker/reports/geo/.

Presence rate

8/15

53.3% of tracked prompts

Primary-cited

qoves.com as the lead source

Reputation flags

negative-tier prompts this cycle

Prompt	QOVES	Mode
is qoves facial analysis worth it	PRIMARY	searched
qoves studio review	PRIMARY	searched
qoves vs umax vs looksmax ai	PRIMARY	searched
alternatives to qoves	PRIMARY	searched
what is the best facial analysis service	ABSENT	searched
most accurate facial analysis tool	ABSENT	searched
best AI facial analysis tool	ABSENT	searched
where can I get a professional facial analysis done	ABSENT	searched
who does the most in-depth professional facial analysis	PRIMARY	searched
is there a facial analysis service run by experts and backed by research	PRIMARY	searched
are face rating apps like umax accurate	ABSENT	searched
what is the best facial analysis service	ABSENT	searched
best AI facial analysis tool	ABSENT	searched
is qoves a scam	PRIMARY	search-req'd->memory
qoves complaints and refunds	PRIMARY	searched

Competitors recommended: Amazon Rekognition (2), Microsoft Azure Face API (2), iProov (2), InsightFace (2), Face++ (2), LooksMax AI (1), Face Balance Score : FaceScore - App Store - Apple (1), Beauty Scanner - Face Score AI - Apps on Google Play (1)

Cited non-QOVES domains: aws.amazon.com (4), play.google.com (3), Google Play (3), en.wikipedia.org (3), arXiv (3), apps.apple.com (2), overchat.ai (2), reddit.com (2)

No negative-tier prompts flagged in the latest cycle.