Artificial Intelligence

Artificial Intelligence

Mar 11, 2024

Mar 11, 2024

Mar 11, 2024

How Pinterest’s programmatic SEO engine drives 300M+ visits per month

Pinterest has always been an anomaly; from its initial record-breaking growth to its consistently high traffic, the company has managed to defy expectations — and even categorization — at every turn.

The company leverages a product-led, programmatic SEO strategy that works at scale to drive over 1 billion monthly visits, about 35% of which come purely through organic search. 

Programmatic SEO, or pSEO, refers to converting structured data into content at scale. It typically involves creating keyword-targeted pages in an automated way to improve visibility on search engine results pages (SERPs). Pinterest’s ethos of user-centricity combined with its programmatic approach to SEO helps the platform achieve remarkable monthly traffic.

Dissecting the numbers behind Pinterest’s pSEO machine

Traffic funnel analytics show a double-corded strategy to drive traffic

According to data from SEMrush, Pinterest relies on a two-pronged approach to drive the vast majority of its traffic:

  1. People going directly to the website (~57% of traffic)

  2. People finding Pinterest through organic search (~35% of traffic)

Together these two traffic sources account for ~93% of Pinterest’s total web traffic which amounts to over 1 billion monthly visits. So, how does Pinterest maintain such a high traffic volume with little to paid search, email marketing, display advertising, or social promotions? A breakdown of Pinterest’s traffic sources reveals some key learnings. 

Organic traffic is driven by a pSEO strategy 

According to SEMrush, Pinterest’s pSEO strategy draws in roughly 35% of its visits each month amounting to around 360 million visits. Pinterest uses three different templated page types to draw in traffic:

  1. Pin pages (“pins”): Pinterest describes these pages as “visual bookmarks” that users create to save content they love. All pins follow the same template that’s been optimized for search engines. 

  2. Board pages (“boards”): Pinterest describes boards as spaces where users can “save, collect, and organize” pins. All board pages use the same template and URL structure:[user]/[collection], essentially creating a massive long tail. 

  3. User profiles: As the name implies, user profiles are spaces where individuals can save their boards and pins. 

According to SEMrush, pin pages pull in around 124.6 million organic visits per month, accounting for about 35% of Pinterest’s organic traffic.

Direct traffic is driven by a product-led strategy

Pinterest’s consistently high traffic is driven predominantly by direct visits which account for more than half of its monthly traffic. A high percentage of direct traffic is typically a clear indicator of product market fit (PMF) and a love for the product among consumers.

From the beginning, Pinterest has prioritized user experience (UX) to draw in a loyal following. The company has entire teams dedicated to UX research and design. and As a result, the company has defied typical social media formats to build something wholly new: a hybrid social media platform and search engine that reflects its users’ demands. Pinterest claims that it is not a social media site, but rather “a visual inspiration platform where people come to search, save, and shop.” 

Ultimately, Pinterest’s product-led approach amplifies its pSEO strategy and vice-versa. Search engines will always be changing to surface better search results that meet user intent. Pinterest’s focus on user experience reduces the likelihood that its pages will be deprioritized as these changes happen. 

Pinterest’s unique approach to internationalization

Pinterest caters to a diverse, global audience across hundreds of countries and territories. In fact, as of 2018, more than half of Pinterest’s monthly active users were outside of the United States. 

To meet the needs of a global audience, provide a localized experience, and improve SEO, Pinterest introduced country code top-level domains (ccTLDs) for certain geographies like Canada, Spain, Japan, and the U.K., and subdomains for other geographies like India, Brazil, Indonesia, and Vietnam.

This strategy improves SEO by ensuring users find relevant content in their language and, in the case of ccTLDs, the local domain format they recognize (e.g.  for those in Spain or  for those in the United Kingdom). 

Below are the traffic statistics across Pinterest’s top geographies with specific tags labeling subdomains vs. ccTLDs.

Breaking down Pinterest’s product and how it created a pSEO edge

As noted earlier, Pinterest’s pSEO strategy relies on the unique architecture of its three page types: pins, boards, and user profiles. Pins and boards are optimized for SEO and user experience while user profiles exist largely for user experience. Below is a breakdown of the structure of pins and boards — the pages that leverage Pinterest’s winning pSEO template. 

