Why SEO Is Not Enough A Benchmark Study
A Comparative Analysis of Brand Authority, Citation Depth, and Generative Search Presence
Focus: D2C Beauty & Skincare Niche | Study Date: April 2026 | Brands: Mamaearth · Purplle · Dot & Key · Plum Goodness · Tira Beauty
1. The Argument
There is a widespread assumption in digital marketing that ranking well on Google will get your brand into AI-generated answers. This study, using Semrush data from five competing Indian D2C beauty brands, demonstrates that this assumption is wrong — and the data makes it impossible to ignore.
Purplle has 108,700 organic keywords. Mamaearth has 71,200. Purplle has 30,000 indexed pages. Mamaearth has 2,732. Purplle has more backlinks, more content, more organic keywords — and yet its AI visibility score is 52, versus Mamaearth’s 77. Purplle is winning SEO by almost every conventional metric, and losing AI visibility by a wide margin.
Tira Beauty has 15,418 pages. Its AI visibility is 58. Plum Goodness has 986 pages. Its AI visibility is 72. More pages is not a signal AI systems respect. Recommendation trust is.
The core thesis
SEO gets you indexed. Content gets you discovered. Brand + PR + Product gets you recommended. These are three different outcomes — and only one of them determines AI visibility. Mamaearth has all three working together. That is the gap.
2. The Data at a Glance — Mamaearth as the Benchmark
All comparisons in this study use Mamaearth as the benchmark, as it holds the highest AI visibility score (77) among the five brands analysed.
| Brand | AI Visibility | Auth Score | Org Traffic (IN) | Org Keywords | Total Pages | Traffic / Page |
| Mamaearth | 77 | 57 | 1.3M | 71,200 | 2,732 | 471 |
| Purplle | 52 | 57 | 1.2M | 108,700 | 30,000 | 38 |
| Dot & Key | 68 | 51 | 964K | 21,800 | 633 | 1,523 |
| Plum Goodness | 72 | 54 | 721K | 27,100 | 986 | 732 |
| Tira Beauty | 58 | 50 | 645K | 62,900 | 15,418 | 42 |
Two numbers demand immediate attention.
First: Purplle and Tira have page counts of 30,000 and 15,418 respectively — yet their traffic-per-page ratios are 38 and 42. That means 97%+ of their pages generate negligible traffic individually. Mamaearth runs 2,732 pages at 471 traffic per page. Dot & Key with 633 pages averaging 1,523 traffic per page and Plum runs 986 pages at 732 traffic per page. The brands with the highest AI visibility are those where every page earns its place — not those that chase scale.
Second: authority score does not explain AI visibility. Mamaearth and Purplle both score 57 on authority. Their AI visibility scores are 77 and 52 respectively. Domain authority is a necessary but insufficient condition for AI recognition.
3. Branded Traffic — Why Percentage Is the Wrong Metric
One of the most common misreadings of this data is to look at branded traffic percentages and conclude that Mamaearth (59% branded) is weaker on brand than Dot & Key (83%) or Plum (66%). This is misleading. It ignores the absolute volume of branded demand — which is the signal AI systems actually respond to.
| Brand | Org Traffic | Branded % | Branded Traffic (abs) | Non-Branded (abs) | vs Mamaearth | AI Visibility |
| Mamaearth | 1,300,000 | 59% | 767,000 | 533,000 | BENCHMARK | 77 |
| Dot & Key | 964,000 | 83% | 800,120 | 163,880 | +33,120 | 68 |
| Plum Goodness | 721,800 | 66% | 476,388 | 245,412 | -290,612 | 72 |
| Purplle | 1,200,000 | 28% | 336,000 | 864,000 | -431,000 | 52 |
| Tira Beauty | 645,400 | 36% | 232,344 | 413,056 | -534,656 | 58 |
Reading this table correctly: Mamaearth generates 767,000 branded traffic visits per month in absolute terms. Plum Goodness, despite a higher branded percentage (66%), generates only 476,388 branded visits — 290,612 fewer than Mamaearth. Dot & Key, at 83% branded, generates 800,120 branded visits — slightly above Mamaearth, but with only 163,880 non-branded visits to show for its entire content operation.
