[You may have noticed that my blog was down for a big chunk of this past week - apparently Typepad was suffering from a MAJOR Denial-of-Service attack. Seems like things are back to normal now - keeping my fingers crossed!]
Personalization is the heart of digital. As a direct channel, digital rewards personalization in almost every form, and there are no shortage of forms. In my last two posts, I’ve described an array of personalization strategies and a similarly long list of analytics techniques to drive those strategies. Originally, I’d intended to match the two lists, but on reflection, I’ve decided to tackle, from a business perspective, the factors that drive choice around the personalization strategies.
For convenience sake, here’s a list of the various personalization strategies I cataloged:
|
Content/Product Related Personalization |
Content-History Strategies |
Next Best Offer (NBO) Strategies |
|
Trigger-based Strategies |
Last Behavior Strategies |
Threshold Strategies |
|
Time & Event Strategies |
Offer Matching Strategies |
Filter-Based Strategies |
|
Crowd-Sourcing |
People-matching and Nearest Audience Strategies |
VoC Based Strategies |
|
User-Directed Strategies |
|
|
With so many different strategies, how do you go about deciding which ones to pursue in your business? It seems daunting, so I thought it might be easier to take the conversation up one level. There are some very basic and distinct types of personalization that commonly exist. These types encapsulate (I think) the personalization strategies listed above:
Personalizing a Marketing Drive: Optimizing a specific drive within a broader experience
Personalizing an Experience: Creating a unique creative environment for a visitor
Suggesting Relevant Content or Products: Optimizing User Selection inside a stable experience
Tuning an Offer: Changing the parameters, size or definition of an offer based on the user’s value, price sensitivity, existing relationship or interests
Every single one of these types might be appropriate for a business; however, there’s nearly always one type of personalization that is dominant. If you’re an online publisher, you might think about it this way:
Personalizing a Marketing Drive – Can be used to determine whether I show a Paywall or a Paper subscription offer to a visitor
Personalizing an Experience – Can be used to optimize the default elements on the home page for every user
Suggesting Relevant Content – Can be used set content recommendations for additional reading or viewing
Tuning an Offer – Can be used to set the type of Paywall offer given to a regular user
If you’re just rolling out a paywall, then personalizing the marketing drives and tuning the offer might be the most important personalization’s to tackle. More often, suggesting relevant content is going to be the dominant form of personalization. Content recommendations get to the heart of what online publishing is about – getting as many users as possible to consume as much content as possible.
For a large-scale online retailer, the personalization types might break-down like this:
Personalizing a Marketing Drive – Can be used to determine whether I show men’s or women’s fashion on the home page
Personalizing an Experience – Can be used to optimize a registry or holiday shopping experience
Suggesting Relevant Content – Can be used set product recommendations for other, new or like products the visitor might consider
Tuning an Offer – Can be used to set the discount offered or the type of products offered on a discount page
Once again, all of these are useful, but suggesting relevant products/content is likely to be the dominant form of personalization.
Does this mean that Suggesting Relevant Content or Products is always going to be dominant? Absolutely not. What both these examples have in common is very high content cardinality. Most online publishers generate LOTS of content. Most large-scale online retailers have tens or hundreds of thousands (if not millions) of SKUs. So the key usability challenge on those sites is getting the visitor to the right content or product for them.
That just isn’t true for many other types of sites. An insurance site may only have one or two different products. Even full-service financial services companies may only offer a few dozen products online. What’s more, visitors arrive at the site knowing exactly which product they need. If I show up at brokerage site I usually know whether I’m looking for an IRA, a trading account for a 529. Suggesting products to me just isn’t very useful.
For an insurance site, Personalization types might look like this:
Personalizing a Marketing Drive – Can be used to determine whether I show an SMB drive or a Home Insurance campaign on the Home Page
Personalizing an Experience – Can be used to optimize the quoting experience
Suggesting Relevant Content – Can be used to determine whether I surface comparison content or explanatory content about a product
Tuning an Offer – Can be used to set the coverage options I recommend to a visitor
Once again, each of these types is potentially useful, but the heart of the Insurance challenge is usually around quoting not content. So the personalization options that are dominant are probably around Personalizing and Experience and Tuning the Offer.
Let me repeat it one last time - depending on the nature and shape of your digital business, any one of these personalization types could potentially be dominant. A cause-based non-profit, for example, might look at personalization this way:
Personalizing a Marketing Drive – Tuning the emotional appeal to the audience
Personalizing an Experience – Giving members unique content
Suggesting Relevant Content – Suggesting topical content to maximize engagement
Tuning an Offer – Can be used to change the order and size of the ask
Of these, Personalizing the Marketing Drive is likely to be the dominant personalization method. It gets to the heart of the digital challenge – getting potential donors to respond viscerally to the cause.
Given these examples, can we work inductively to the underlying principles? Perhaps.
Cardinality is clearly important. Where user-choice is very high, the digital challenge is to navigate that choice. All large scale ecommerce sites, online publishing sites, most travel booking and advisory sites, sites like Netflix or Opentable all face the same challenge: too much content to fit into any conceivable real-estate or basic search strategy. So the secret sauce for all these types of sites is personalization based on content or product suggestion. Cardinality is a key element to selecting the right type of personalization to pursue.
After cardinality, you should consider whether your central digital problem is getting different kinds of people to complete the same journey or providing fundamentally different journey’s depending on the type of person. Think about this as Mono-vs-Multi Journey. If you’re selling auto-insurance, you want every visitor to take the same journey. In such cases, you’re not likely to deeply personalize the experience. You’re going to focus on tuning the marketing drive. On the other hand, if you have radically different types of visits, then tuning the marketing drive is probably going to be much less important that personalizing the actual experience.
What about tuning the offer? When I used to do DR Marketing, we always said that it was Offer, Targeting and Creative that drove performance. And of these three, offer was by far and away the number one factor. But the truth is that many businesses can’t really tune the offer. Economics, brand policy, regulation, or simply the nature of the product may make the product offer fixed not variable.
Where offers can be tuned, then offer personalization will often dominate over Personalizing Marketing Drives as the best path to high performance.
So that’s it – my handy little internal decision-tree for deciding at a high-level what type of personalization to pursue. Is high-cardinality choice the key to digital success? Is the business focused on one journey or many? Can we tune the offer?
Put this way, it doesn’t seem so hard and I feel better about going on to the next step – matching personalization strategies to these basic types. From there, I’ll tackle the matching to analytic techniques. Promise!
Comments