If you want to maximize your performance on any digital platform, you need to align your presentation with what its users are searching for at any given time.
This is particularly relevant on Pinterest - now framing itself as a product discovery platform, Pinterest is increasingly working to better connect product matches with user interests, in order to help prompt users to buy, and build its eCommerce credentials.
Given this, if you want to maximize your Pinterest performance, you need to understand how Pin search works, and how you can create Pins which align with search and discovery within the app.
Pinterest has provided various insights on its systems in the past, and this week, Pinterest's engineering team provided another overview of its classification processes, which may help to further improve your understanding, and your Pin creation strategy as a result.
Over on the Pinterest Engineering Blog, the Pin team has outlined how its Pin2Interest - or "P2I" - process categorizes Pins based on the available data.
For example:
Pinterest says that for this Pin, its P2I system would extract the following classifiers:
- (3, dogs, 0.83)
- (2, mammals, 0.7)
- (1, animals, 0.82)
- (3, skiing, 0.83)
- (3, snowboarding, 0.79)
- (2, winter sports, 0.75)
- (3, snowboarding, 0.79)
- (2, winter sports, 0.75)
- (3, ski trips, 0.76)
- (2, travel ideas, 0.65)
- (1, travel, 0.71)
The data displayed here is in this order - "taxonomy level", "label", "score".
- Taxonomy level relates to popular concepts that appear in Pins, with each having around "10 levels of granularity, with 24 top-level concepts". So, for example, 'Home Decor" might be the top-level concept, and within that, there would be a range of more specific sub-topics.
- Label relates to the specific topics and/or subjects identified within the image, which comes primarily from the accompanying Pin text, but can also be inferred by visual classifier
- And finally, score relates to how much of a match Pinterest's systems think the Pin is for each identified label, based on the information provided.
It's worth noting, too, that some of the topics and matches are inferred based on Pinterest's overarching understanding. For example, as you can see in the above listing for the Pin image, "mammals" is listed, as is "snowboarding", both of which are not mentioned in the Pin description, but are contextually related to "dogs" and "skiing". This is how Pinterest's system matches up elements of the Pin to ensure it measures for relevance when incorporating the various factors.
The important note here is that you need to use Pinterest's "Guided Search" within your Pin preparation and planning to see which topics and searches relate to your specific industry. As with Google SEO, you don't want to go overboard in mentioning every key term within your Pin descriptions - which, as you can see from this process, is largely unnecessary anyway - but you do want to ensure that relevant terms are included, helping Pinterest's system better match your content to relevant user queries.
This is also how Pinterest displays relevant ads - in relation to the above skiing Pin, Pinterest provides this example of related ads displayed alongside it.
As explained by Pinterest:
"The P2I signal is used to enforce interest targeting by making sure the query Pin shares the same interests as the ones targeted. In the example above, the two Promoted Pins circled in red share the same Travel interest as the query Pin, while other candidates without any overlapping interests have been filtered out."
In this way, Pinterest can show users more relevant ads for each query. This example is not highly targeted, using "travel" as a match for relevant ad content, but it shows how Pinterest's system seeks to display more relevant content.
Essentially, there's no secret trick or hack that you can use, but by better understanding Pinterest's matching systems, you can improve your Pinterest marketing approach, and ensure your Pin content is displayed to interested users.
You can read more about Pinterest's P2I system in this post.
