Kissmetrics Blog

A blog about analytics, marketing and testing

Built to optimize growth. Track, analyze and engage to get more customers.

Behind the Scenes of Behavioral Advertising

Today we’re going to look at some of the little known details and services surrounding the world of behavior advertising. And just in case the term “behavioral advertising” is confusing to you, let’s define it real quick:

Behavioral advertising is a technique used by online advertisers to present targeted ads to consumers by collecting information about their browsing behavior.

Several pieces of data may be used, such as:

  • The pages browsed on a website
  • The time spent on the site
  • The clicks made
  • The recency of the visit
  • The overall interaction with the site

All this data creates a user persona or profile that can be used to segment the audience into certain types. People with similar online behavior can be clubbed together into one segment. Then, these segments are shown ads that cater to their interests.

According to a survey by the Network Advertising Initiative, conducted with 12 advertising networks, targeted advertisements based on user behavior converted 6.8% compared with non targeted ads at 2.8%.

How Behavioral Advertising Works

Lots of platforms work by inserting a cookie into a visitor’s hard drive. Advertising networks have a larger demographic of users at their disposal. Being able to serve different sites, they can easily divide the populace into segments.

The first step in serving behavioral ads is tracking who the users are and what they do online.

Tracking refers to the process of collecting and processing data, while targeting refers to using that data to offer personalized solutions or advertising.

In order to understand how tracking works, we need to define a few terms:

HTTP request: Generally a tracking pixel or JavaScript sends client information to the ad serving company.

The user is identified with the help of a cookie in which unique IDs are stored.

Furthermore, combining HTTP request headers with UserAgent strings can lead to even more precise User Identification.

Flash Cookies: Flash allows advertisers to place Flash cookies known as local shared objects with the ability to store client information. The Information can range for preferred volume settings or even unique identifiers.

One of the notable advantages with Flash cookies is that they are outside the browser. They persist even when the user closes the browser, and they carry through when he/she opens another.

While standard HTML cookies store 4kb of data, Flash cookies can store up to 100kb.

There are mechanisms which keep track of the user even when he/she deletes the cookie. For example, say cookie named 144587 is attached to my IP. Imagine that I delete the cookie. The next time I log in, the server attaches a cookie with a different name, say 8654977, but the software knows that both cookies relate to the same person. This is called cookie respawning.

Etags: Etags are a feature of the browser cache. The cache saves a previously visited website in the memory, and if a visitor is inclined to visit the site again, the saved site is shown. Advertisers use the cache to store unique identifiers pertaining to the user.

Etags are difficult to remove because deleting the cache would make browsing slower.

HTML5: HTML5 makes use of a cookie called HTML5 local storage. Even if the user closes the browser, the cookie persists until deleted willfully. Compared with Flash and simple HTML cookies, HTML5 cookies can store data up to 5 Mb.

It is preferred because users don’t need any plugins like Flash to run it.

The EverCookie: The EverCookie uses the combined features of HTML5 cookies, Etags, Flash cookie, etc. If a user deletes one identifier, say Flash cookie, then the other identifier, say Etags, is enough to activate the EverCookie.

AOL has explained it beautifully with images of Penguin. When you visit a sports site or gourmet site, for example, an ad company sends a cookie to your computer. Later, when you visit some other site, the ad company reads the cookie and displays an ad related to gourmet or sports.

AOL penguin explanation

Online Behaviors That Advertisers Use

Let’s go into more detail on two pieces of data that advertisers use to create targeted ads:

IP address & geolocation: Targeting your audience by where they live is one of the most basic segmentations you can make. You can show discounts and sales targeted to a specific region. You also can blend urgency into the messages, such as: “This offer is only for Utah readers and expires in 10 hours.”

There are even more metrics you can add, like the person’s country, state, region, and city.

Verve, after comparing data from its 2,500 mobile targeted campaigns, found that mobile ads that are geo-aware performed 50% better than the rest of the campaigns.

In another study, Central Market saw a 4.1% click-through rate location based targeting.

Recency of visits & whether repeat visitors or new visitors: A repeat visitor can be shown related content based on what they searched for in the past.

You might have seen yourself that, when you visit some sites to search for certain things, you later see ads of those sites elsewhere.

Amazon does this. They even email users based on the products they browsed earlier.

Personyze is a platform that offers real time personalization based on user data.

How does that happen? It’s called retargeting, and we are going to learn a bit more about it.

How Advertisers Use Retargeting

When a user visits a site, he/she is expressing interest in the site. Later, the user can be shown ads related to the site.