Purplle generates only 336,000 branded visits — less than half of Mamaearth — despite comparable total organic traffic. This is the most telling number in the entire dataset. Purplle has traded brand depth for keyword breadth. Its enormous non-branded keyword base (864,000 non-branded visits) tells us it can be found for product and category searches — but the brand itself is not generating the kind of direct, intent-driven demand that AI systems recognise as a signal of brand authority.
Why absolute branded traffic matters to AI
When millions of people search for a brand by name, that is a real-world signal of brand trust and recall. AI models — trained on internet-scale data — absorb this signal indirectly: through the volume and sentiment of brand mentions, reviews, social discussions, and editorial references. Percentage splits are a relative measure. What builds AI recognition is the absolute weight of brand evidence in the world.
The Dot & Key nuance is important here. Its 83% branded share and 800,120 absolute branded visits indicate extraordinary brand loyalty — people love the brand and go directly to it. But its 163,880 non-branded visits mean the brand has almost no presence in discovery contexts: skincare routines, ingredient queries, category comparisons. For an AI model responding to “best vitamin C serum India”, Dot & Key may not even appear — not because the product is not excellent, but because the web has too little non-branded evidence to surface it in that context.
4. More Pages ≠ More AI Authority
The clearest structural proof of the SEO-vs-AI gap in this dataset is the relationship between page count and AI visibility. If SEO scale drove AI recommendations, Purplle and Tira would be leading. They are not.
| Brand | Total Pages | Blog Pages | Product Pages | Traffic / Page | Backlinks | AI Visibility |
| Mamaearth | 2,732 | 1,030 | 1,359 | 471 | 20,300 | 77 |
| Purplle | 30,000 | 1,155 | 20,924 | 38 | 54,200 | 52 |
| Tira Beauty | 15,418 | 533 | 11,485 | 42 | 51,400 | 58 |
| Plum Goodness | 986 | 301 | — | 732 | 101,100 | 72 |
| Dot & Key | 633 | 307 | — | 1,523 | 11,300 | 68 |
The Purplle Case: 30,000 Pages, AI Visibility of 52
Purplle’s 30,000 pages include 20,924 product listings and 4,783 category pages — a classic multi-brand retailer index. At 38 traffic per page, these pages function as SEO infrastructure: discoverable, rankable, but thin. An AI model asked to recommend a beauty brand does not consult a product listing page. It draws on editorial content, reviews, brand mentions, and expert citations — precisely what Purplle’s scale-led strategy has underinvested in relative to its page count.
Purplle’s 1,155 blog pages contribute 10.3% of total traffic — a lower share than Mamaearth’s 1,030 blog pages contributing 17.6%. The issue is not absence of content — it is absence of depth and topical authority. Volume without expertise does not build the recommendation trust that AI systems are calibrated to recognise.
The Tira Case: 15,418 Pages, AI Visibility of 58
Tira faces a structural challenge that page counts cannot solve: it is a multi-brand retailer. When an AI model answers “which serum is best for oily skin?”, it will recommend the brand — Plum, Mamaearth, Minimalist — not the platform that lists them. Tira’s 11,485 product pages are effectively marketing another brand’s story. Its 533 blog pages generating just 3% of traffic confirm that editorial authority — the one channel that could differentiate Tira’s own voice — has not been built at scale.
What Mamaearth Does Differently
Mamaearth’s 2,732 pages at 471 traffic per page means nearly every page it has published earns consistent traffic. Its 1,030 blog pages — covering ingredient science, skin concerns, and beauty routines — attract 17.6% of total organic traffic. These are not thin SEO pages. They are the type of in-depth, expertise-signalling content that editorial sources, dermatologists, and review platforms cite — and that therefore makes its way into the training data and citation patterns of AI models.
