Tagging your content and building visitor segments can improve reporting on the performance of your content and enables you to increase visitor engagement. For our clients, we have setup a system to automatically classify content with consistent tags and build visitor segments to achieve these improvements.
Websites often serve multiple types of content such as blog posts, news articles, brochures and so on. As the owner of a website or app, you are interested in how your content performs and serves the different visitors of your platforms. Or more precisely, people will visit your website with a certain goal or intent. They might, for example, be looking for information about a ballet show. A good website will quickly provide the users with the right content necessary to reach their goal. Therefore, an important question is, are your visitors able to reach their goals through your content?
Plainly looking at regular performance indicators, such as content views or the time that visitors interact with your content, might be insufficient to determine how your content performs and whether visitors reach their goals. It could happen that your content is fine, but you are showing it to the wrong visitors.
This could for example be the case when you are analyzing content on ballet over a certain period. However, during this period you have a campaign running which attracts a huge amount of visitors to your website who are interested in soccer. Content on ballet would then probably underperform if you would not take these visitor interests into account. Namely, visitors that are interested in soccer might not read content on ballet or interact with it very superficially. Reporting on how your content serves your visitors should therefore incorporate your visitor interests as well.
The other side of the coin is that knowing the interests of your visitors allows you to serve personalized content to your visitors. Once you identify a visitors interests on your website, you could for example serve him or her with content about a ballet show. Serving your visitors with more relevant content will improve their experience with your website and increase their engagement with your platform. Recommender solutions like Harvest Combine can take this personalization even further, however we could write a whole separate article about that (and we probably will).
The first step in the program described above, is correctly classifying your content. Many companies are having difficulties when it comes to tagging their content consistently. It is often the case that content is written by multiple editors, and that each editor tags their content differently. Taking this task away from the editors and automating the process can be a good way of achieving consistency (you might want to involve the editors by letting them “check” the tags the algorithm came up with).
We have built such a tagging system using the Harvest Store. The Harvest Store gives you real time access to the raw data that you collect from your website (and other sources). The Harvest Store periodically checks if there is newly published content. This content is then automatically acquired from a content feed and gets tagged using a natural language processing solution. The resulting content tags are then stored and made available as new “data” in the Harvest Store. With this setup, the content tags are available to any report you might see in the Harvest Store, creating the possibility to split your informations, based on content tags.
Having your content tagged consistently and having these tags widely available allows you to take the next step: determining your visitor segments. For this, it is crucial to have access to well-organized data on visitor interactions with your tagged content. Through Harvest Collect and Harvest Store, we can set this up easily. Subsequently, a scheduled search within the Harvest Store periodically analyses the resulting data. Roughly speaking, a visitor is then classified as being interested in a certain subject if it is observed that he or she interacts more than average with the corresponding content. These user interests are stored coupled to a user identifier and again made available in Harvest Store for reporting, as well as through an API endpoint to be used for personalization purposes.
In conclusion, the Harvest Store enables you to build a framework to automatically and consistently tag the content on your website or app. It also provides you access to the raw data on visitor interactions with your content, which you can use to build solid visitor segments. This enables you to improve reporting on content performance and increase your visitor engagement by serving more relevant content.