Overview: Blueshift's Recommendation Studio, built with patent pending Artificial Intelligence technology, is the most flexible and transparent system for building recommendations that can be used by non-technical marketers. Blueshift automatically computes various forms of recommendations: based on events (re-targeting), collaborative filtering, affinity based, trending items, catalog change based etc. These recommendation types can be sorted, filtered and combined in different ways to suit your business needs. Recommendations can be configured from a visual studio that offers previews for different users, and once configured, can be attached to any email, web or mobile template with 1 click.
1. Are unique recommendations available for every user? Is this a native capability or through a partnership/ additional integration?
Yes, unique recommendations are available and can be easily configured by non-technical marketers. Recommendations are powered by Blueshift's patent pending technology.
2. How are recommendations generated by your system?
Blueshift generates different types of recommendations from 3 types of datasets: events, user attributes and catalog. Various forms of recommendations are automatically computed and can be sorted, filtered and combined in different ways in a visual studio.
3. What forms of recommendations are available?
Blueshift offers a comprehensive range of recommendation types: based on events (re-targeting), collaborative filtering, affinity based, trending items, catalog change based etc. All of these types can be combined, sorted and filtered in unlimited ways.
4. Describe how recommendations can be deployed in email.
Recommendations can be deployed with 1-click in any email template that has dynamic blocks. Simply click the “Products” tab next to the template and select the recommendation type you intend to use.
5. Describe how recommendations can be deployed in mobile app push notifications.
Recommendations can be deployed in mobile push notifications by associating them with 1-click to the templates.
6. Describe how recommendations can be deployed on websites.
Recommendations can be deployed to websites either using a fully rendered unit serve by Blueshift in an iFrame or by calling on the JSON endpoint of the Blueshift API. In either scenario, you would follow 3 steps:
- Configure a slot: The slot defines a certain area of the website where you intend to use the recommendations (e.g. homepage top banner). You could serve different types of recommendations to different segments of users within the same slot.
- Configure a template: The template can be an HTML fully rendered template or a JSON equivalent that can be accessed through the API
- Create a live content campaign: The campaign associates the slots with different segments and templates
7. What kind of latency can we expect for website recommendations?
Our recommendations API is expected to return results within 350ms and fully rendered widgets are returned in 500ms.
8. Can we import recommendations from our in-house team? If yes, describe how.
Yes, for certain specialized cases, you may want to import recommendations from your in-house teams. Blueshift enables your teams to publish these recommendations either using a file import or through bulk API calls. Once published, these recommendations will be available to your marketers in the Recommendation Studio.
9. Can we combine multiple types of recommendations into 1 email, with some recommendation blocks coming from our team? If yes, describe how.
Yes, Blueshift supports the notion of "blocks" of recommendations, with each block being powered by a different recommendation scheme. Later blocks can refer to elements in previous blocks to build a story, as well as avoid duplication of content.
10. Does your solution offer a visual studio to preview recommendations and add filters?
Yes, we are unique in the ability to offer marketers the ability to configure recommendations through a visual studio.
11. Does your solution offer predictive scoring on customer actions? What scores are available?
The solution uses AI to compute scores that indicate a user's likelihood to complete certain actions, as well as user affinities and preferences. Predictive scores are offered as a premium feature. Scores can adapt to your business needs. For instance, some of the types of scoring offered include Purchase intent scores, Churn scores, Category affinity scores etc.
12. What kind of analytics do you offer to understand the predictive scores?
Blueshift offers visualization to help marketers understand the predictive scores:
- Visualization of the correlation between score percentiles and the desired goal being modeled
- Visualization of the top few data elements that influenced the score most, as discovered by our Artificial Intelligence system.
13. How often are scores updated?
Scores are updated daily by default. Faster options may be available depending on the nature of data and the specific use cases.
Updated almost 3 years ago