Personalized Content Nice in Theory Challenging in Practice

Amid the intense focus on vehicles for marketing -- social media, omnichannel merchandising and what have you -- heightened interest in content and its role in marketing is a welcome development. Perhaps content marketing hasn’t taken center stage before because everyone does it, and always has. If what you tell or show prospective customers has always been part of your marketing efforts, what has changed to justify the increased interest?

While the goals and concepts of content marketing have remained the same, the evolution of technology and the internet have largely taken over media and delivery channels, changing the execution of those concepts in ways that are forcing a radical rethinking of how best to engage rapidly changing target audiences.

Those changes, especially in the habits of the target audiences, can be seen in several ways. For example:

  • InteractivitySocial media, primarily Twitter, has popularized short, as little as one word -- or "1" character "4" that matter -- exchanges, and with them a propensity for Internet users to avoid longer communications including marketing content.
  • Personalization: Increased dependence on interactive social media, especially Twitter and its counterparts, means that prospective customers respond best to content messages that appear to be just for them (“Welcome Back, Bill “, etc.)
  • Layering and Impatience: Give today’s twitterheads too much content and they’re apt to become impatient and move on. Layering -- showing viewers just enough to keep their interest and letting them opt for more if they wish -- is an effective way of dealing with an audience increasingly prone to impatience.

Personalized Content: Sounds Nice, But How Do We Do It?

As you inventory the ways you can and want to engage your target audience with personalized content, you face the question,“How can I create the flexible content my marketing strategy demands?”

Fair question, especially if you sell a wide range of products to a wide range of prospective customers. This is the point where getting your IT people involved makes real sense because working backward from the customer interactions you have designed, you will end up where the source content must be created, managed, revised, organized and delivered, and that can raise some serious functional and technical questions that will bedevil you if you don’t get them correct.

Manage Pieces, Conditions or Both

Of these questions, perhaps the most critical is how you will capture and store your source content. You can use word processing, database fields, some flavor of XML or a mix of all three to create a content environment that allows you to identify, prepare and deliver just what the interaction in process requires. Each of these approaches brings with it different technology, different costs, different procedures and different ways your creative people must work.

Here are some of those approaches and a perspective you can use to differentiate them for your needs:

  1. Write a complete version of your content for each type of interchange in each customer session, store them on the file system or in a content management system and select them based on the activity on your pages. This approach is straightforward for the author but suffers from all manner of complexities as your content grows and your personalization becomes more granular. It also creates lots of duplicate data -- all the common content in each document -- each copy of which must be updated every time a change is made to any of the common content. Unless your content needs are really simple, this may be one to avoid so we’ll have no more to say on it.
  2. Break your content into sub-components, store them separately, then collect and deliver them as a stream to meet the particular needs of each interaction. This approach minimizes duplication by storing the vast majority of content only once for use wherever it is needed. However, it also can create a blizzard of increasingly smaller components that make little sense by themselves, can be difficult to author and can be delivered only when selected by a “map” that identifies which components go where -- in itself often a complex undertaking. There are standard protocols for this type of “component content management” and multiple content management systems designed to handle them. This is a powerful approach but should be approached carefully lest its complexity gets away from you.
  3. Conditional Processing, as it's called in the DITA (Darwin Information Typing Architecture) standard, important because of its wide use but not the only way content can be made conditional: the DocBook standard for technical documentation, for instance, calls it Profiling. This technique, based on the use of XML attributes, tags variants within master document components so that at processing time, the desired variants may be extracted without breaking the content into ever smaller components. In DITA, there are four named conditional tag categories (product, platform, audience, revision) with another catch-all named “otherprops” to be used for anything else -- DocBook uses 14 in much the same way. This architecture is important because it allows master document components to be structured for use in a wide range of situations, all without increasing the fragmentation of the source content. There are software products available for all phases of conditional processing support, from authoring to management to conditional publishing.
  4. A Combination of component content management and conditional processing: in a component-based environment like DITA, conditional processing can be used to minimize the volume and granularity of components by making each one usable for a range of interactions. This approach, properly done, is likely the best way to achieve a high degree of personalization without undue complexity and volume of content components.

Finding Common Ground

When you sit down with your IT support staff to explore how to best generate a workable personalized content marketing approach, you must find a level of common ground that takes into account what systems and expertise they have in-house -- or can acquire -- against what will be best for the interactivity and personalization your audience will respond best to, as well as your resources to create and manage it. The key term here is “common ground” because IT often wants to use what they have and know best while you may need something different.

Technology groups and the vendors they favor often start from the capabilities of their favored software and work backward to create a set of requirements they can best handle: a little like developing the cure and looking for a disease for which it is applicable. This usually ends up failing to fully or properly meet the real requirements their users are facing and just as often ignores the resources and constraints under which the users must operate. To avoid this “solution looking for a problem” trap, it’s critical that user requirements are on the table from the outset and are fully taken into account as technical approaches and tools are evaluated and selected.

While you may not get everything you want, your functional needs must part of the decision process as to what the final technical solution will be. Plan to end up and live somewhere between functionality and technology, but don’t give up the things you need for the technical resources already in-house.

This can be testy at times, but it can be done, and you, IT and your business will better off if you can reach a level of compromise that makes sense for all.

Title image by Luke Loughead (Flickr) via a CC BY-SA 2.0 license