Let me clarify with an example. I was searching for a company called IfeelGoods on Google. Minutes later, I logged into Facebook and found this ad from IfeelGoods in my news feed.

ifeelgoods

It doesn’t end with that. Days later, when I am reading a random piece on Salon, right there before me is an AdSense ad of IfeelGoods.

salon advertisement

Here are a few KISSmetrics articles that go over retargeting in detail:

Tools for E-commerce and SaaS Businesses

Here are a few tools that e-commerce and SaaS businesses can use for behavioral targeting:

RichRelevance: RichRelevance has served more than 1 billion personalized product recommendations. It has delivered more than $10 billion in attributable sales to its retail partners like Walmart, Target, Sears, Best Buy, etc. It’s the perfect solution for e-commerce site owners.

richrecs

You can sign up for RichRelevance here.

Related Products Management by CommerceStack: This is a Magento extension that integrates user data into the shopping experience by generating relevant up-sells, cross-sells, and down-sells. It’s currently free.

The below examples show you how the Products Manager can be used to add related products and manage them with a few clicks.

magento admin panel

8 magento admin panel

Visit the CommerceStack installation page here.

Personyze: Personyze is a behavioral targeting tool that collects and analyzes information on the user and makes real time recommendations. It can display targeted banner ads from a range of data like the keywords used for search, click behavior, etc.

An example of a site offering personalized recommendations with Personyze is shown below.

Personyze personalizes the webpage for each user, as follows:

  1. Greets the user with his/her name, “ Welcome Brian,” as you can see
  2. Products shown based on previous searches
  3. Down-sells and cross-sells
  4. Up-sells
  5. User guides
  6. Reduced prices to attract customers

monkey gadget

You can sign up for Personyze here.

Intelligent Cross-Sell by CNET: This tool is a boon for those in the consumer electronics industry. CNET has partnered with RichRelevance to help merchants optimize their cross-sells. The tool taps into CNET’s database of 5 million digital goods and comes up with relevant recommendations.

The example below shows a product comparison table powered with the rich data base of electronic products on CNET.

e storesite product comparison

Get started with Intelligent Cross-Sell here.

CoreMetrics Intelligent Offer by IBM: The tool generates product recommendations based on past purchases and business rules. It delivers smoothly across email, mobile, and web platforms.

11 dynamic recommendations emarketing spot integration

12 dynamic recommendation storefront

Get Started with CoreMetrics here.

Magiq Dynamic Personalization Software: Magiq builds and maintains long term customer behavior records which can be used to offer personalized content. Magiq records key user behavior like the place they are from, whether they have registered on the site, search keywords used to reach the site, etc. It also categorizes active customers as “ready to buy,” which means if the sales team pitches them with an offer, they probably will buy the product.

Swimshop.co.uk was able to achieve a 61% conversion on cart abandonment emails with Magiq.

Whenever a user abandons his/her cart, a trigger is fired. The trigger can either send an email offering a 15% discount or show the banner ads of the products on subsequent visits to the site.

Here is a video that shows the entire process of remarketing via Magiq:

Sign up for Magiq here.

Facebook interest based targeting: In a post here, Tommy Walker talks about how a True Value ad got into his news feed. In his own words: “Well … I’m 27, I’m married, I’m a male, I’m a father, and I’ve Liked the pages Saving Money, Super Coupon Lady, and DIYnetwork, among other things that would signal I’m a responsible adult living on a budget and like to take care of things around the house.” He also suggests the perfect tag cloud to target people who’d be excited about a $5 True Value coupon.

facebook interest based targeting

Get started with Facebook Advertising here.

On Facebook, people enter a lot of data that makes it easy to target them with highly personalized offers. They like pages, so that indicates their interest in certain products or businesses, they mention where they live, and so on.

Have You Tried Behavioral Advertising?

As you can see, behavioral advertising helps you provide consumers more personalized offers. It helps you get into their shoes and understand what they need. You can use the tools to better serve your customers. How have your experiences been with behavioral advertising? Leave your thoughts in the comments below!

About Kissmetrics

Kissmetrics combines behavioral analytics with email automation. Our software tracks actions of your users across multiple devices allowing you to analyze, segment and engage your customers with automatic, behavior-based emails in one place. We call it Customer Engagement Automation. Get, keep and grow more customers with Kissmetrics.

 

 

About the Author: This article was written by George Mathew.

  1. Very informative post . I appreciate your thoughtful writes. thanks for sharing sucha
    helpful thoughts.

  2. re-marketing is a very powerful tool for advertisers. I wonder where all of this is going in the future.

  3. Nice Blog For Behavioral Advertising.

    Thansk For The Post Thist Blog

3 comments

Please use your real name and a corresponding social media profile when commenting. Otherwise, your comment may be deleted.

← Previous ArticleNext Article →