The personalization story –

Personalization is one of the hottest buzzwords in the industry. The problem is that few people understand what personalization technology is, how it can be implemented, and whether or not the return is worth the effort.

This series of columns will examine the questions of what personalization is and what it isn't, why it should be considered as part of the application environment, how it should be deployed as part of the end-user experience, how you can choose the appropriate technology for your environment, and how a business and financial case can be made for technology acquisition.

Personalization involves an application that computes a result, thereby actively modifying the end-user interaction. Personalization really has only one goal, which is to deliver some piece of content (an ad, product, or piece of information, for example) the end user finds so interesting that the session lasts at least one more click. The more times the end user clicks, the longer the average session lasts; longer session lengths imply happier end users, and happier end users help achieve business goals.

The coding of the algorithms derived from statistical model(s) is really the core of any personalization engine. The engine's implementation must be able to coexist in the anticipated (production) environment and -- even more important -- be fast enough to return results within an acceptable time frame. Unsurprisingly, the effort to create an accurate and fast set of algorithms is so great that software vendors in this space all patent their technology. Therefore unless you are operating within a truly unique environment, you should definitely buy personalization software instead of building it in-house.

Caveats about stating prices notwithstanding, it can be said that the software component of a personalization solution generally costs between $100,000 and $1 million but "typically" runs around $200,000. Of course, each vendor uses a different pricing model, but vendors usually base the cost on concept of use (by number of CPUs, transactions, or URLs, for example). When calculating at the total cost of ownership (TCO) of personalization, you should include the price of software purchase and maintenance, in addition to the costs of the servers and their maintenance.

What personalization is not

In order to better understand personalization, let's first take a look at what it is not. In technology terms, personalization is not:

  • Customization: Customization involves end users telling us exactly what they want, such as what colors or fonts they like, the cities for which they want to know the weather report, or the sports teams for which they want the latest scores and information. With customization, the end user is actively engaged in telling the content-serving platform what to do; the settings remain static until the end user re-engages and changes the user interface.
  • Customer relationship management (CRM): Personalization can be a key component within a comprehensive CRM strategy. CRM should take a larger view of the relationship between customer and company and help achieve high-level business goals. Personalization is the grassroots execution, figuring out the next interaction between the customer and the Website.
  • Content-serving, image-serving, database look-ups, page layout design or modification: These are all distinct technologies in their own right.

In business terms, personalization is not:

  • Headlines or features: New or important content that has general appeal is not personalization.
  • Sale items of the day: Sale items are perceived to be of interest to every shopper.
  • "No-brainer cross-sells": When a customer purchases a product, offering accessory items, such as a case for a PDA or batteries for a Walkman, is not personalization.

What personalization is

Personalization is a computation-based application that takes a well defined set of inputs and returns one (or more) recommendations for a piece of content to be immediately served to an end user. Personalization is designed into a Website: the "content server" calls the personalization application (or engine) at useful points in the visit flow (or a small part of every page).

Typical visitor flow with a personalization engine engaged

Personalization is content that is specific to the end user based on implied interest during the current and previous sessions. The technology is cognitive because it "learns" what visitors to a site want by "observing" their behavior. It has the ability to adapt over time, based on changes in a site's content or inventory, as well as changes in the marketplace. Because it observes end users' behavior, personalization has the ability to follow trends and fads.

Consumer sites that have successfully implemented personalization tend to remain quiet about their technology, almost to the point of secrecy. Most businesses see personalization as a competitive advantage (ranging from strategic to high tactical) and don't want to give the technology away to their competitors. These companies are also aware of the public's sensitivity to privacy, and they don't want their customers to misunderstand their objectives in implementing personalization technology.

The classic personalization example is's Website. The book retailer was a pioneer in the personalization space, using collaborative filtering technology to make reading recommendations to each customer based on the purchases of other customers. Amazon has devoted significant resources to the site's personalization capabilities and has implemented multiple versions of a number of personalization technologies.

Business applications of personalization technology

Let's first look at examples of inappropriate personalization implementation:

  • Sites that have a simple, well-defined business model, such as delivering winning lottery numbers or telephone book services and nothing else.
  • Content sites that are focused, have a simple structure, and contain little content, such as a local airport site that describes directions and services.
  • Commerce sites that feature a few simple products, such as coffee from a single roasting house or men's underwear from a single designer.
  • Email campaigns for follow-up sales based on product affinity, such as selling the case or battery after the initial sale.

By excluding these broad categories of sites, we can examine sites with business objectives that could benefit from personalization:

  • Information-heavy content sites, such as technical resource, financial information, or equipment manufacturer support sites. This also includes sites that need to house full documentation but could benefit by applying the experience of "power users" to the general audience.
  • Commerce sites that have a large number of SKUs, such as movie and entertainment sites or large retail sites that carry multiple lines of kitchen products. These are sites that carry, sell, and promote various types of goods and want to leverage user interaction beyond simple segmentation.
  • Commerce sites that want to leverage direct user taste and preference. These sites might ask end users for ratings on items they like and dislike and use that as the basis for personalization.
  • Commerce sites that actively carry, sell, and promote "trendy" items. Personalization allows these companies to catch and capitalize on fast-moving consumer trends without expensive and slow data mining efforts.

What's up next

Stay tuned for the next column, as we discuss specific personalization technologies, such as offline data mining, collaborative filtering, Bayesian Networks, neural networks, and natural language processing. You'll learn how each form works and how to choose the correct technology for a project that may be at hand.

ITWorld DealPost: The best in tech deals and discounts.
Shop Tech Products at Amazon