Ratio - Header
Blog Sitecore

Content journey personalisation on Sitecore - what's the impact for you?

10.10.2017

 

As many of you will know, Ratio does a lot of work with Sitecore customers implementing the marketing capabilities of the platform, but it's left us with very little time to take a step back. However, I wanted to quickly share some experience and results we have had with content journey personalisation, which is something we've been asked about a few times in terms of the effectiveness versus effort of this type of activity.

Firstly, what do we mean by "content journey personalisation"? When we look at client analytics data, especially in the B2B sector, we see that usually there are several visits a user goes through, with each of these steps having different levels of engagement with different types of content until they finally convert (or indeed there may not even be a defined conversion point for a business where sales are primarily conducted offline).

Therefore one of the activities we undertake is mapping content against visit number so that we can try to:

1. Gradually influence their journey by driving them close to call-to-action based on what we know about the average visits before converting

2. Drive them quickly back into the sales funnel based on their last point of departure

3. Keep the journey and messaging relevant 

A typical content journey funnel might be:

  • Visit 1: user views a product or service page and leaves. We build segment messaging around this as well using things like GeoIP and business name to make the experience relevant and direct them to relevant content so that we are influencing from first visit - more on that next week
  • Visit 2: user served homepage banner showing benefits of product or service
  • Visit 3: user served homepage banner directing them to a call to action

Pretty simple in theory, but there's obviously a lot of work that goes into precisely what we are showing to who and at what time, content creation, as well as logic around what we do if they have already converted or how long we maintain the journey for.

For one of our clients we also had to create custom personalisation rules based on goals. If you've ever used the Sitecore rules engine you'll know that it's often not really capable of dealing with real world marketing and journey scenarios, so we built out a rule based on goals or events triggered in previous x number of visits to allow us to more accurately personalise based on engagement.

We then implement this and test it over a period of month. A word to the wise: the personalisation testing engine in Sitecore has been something of a challenge and doesn't seem to actually measure test results as you would expect (we are fighting this battle at the moment), so we have been doing analysis through Google Analytics to more accurately measure impact.

So to cut a long story short, does it work? Indeed it does. Where we have trialled this so far on a couple of client sites we have seen anything up to a 20% level of engagement (users who have been served the banner and then clicked through) just with personalised homepage banners depending on the journey stage, and on one client that has also led to a small but valuable increase in lead conversions.

As an aside, we are also extracting data about who these engaged users are, and providing this to the sales team as a lead-scored report - invaluable for those organisations I mentioned before where the phone is mightier than the form, and it helps us to demonstrate the value of the website to the sales process.

We are phasing in more data collection and analysis of this to allow us to do full end-to-end attribution for these content journeys, and I'll blog more about that as we have it, but I wanted to share some of our experience with you if you're thinking about embarking on this route yourself. It takes time and effort but it does work.

Next week - Account Based Marketing and our experience, which is something that is coming up again and again. We're super excited about it... Happy weekend!