5. What Actually Builds AI Visibility — The Three-Layer Model
The data points to a consistent pattern across all five brands. AI visibility is not determined by any single metric — it is the product of three distinct layers operating in concert, each owned by different teams within the organisation.
| Layer | What It Does | Who Drives It | Mamaearth Evidence |
| SEO | Gets you indexed & ranked | SEO & Tech teams | 71,200 keywords · 2,732 pages · DA 57 |
| Content | Gets you discovered in informational queries | Content team | 1,030 blog pages · 18% of total traffic |
| Brand + PR + Product | Gets you recommended by AI | Product · PR · CX · Performance | 767K absolute branded traffic · 20.3K backlinks |
Layer 1: SEO — Gets You Indexed
SEO is the foundation. Without it, no organic discoverability. Mamaearth scores 57 on domain authority, holds 71,200 organic keywords, and has a technically clean site. But Purplle also scores 57 on authority and holds 108,700 keywords. Both have strong SEO foundations — yet one is recommended by AI and the other is not. SEO alone does not predict AI visibility.
Layer 2: Content — Gets You Discovered
Content is what gets a brand into informational search results and — more importantly — into the body of knowledge that AI models train on. Blog depth matters more than blog volume. Mamaearth’s blog pages average significantly higher traffic per page than Purplle’s, suggesting genuine topical authority rather than keyword stuffing. Plum’s 301 blog pages achieving 12.1% of total traffic is the highest efficiency ratio in the group — a result of focused, expert-led content that earns both links and citations.
Layer 3: Brand + PR + Product — Gets You Recommended
This is the layer SEO cannot buy. It is built by:
- Product teams that create quality products generating authentic reviews on Nykaa, Amazon, and Google
- PR teams that earn mentions in Vogue India, Cosmopolitan, The Ken, and health/wellness media — high-authority sources that appear in AI training data
- Customer experience teams that drive satisfaction, reducing negative sentiment that AI models are trained to avoid recommending
- Performance marketing teams that build brand awareness at scale, increasing branded search volume and the density of brand mentions across the web
Mamaearth’s 767,000 absolute branded visits per month, 20,300 backlinks, and 8,500 AI mentions are not the output of any single team. They are the cumulative result of every function in the company doing its job well enough that the internet — collectively — has formed a strong, positive, repeated association with the brand. That is what AI models learn from.
6. Individual Brand Profiles vs. Mamaearth
6.1 Mamaearth — AI Visibility: 77
The benchmark
Mamaearth is the only brand in this set that has all three layers working at scale simultaneously — SEO authority, deep content, and absolute branded demand. That combination is what a 77 AI visibility score looks like.
What Mamaearth does that others do not: It distributes traffic across all content types — Homepage (30%), Blog (18%), Category (27%), Product (25%). No single page type dominates. For an AI model, this means Mamaearth appears in a wide variety of query contexts: brand searches, ingredient questions, product category comparisons, and specific product needs. Each query context is a separate opportunity to be cited.
Its 8,500 AI mentions across ChatGPT (2,400), AI Overviews (1,800), AI Mode (1,200) and Gemini (3,100) confirm consistent cross-platform presence. No competitor has this breadth of AI platform coverage simultaneously.
6.2 Plum Goodness — AI Visibility: 72
The most efficient AI performer in the set. With 3.5x fewer pages than Mamaearth and 44% less organic traffic, Plum achieves 72 AI visibility. The driver is not scale — it is quality signal density. Its 101,100 backlinks — by far the highest in the set — reflect years of earned media, ingredient credibility (clean beauty positioning), and third-party citations that LLMs weight heavily.
Plum also demonstrates the correct reading of branded traffic. Its 66% branded share looks strong, but the absolute figure of 476,388 branded visits is 290,000 less than Mamaearth. The gap to close is not percentage — it is absolute brand demand. If Plum can grow total traffic while maintaining its content quality and backlink velocity, it is structurally positioned to challenge Mamaearth on AI visibility.
