— April 4, 2018
With so much buzz around “personalization” these days, it can be tough to find the best approach to actually getting started. How do you set your goals? Should you use rules or machine-learning algorithms? What data sources do you need? What channels should you personalize? What format do personalized experiences generally take?
These are all foundational questions that you should be able to answer if you are planning a personalization strategy. We sought to answer these questions in our newly updated version of our popular eBook, The Ultimate Personalization Planning Guide. We have made it available ungated on our website, so check it out if you need help answering any of these questions.
And for a quick glance at the path to successful personalization, check out this handy infographic we’ve put together.
As with any new marketing initiative, it’s important to begin by setting your goals. Personalization shouldn’t be viewed as a separate initiative from the rest of your marketing and customer experience, so it shouldn’t be too difficult to determine your goals. Your personalization goals should be an extension of your business goals. Increasing conversions, decreasing bounce rate, driving upsells and increasing retention are all common goals.
Good data plays a critical role in successful personalization, so the next step is identifying your data sources. Cross-channel behavioral data (such as content viewed, time spent, downloads, etc.), attribute data (such as location, industry, loyalty program status, etc.) and explicit data (such as from surveys or forms) can all be used to create a clear picture of each individual’s affinities and in-the-moment intent.
Next, you need an understanding of the personalization approaches. Rule-based personalization allows you to target broad or narrow groups of people with a specific message or experience. Machine-learning algorithms allow you to dynamically present the most relevant content or experience for each and every visitor. Each approach has its ideal use cases and most marketers use a combination of both in their personalization strategies.
You also need to work on defining your personas and journeys if you have not already done so. Document what you know about each of your key buyer types, and map out the steps they take in their research and buying processes. This exercise will help you when you begin planning your personalization campaigns.
Next, begin segmenting your audiences. In what meaningful ways can you categorize your site visitors, mobile users, customers, etc.? These segments will be used for analysis (so you can understand the key differences between these groups) as well as in your rule-based targeting campaigns.
A good personalization platform will give you the ability to work on selecting and refining your machine-learning algorithms. Choose your base algorithm, add in filters by including or excluding specific criteria, include boosters to account for each individual’s preferences, and include variations to keep algorithmic experiences fresh. This approach is critical to giving you control over your machine learning-driven experiences.
Next, you can start designing your cross-channel personalized experiences. Consider how personalization can be used to affect experiences on your website, web application, mobile app or site, email, on-site search, digital advertising, or in-person channels.
Finally, make sure you are continuously testing, measuring and iterating on your campaigns. Ensure your campaigns are effective by measuring their performance against a control and always strive to find ways to improve your personalized experiences.
Of course, before you can truly get started with personalization, you need the right partner. To learn more about what to consider when selecting a personalization platform — and for more detail on each of the topics covered in this blog post — download The Ultimate Personalization Planning Guide.