How does the model work?

Made With Intent uses a single universal model that is trained across all clients and eCommerce sites.

The Intent Base model is what we use for our product and all backend inferences. It's a Large Language Model (LLM) using the Tensorflow architecture to generate all the output we see in the dashboard.

The model reads in a text sequence of a user's recent events, including pages, actions, clicks etc. As it's an LLM and trained on web tagging, the inputs can be (literally) read sequentially like the journey of a user, but outputs are weighted based on what it believes the most important things are, such as operative words and the sequence of events etc.

An example would be somebody hitting a well-performing product page. The model will then learn that this page means something significant, as well as the events leading up to/after landing on it.

Do you support non-English languages?

Yes! Whilst our model tracks text from each user’s browser, it has been designed to read the code elements from the website directly. The language inputs only make up a few of the +250 the signal we track.

Why do you use a universal model over a site-specific model?

The Intent model is a single model and is trained on all our online retail clients. We found that a universal multi site trained model worked best across retailers than individual site specific approach.

All our websites are continuously fine-tuned to our models to get the optimal performance for all our clients. Our models are continuously iterated on and optimised for accuracy and performance. This is only made possible by only having to do this for all sites at once.

A universal model trained across multiple retailers increases the accuracy over any model trained only for a specific website due to the sheer volume of data availability.

A universal model means a client benefits from user behaviours not yet experienced on their site specifically. For example, imagine launching a new product range. How can we understand users’ intent if this is totally new behaviours and sequences? The universal model has seen hundreds of millions of interactions and would have no problem in interpreting users’ intent, even if it's new to your site.

Despite continuous fine-tuning to clients’ data, if a client only has a small sample of data from a particular segment of users then they will have a hard time confidently interpreting the behaviours and intent. For example, only a small amount of traffic is sourced via YouTube and conversion rates for this group are therefore highly volatile from week to week. The universal model will be able to confidently interpret these users.

Is my dashboard and scoring specific to my website(s)?

The short answer is yes. Despite our models being trained across all our retailers the model predictions are specific to the selected site in the platform. You will only ever see the data that’s specific to you, benefitting from both the universal model’s power but specific training on your data.

Is my data still private in this universal model?

Yes. All client data is partitioned in our data stores. Then anonymised and scaled prior to being sent for model training.

The model itself holds no private information. It only contains a mathematical algorithm of numbers and weights designs to output a score from a pattern of numerical data.

Is the model up to date? How often is it trained?

We go through a quarterly training cycle with our models to ensure they are continuously fine tuned and monitored for optimal performance.

How live is the data in my dashboard? How often is it updated?

Data is currently available up to the previous day within the Intent Performance Score and our Segments. However our real-time data health check provides you with a view of the metrics we are receiving, in real-time