Examining the structure of Pinterest pin pages

By examining Pinterest’s pin pages we can see the Pinterest pSEO template at work. When a user creates a pin page, they are prompted to include the same elements (descriptions, titles, images) so that each pin page has an identical structure, regardless of the user who posted it. This structure includes: 

  1. Image: Pinterest touts itself as a “visual discovery” engine. To satisfy user intent for visual discovery, the image is the most prominent aspect of the pin.

  2. Link: The source link is included at the top of the pin page for users to easily learn more about the context of the image from the original source.

  3. Title: The title is defined by the user who creates the pin. It’s used as the H1 of the page and the meta title.

  4. Description: The description of the image is also defined by the user and is used as the meta description for the page.

  5. User profile: Users can click to view the profile of the user who posted the pin, enabling further discovery.

  6. Comments: Users can comment on pins and also “like” pins. This functionality is intended to increase engagement and dwell time on the page which is currently a factor in Google’s page ranking algorithm.

  7. Related content: This section offers related content to users so they can discover additional content. Again, this functionality is intended to keep users engaged and on the site longer. 

With this template, Pinterest can create thousands of keyword-targeted, SEO-optimized pages. While some pins end up pulling more weight than others from an organic traffic perspective, the creation of pages is near-automatic. This simplistic page template also optimizes user experience: All pins are curated in a similar format, allowing users to view a concise array of choices.

Examining the structure of Pinterest board pages

Pinterest boards are designed for users to share their personal tastes in a curated format. Each board holds a collection of pins that are all categorically related. Every board includes: 

  1. Board title: This is the page H1 and meta title, which is defined by the user.

  2. User: Each board is owned by a user who is linked at the top of the page.

  3. Share button: This feature enables users to share an entire board on other social channels. 

  4. Related boards: This feature encourages users to explore additional content and is intended to increase engagement and time on site (factors in Google’s page ranking algorithm).

  5. Pins: Each pin page is displayed as a card on the board. The card includes the pin image, title, and description. 

Again, the board format enables Pinterest to create thousands of keyword-targeted pages with one template while also catering to user desires — in this case, to curate and share their unique taste and style with friends. 

Pinterest’s philosophy on SEO experimentation

Pinterest runs a large number of measurable experiments to understand how different SEO strategies and hypotheses perform on Pinterest’s pages. Then, the team makes adjustments to quickly expand successful experiments and halt any failing tests as soon as possible.

In 2015, Pinterest documented its A/B testing framework for SEO experimentation as follows:

  • The team identifies a potential test and a group of pages that meet the parameters for the test. The team then applies the test value to half of the pages.

  • Within a week or two after the experiment start date, the team collects traffic data from both the test pages and the control pages.

  • Data is normalized to account for discrepancies that exist between the different test groups. The team then analyzes the average traffic increase or decrease between the test and control groups.

  • If the average organic traffic of the test pages is higher than the control group, the team implements the test value on all pages.

This approach enables Pinterest to apply hundreds of small changes over time to increase organic traffic. By applying page alterations incrementally, Pinterest can also adapt quickly to changes in how Google and other search engines prioritize content. 

Build a pSEO engine with daydream 

Building a high-performing, end-to-end pSEO engine like Pinterest’s in-house is no simple feat. It requires sizable up-front and ongoing investment. Plus, significant lead-up time to build the infrastructure required for cross-functional teams to tightly coordinate on a complex set of tasks. 

The daydream platform enables businesses to build pSEO programs without the expensive price tag. Our platform automates all of the key workflows involved in building a robust, scalable, and durable pSEO engine. If you’re interested in using daydream and joining our growing list of customers including Notion and Tome, email us at [email protected] to start the conversation.

Join Our Waitlist

Ready to make SEO work for your company? Join our waitlist.

Related Notes

© 2024 daydream Labs, Inc. All rights reserved.

© 2024 daydream Labs, Inc. All rights reserved.

© 2024 daydream Labs, Inc. All rights reserved.