Website Retargeting

How Activity Bias Screws With Conversion Rates

If you’re in business online, you’re in the business of getting visitors to your website.

You spend hundreds, if not thousands, of dollars on internet marketing – making sure those Ads appear in the right place, at the right time and with the right message.

Your conversion rates are encouraging and you call your campaign “a brilliant success!”.


New research suggests this may not be entirely the case.

You’re in the Business of Persuasion

Whether you’re a Marketer, Employer, Manager, Politician or Parent, you’re in the business of persuasion.

The term ‘Activity Bias’ originates from the world of Behavioural Economics where its meaning relates to the situation when an individual, given the choice of taking action or doing nothing, chooses to take action.

Recently the term ‘Activity Bias’ has been turning up in research on user behaviour when interacting with online advertising. Here it is being used to explain how marketing analysts and advertising agencies are inadvertently over estimating the cause and effect relationship between viewing an online ad and sales conversion (buying).

The research has identified three significant hidden variables that are not currently accounted for when assessing return on investment (ROI) and other success metrics for online advertising campaigns.

Retargeting – The Ads that Follow You Around

Retargeting Ads implement behavioural targeting using cookies and make a bunch of assumptions about your personal preferences based on pages you’ve viewed or things you’ve clicked on. Many people find them creepy and their use is often described as Ad Stalking.

Chances are good you’ve experienced this yourself – after visiting a particular website you may have found that almost every where you go you see the same Ad or an Ad for the same business/product/service; this is retargeting Ads in action.

In terms of their impact on Activity Bias…

It is difficult to isolate data that conclusively defines the purchasing influence of a visitor to your website.

Is it that the visitor has specifically visited your website or is it because they have seen your Ad on some other website?

It’s not clear-cut because the visitor’s browsing activity determines exposure to an advertising campaign. I.e. It might be the most recent Ad campaign or one that was run some time in the past.

In plain speak – the user would not see your Ad unless they had been on your website in the first place. So was it their visit to your website or your Ad that convinced them to buy?

Assuming that clicking on the Ad link to return to your website to make the purchase has problems too, how can you know with certainty that they didn’t click the Ad because they were too lazy to type in the URL?

Browsing Behaviour is Unpredictable

There is a level of abstraction between how users behave and the reporting your analytics software shows you.

For ease of use and to make complex data more user friendly, most website analytics programs provide easy to consume and understand reporting charts. Making it relatively simple to understand in averaged and general terms what large numbers of people are doing when visiting your website.

Group behaviour looks consistent, but individuals are chaotic.

The dilemma here is that these reports and tools take away a lot of the individual human chaos and lull us all into a false sense of calm and constant engagement. This is very far from the reality of how people use the Internet and visit websites.

Think about the last time you were online when you were just generally looking around. May be it was in the evening when you had no time constraints and there wasn’t anything in particular you had to be doing, so you thought about checking out the latest technology releases and found yourself on review websites and price checking.

Now recall a different time when you needed to get last minute flights and accommodation booked.

Chances are that for the last minute flights and accommodation scenario, you fired up your web browser went to at most five websites and got what you needed. A rather quick and intense online engagement.

Compare this quick experience with when you had lots of time to browse. Not only did you spend more time on each page you viewed it’s highly likely that you looked at more than one page per website you visited across a much larger number of websites.

Standing out in the crowd

This is point two. Users do not browse the web in a consistent manner over time. In fact, users show large variation in their browsing behaviour over time, browsing very different numbers of pages from the one day to the next.

So when users are having a high consumption day (viewing lots of web pages) they are even more likely to encounter you Ad and see it more frequently through out that day than on other days when they’re relatively inactive online.

3. Lots of Browsing on Your Website Means Lots of Browsing on Other Websites

The third and final hidden variable that’s often over looked is to do with exposure and competitive impact. Research has found that someone who browses a website more than usual on a given day is also likely to be browsing other websites more than usual.

This is no different to a billboard that advertises a restaurant increasing sales for all the restaurants in the area, or an Ad in a supermarket promoting a brand of cheese that increases that brands sales by reducing the sales for other brands of cheese.

Transactions will continue to occur. Sales will continue to be made. Your interest is in knowing which business benefits most from the additional exposure – Yours or someone else’s.

What Knowing About Activity Bias Can Do For Your Business

You may have gathered, Activity Bias is not a new phenomenon. But it is gaining wider attention as technologists, marketers and C-level executives develop an increased awareness of it’s role in online transactions and marketing.

The reason why it matters to you?

Knowing that activity bias exists in your own interaction with clients, partners and staff helps you recognise that there are better questions you can, and should be asking when reviewing your analytics data, assessing the success of a campaign and working out if a spike in activity on your website is to your or your competitors advantage.

We make our world significant by the courage of our questions and by the depth of our answers. – Carl Sagan


  1. Internet Activity Bias Causes Lumpy User Behavior
  2. Here, There, and Everywhere: Correlated Online Behaviors Can Lead to Overestimates of the Effects of Advertising
  3. Are Cognitive Biases on the Increase?

About the Author

SemantiaSemantia is a strategic technology services and advisory company. Through its specialty business divisions the company provides a comprehensive range of software development, systems integration, semantic web development, business process outsourcing, digital marketing, online marketing automation, conversion rate intelligence and business IT insights and advice. Semantia collaborates with clients to deliver solutions optimised for business success including increasing performance in sales and marketing, reducing errors and processing times for data & identifying new insights from analytics for leaders to make better informed business decisions.View all posts by Semantia →

Semantia Launches Calc123
A Calculation Engine for creating high-performance SaaS solutions using Excel®



Lost your password?