Personalization - Content recommendation - The Future
In this series about Personalization, I covered some of the steps that can be taken on a personalization journey. From personalization achieve by handcrafted segmentation, to a one-on-one personalization experience done using the wizardry of artificial intelligence algorithms. In this article, I want to cover some ways we can help our customer move forward on their personalization journey and how we can take advantage of data to go even further.
Ways to user personalization
What are ways companies are using real-time, 1-to-1 personalization to drive engagement and conversions?
Here are some examples:
- Tailor your homepage for different visitors from the information we have about the users. We can taylor the homepage to be more relevant to the visitor, even if they are still anonymous.
- Capture email addresses at the right time, and target email capture campaign for specifics users using visitor groups.
- Make a strong first impression by presenting users with relevant and helpful messaging in the moment to engage them and reduce the bounce rate.
- Help prospects discover and engage with relevant content by promoting the most relevant assets (article, video, e-books) to each person individually, engaging prospect and guiding them along their journey.
- Recommends products to visitors to guide them along their buying journey.
Those are all ways you could implement in your personalization strategy. The best way to apply those strategies is to use data and algorithms as much as you can. Unless you are using data to help guide you with your personalization strategy, you will mostly be slowed down by some HIPPO-based decisions. HIPPO stand for HIghest Paid Perso Opinion. Personalization strategies decided solely by people without being based on data analysis will be based more on feeling more than on actual metrics and facts. This could prevent you from seizing opportunities in your market and it’s very hard to have some objective conversations on your strategy without knowing what is working or not. So, the more you lean on your data to make some decisions, the better are your chances to get to your expected outcomes.
Experimentation
One way to get relevant data that can be applied to offer a better personalization if to perform some experimentations. Experimentation as its roots is the action of experiencing, learning and discovering from personal or scientific experiences. By having an easy way to try, collect some data and analytics and applying those to an online experience, we could achieve an even better personalization experience. With the integration of the Episerver and Optimizely platforms, we now have the best of both worlds. We have a CMS and Commerce platform, integrated natively with an Experimentation platform, which allows us to run experimentation natively on our site content. In such an implementation, the easiest way to perform some experimentation would be in the the form of A/B testing. A/B testing can be applied by delivering different experiences to different sets of visitors, collect the behaviours of the visitor and analyse which experience got the better results according to our targets. We can then apply personalization based on the results of the experimentation and deliver experiences that are more suited to our visitors.
Data Platform
So, by applying all those personalization techniques, we gathered a lot of data about our visitors,
collected in the form of:
- data stored in the Visitor Intelligence Profile
- data tracked and stored in Analytics solutions
- data that we track on users and content from a Content Recommendation implementation
- data gathered from experimentation
This makes a lot of data that we could use to understand our visitors. But as of right now, it’s spread all over the place. If we could centralize this data, we could take advantage of all of it and provide a more relevant experience to our visitors. This is where the Optimizely Data Platform (ODP) comes into place. ODP unifies the customer profile under one centralize data warehouse. This data is available for creating segment, defining advance reporting and activating on this data to achieve your desire outcomes. This data also can be used to render more relevant experience to our visitors, by using this centralized source of data to achieve personalization.
Data Core Service
But customer information is not the only information that is tracked and analyzed on a website. Assets, behaviours and customers information are gathered all over the visitor experience.
We have information about:
- content, product catalogs, orders
- events and interactions
- experimentation results
- customer profiling
What we would then need is to centralize all this data into one single warehouse, where we could harmonize, understand and act on this data. This is where Data Core Service comes in. Data core service is the « connective tissue » that unified the DXP. If connects data from your website, B2B or B2C commerce site and experimentation results into one source. All of this allows you to have dashboards and analytics covering your full platform and act based on actual metrics. All this data would be available to your Optimizely product, but it could also be exposed to your BI using Snowflake to be used outside the websites.
So, no matter where your client maturity is on their personalization journey, you can aim to take the next step by gathering a little bit more data, try to have it as much harmonized as you can and act on this data to deliver a relevant experience to your visitors, instead of letting some human feelings define your personalization strategy.