Targeting Based on Category Affinity

One of the more interesting things we’re doing at OTTO Digital is pushing the envelope on content targeting and re-targeting based on source and behavior. Presently we’re working with a major retailer on delivering contextually relevant content on their homepage to return visitors (either in-session or on subsequent visits) based on what category the user has shown an affinity and interest in.

We define category affinity (in this instance) based on the first category page a user visits. This can be a visit from the homepage to a category page or a direct visit into the category page from search or any other source. When the user comes back to the homepage the content that is delivered to them is changed based on their category interest.

Currently we have created “universes” of users for our client’s six major categories that we will target content to. The idea is that as a user you become a “Women’s Shoe” person or a “Men’s Suit” person and are targeted for the delivery of content and/or products that are contextually relevant to your universe.

Over the last 5-day period 18% of the users that entered a category universe returned to the homepage where they were targeted with content relevant to their category affinity. The results of this targeting have been outstanding. There was a 31% increase in conversion rate to users that received a targeted content on the homepage vs. those that did not. Average Order Value increased 11% and Revenue per Visitor increased a whopping 42%.

An important point: There is a major distinction with this type of targeting from the more generic behavioral targeting (BT). BT takes place over an ad network and the display ads passively have to vie for user attention with other active priority of users. Affinity Targeting takes place on the client’s site and during the moments of engagement. It is based on intent and focused on “right time, right place and right message.” Classic BT can only focus on “right message.”

Targeting users to deliver contextual relevance is a key initiative for us this year and I look forward to sharing much more about these efforts over the coming months. We are working on this not only with retailers where it drives lift in revenue but with information and content providers where this drives engagement. We are working on creating universes not only from site based actions but from search behavior. Interesting results await…



, , ,



4 responses to “Targeting Based on Category Affinity”

  1. Dave Avatar

    I’m wondering how you measured results on this. How are you comparing A/B group results.
    Lets say Group A is the product affinity set, and Group B is the non-targeted set.
    Did you measure results like this:
    Group A: Target page
    Group A: Non-target page
    Group B: Non-target page
    or like this
    Group A: Target page
    Group B: Non-target page


  2. Jonathan Mendez Avatar

    We measured all return visitors to the homepage against the targeted group. The targeted group had to visit a category page to be entered into the campaign. The other group there was no such restriction. The only requirement was that they were return visitors and they hit the homepage. So in this manner it was not a true A/B. we did it this way because we didn’t want to divide the category traffic in two groups since it would have taken much longer than the 5 days to get confidence levels. the point here was to validate the concept. Now we spend our time looking at what to target instead of whether to target. A much better use of the clients time and money in my opinion.


  3. Bob Avatar

    Can you recommend a good site for Affiliate Marketing Best Practices?


  4. elalayzipFamy Avatar

    plz ant one know arbic translate that for me
    all love (kiss)


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a website or blog at

%d bloggers like this: