Before I get started, let me be clear, I am very much of the belief that a personalized customer experience can grow conversion by at least 20% compared to a contextually irrelevant site. Relevancy matters, so please put away your torches and pitchforks and let me clarify. I have started to see a trend where organizations are overdoing personalization on a single page. That is to say, one specific page of content is forced to carry the personalization load of many different products, categories and sub-sets of categories, making for a (let’s say) rich set of rules. You know…let’s say, “breaking the bank” set of rules.
This is when personalization has gone too far. Fortunately this can be resolved by bucketing products into different landing pages and personalizing the categories and sub-sets therein.
For sake of argument let’s say you have 10 different versions of one product or service: Blue, Red, Green, Orange, Yellow, Purple, Black, White, Magenta, Cyan. If each of those have 10 personalization options under them, combinations of component personalization, subsets, etc…managing them starts to gets unwieldy.
Here are 7 reasons I think you should divide content up into separate landing pages
Example: acme.com/widgets/blue, acme.com/widgets/red, and so on.
- SEO: Web crawlers will read the default content of each page and take into account the URL. If all lumped into one giant personalized page, that URL equity is lost because it won’t contain the specific color in the product URL. Also, the default content may be so agnostic that no one product ever sees content equity. While I agree that the default page in the lump setup could include a component listing ALL the color products, it will still not likely be as good as a dedicated landing page with multiple mentions, and a general structure liked by Google and other crawlers.
- Manageability: Content teams can divide and conquer per bucket and focus on personalizing (and testing) their specialized content. Workflow would likely be cleaner and focus will certainly lead to better content. Not to mention, less likelihood of errors and personalization conflicts.
- Reporting: It will be easier to visualize trends in Sitecore analytics with separate landing pages. Develop a report to show a line graph or pie chart with all buckets to visualize which page is performing better and then test/personalize within in to improve each bucket. This will be a quicker way to see which color product performs better and reacts better to which test and what personalization choices. With the addition of facets and tagging, drill down further from a more focused starting place. It will also help if divided up to see which content author is performing better on page/bucket improvement.
- A/B/N Testing: Testing across all buckets and all possibilities would yield a very long test time, especially if doing multivariate tests. By parsing out into buckets, test times will be lower and Sitecore will better be able to determine traffic/bounce (and other factors that drives test timing) per bucket instead of each bucket criteria being influenced by the others.
- Speed: The more complex the rules are for the giant lump master page – especially if it includes rules directly on content – page load will be affected, which can certainly influence SEO, user experience, bounce rate, etc. With many different persona combinations on a page, increases in the number of cached variations have to be managed, and can increase server load.
- Content: Related to Manageability above, but worth calling out, focus on each bucket will lead to better clarity of content and clearer testing and personalization vision.
- Profile/Patterns: Bucketed color/products will allow for more refined placement of Profile cards on each page and in the long haul reveal more refined patterns for further personalization and cyclical improvements.
Consider breaking out bucketed pages before leaving all of the content personalization to the Conditional Personalization control. It will yield better SEO performance, a cleaner management of workflow, and probably lower blood pressure #notadoctor.
I invite you to share thoughts, additional reasons and constructive opinions in the comments or tweet @rickbperficient.
Agreed!