Daily Insights

Ulta vs. Sephora: All Search, All Shades

By: Ried Niziak | Sep 07, 2017

Online wholesale partners represent a significant portion of e-commerce revenue for many brands. As a result, shifts in a retailer’s strategy can significantly impact brands. Google drives approximately 40% of top retailers’ online traffic, so understanding how retailers invest and perform on the platform can provide valuable insights into their digital strategies.

Our goal was to develop an analysis and corresponding visuals to depict differences in search strategies for competing retailers. This led us to calculate and compare the amount of traffic retailers receive from specific search keywords. Our first test case analyzed beauty retailers Sephora and Ulta.

Methodology: L2’s data partner SEMRush estimates the amount of organic and paid traffic that specific Google search queries drive to domains. Working with this data, we compiled a list of the top search queries driving traffic to each retailer, which allowed us to estimate the amount of traffic driven on a query-level basis. For example, we wanted to be able to attribute traffic driven by “mascara,” no matter whether the query was “lengthening mascara” or “waterproof mascara.” Our list of top keywords included any query that was responsible for at least 0.01% of a retailer’s organic or paid traffic, resulting in 1,931 keywords for Sephora and 1,947 for Ulta.

To efficiently aggregate them, we cleaned the text of each query. We removed stop words (e.g. “and,” “of”) and stemmed each word, which is a technique to reduce words to their root form. This was important in order to treat plurals (“cream” and “creams”), tenses (“dye” and “dyed”), and instantiations (“contour” and “contouring”) as the same word. Next we identified the unique 1 and 2 n-grams in each query, which separated all one- and two-word phrases found in the search queries. For example, the search query “face cream” would be broken into three terms: “face,” “cream,” and “face cream.”

We could then sum the amount of traffic each retailer received from queries with particular words. In April 2017, for example, “lip gloss” search queries contributed to 13,000 organic search visits to Sephora and 8,000 organic search visits to Ulta. We calculated this traffic for all 5,500 unique n-grams identified. Finally, we removed any terms that contributed to less than 1,000 total visits and 100 paid visits, as well as terms for which a single retailer controlled more than 99% of organic or paid traffic. This resulted in a final list of 228 core competitive terms.

Results: We graphed the results by the share of organic and paid traffic each retailer controlled, which enabled us to extract insights as to the focus and strengths of each retailer.

Ulta vs. Sephora

Highlighting only hair care terms reveals that Ulta commands organic traffic in the category. Ulta also owns the largest share of paid traffic, but this difference is less extreme, indicating that Sephora has not completely forfeited the category.

Ulta vs. Sephora

The retailers are split between fragrance and nail terms. Sephora controls the largest share of organic and paid traffic related to fragrance terms, while Ulta controls the largest share for nail terms.

Ulta vs. Sephora

For skincare terms, Ulta maintains dominance in organic and paid traffic. However, high volume terms like “cream” and “face” are much more competitive between the two retailers than long tail terms (e.g. “sunscreen,” “acne”). In addition, Sephora makes specific investments in problem and solution terms, like “acne,” “pore,” and “treatment.”

Ulta vs. Sephora

For branded terms that drive traffic to both retailer sites, Sephora is the organic and paid traffic leader. In fact, Sephora has the highest share of organic traffic on 66% of branded terms and of paid traffic on 61% of branded terms.

Trendy brands bareMinerals, Urban Decay, and Anastasia Beverly Hills are the most competitive terms. Interestingly, while Sephora owns almost 90% of the organic traffic associated with Anastasia Beverly Hills terms, Ulta makes signifigant paid investments on the same keyword set. As a result, Ulta captures 70% of the paid traffic driven by “Anastasia Dipbrow” terms. This uncovers a potential strategy to draw loyal Anastasia customers from Sephora via one of the brand’s most popular products.

Ulta vs. Sephora

Conclusion: At L2 we pride ourselves on the ability to not only identify trends, but to quantify them. This analysis is a useful tool for highlighting the changes in the competitive landscape. We can identify which product categories (e.g. “lip gloss”) or brand terms (e.g. “Clinique”) currently command the highest investment from retailers and can track how traffic drivers have changed over time.

This approach can help brands identify potential retail partnerships. Understanding a retailer’s strengths and investment priorities on search provides insights into the type of consumer the retailer is targeting and subsequently driving to its site, and whether those consumers align with a brand’s product assortment and price point.

The analysis is inherently limited because the research is based on mathematical projections as opposed to primary source data. Historically this has resulted in uncertainty regarding absolute traffic numbers and accuracy of the model’s results for much smaller retailers. This analysis is best used to measure relative differences between domains and to track changes over time.