6.3 Dot & Key — AI Visibility: 68
The loyalty paradox. Dot & Key has built extraordinary brand loyalty — 800,120 absolute branded visits, the only brand to exceed Mamaearth in this metric. Consumers who know Dot & Key go directly to it. But its 163,880 non-branded visits mean the brand has almost no footprint in discovery contexts. An AI model answering a category query has insufficient non-branded evidence to surface Dot & Key.
The implication for Dot & Key is strategic: it does not need to fix brand strength — it needs to build category authority. Deeper content around its hero ingredient categories (Vitamin C, SPF, Ceramides, Retinol) — expert guides, dermatologist endorsements, comparison content — would create the non-branded discovery surface that AI systems need to cite the brand in open-ended recommendation queries.
6.4 Tira Beauty — AI Visibility: 58
The retailer ceiling. Tira’s AI visibility challenge is not solvable by more pages or more SEO. It is structural: AI models recommend brands, not platforms. Tira’s path to higher AI visibility runs through editorial authority — original beauty guides, trend reports, expert advice, and a distinct editorial voice that makes Tira itself the recommendation, not just the destination for brands that are already recommended.
Its 9,100 AI mentions (the highest in this set) but only 58 AI visibility is the most instructive data point. Volume of mentions does not equal quality of recommendation positioning. Tira is mentioned frequently — likely in product listing contexts — but not recommended authoritatively.
6.5 Purplle — AI Visibility: 52 (the instructive failure)
The cautionary case. Purplle is the brand that proves the thesis most forcefully. It has more organic keywords than anyone else (108,700). It has a comparable authority score to Mamaearth (57). It has more backlinks than Mamaearth (54,200 vs 20,300). And yet it scores 52 on AI visibility — 25 points behind Mamaearth.
The reason becomes clear in the branded traffic data: Purplle generates only 336,000 absolute branded visits — less than half of Mamaearth’s 767,000. Despite identical organic traffic volumes (~1.2-1.3M), Purplle’s brand has not converted its SEO scale into brand conviction. Consumers find Purplle for product searches but are not choosing it as their go-to brand. AI models mirror this reality: they will surface Purplle pages for specific product queries, but they will recommend Mamaearth when asked who to trust.
7. A Framework for AI Visibility
SEO → gets you indexed | Content → gets you discovered | Brand + PR + Product → gets you recommended
These are not sequential steps. They are parallel investments. Brands that run all three simultaneously — as Mamaearth does — accumulate AI visibility. Brands that over-index on one layer while neglecting the others — like Purplle on SEO, or Dot & Key on brand — achieve partial visibility at best.
The practical diagnostic: If your branded absolute traffic is growing year-on-year, if your content is earning third-party citations, and if your product reviews are consistently positive — your AI visibility will follow. If any of these three is absent, no amount of SEO investment will close the gap.
9. Conclusion
The data from five Indian beauty brands — analysed through AI visibility scores, absolute branded traffic, page-level efficiency, content architecture, and backlink profiles — points to a single, clear conclusion: AI visibility is a brand outcome, not an SEO outcome.
Purplle proves you can win SEO and still lose AI visibility. With 108,700 organic keywords, 30,000 pages, and domain authority equal to Mamaearth — it scores 25 points lower on AI visibility. The gap is not technical. It is a brand gap: insufficient absolute branded demand, insufficient content depth, and an SEO strategy that has prioritised keyword breadth over recommendation trust.
Mamaearth proves that AI visibility is achieved only when SEO is supported by brand strength, content depth, and trust signals. Its 767,000 absolute branded visits, 471 traffic-per-page ratio, 8,500 AI mentions, and distributed content architecture are not the result of any single team’s work. They are the output of an organisation where product, content, PR, performance marketing, and customer experience all contribute to a brand the internet — and therefore AI — has learned to trust and recommend.
The question to ask your team
Stop asking ‘how do we rank for more keywords?’ Start asking: ‘are we building enough absolute brand demand? Is our content deep enough to be cited? Is our PR generating enough independent validation? Are our products generating enough positive sentiment?’ The answers to those questions are your AI visibility score.
Data: Semrush | All figures approximate
