Last Chance! Can warnings reduce online spending?

Benjamin Timmermans, Eva Zinger-Mityok & Monique van Maare
Posted on March 01, 2018

The temptations of buying online 

Shoes, concert tickets, electronics…  These are only a few of the increasing number of things we buy online. Our worldwide consumption of online goods has increased to 1.9 trillion in 2016 and is expected to grow to USD 4 trillion in 20201. As companies figure out how to digitally entice us with their latest products – fuelled by powerful analytics that improve products’ discovery and their attractiveness – it is increasingly tempting to spend too much money online.  As most of us know, these purchases do not always serve our long-term interests. If we are not careful, we unnecessarily duplicate products that stock our shelves and wardrobes, and slow down our future saving goals with expensive interest rate charges as a result.    

“"The pain of paying influences our decision to buy."”

The "pain of paying"  

So how to avoid spending too much? Studies on buying behaviour have emphasized that one way to do that is to raise the “pain of paying”. This uncomfortable feeling that comes with paying differs across payment methods and influences our decision to buy. Using a credit card is associated with a lower “pain of paying” – and therefore higher levels of spending2 – than paying with cash. An increased pain of paying is like a “moral tax”, as researcher Dan Ariely calls it3, because it makes us less inclined to make the purchase.

Similar to credit cards, online buying has been associated with a medium to low level of “pain” and stimulates spending. This raises the question how irresponsible online spending that causes debt could be reduced. In our research we study if highlighting the consequences of going ahead with the purchase increases the “pain of paying” and as a result helps people to control their online spending in a responsible way.

Ask the crowd

To test our hypotheses, we want to benefit from the "wisdom of the crowd". At the IBM Center for Advanced Studies in Amsterdam, we have a longstanding collaboration with the Vrije Universiteit in Amsterdam on the CrowdTruth4 project. We have found that crowdsourcing is a well-tested method that allows quick, anonymous, cheap, realistic and reliable testing of research scenarios. For these reasons, we are using the Crowdflower platform (  to gather our human data.  

As part of our research, we ask the crowd to complete fictitious purchases online for a desired product. By varying the conditions under which these purchases are presented, we want to understand which factors may drive a buyer to cancel their purchase. 

It works like this: The participant is shown an online payment screen with a warning message that highlights a negative effect of completing the purchase, such as increased debt or decreased savings. The warning messages vary in severity, concreteness, emotional impact and the endurance of the consequence (i.e., short or long term consequences). For example, people received messages such as: “Are you sure? You don’t have sufficient balance to complete this purchase right now” and “Please note – this purchase amount equals approximately 20 hours of work”. With the findings of this research we expect to be able to provide a clear picture of what type of warnings are most effective in increasing the "pain of paying" – thus avoiding irresponsible spending that could lead to financial problems.


As with all research projects, we started by defining our concepts and designing a questionnaire to test our warnings. In the meantime, we have already received over 1,000 responses in Crowdflower and are currently analysing them. The initial results are promising: they  show that warnings could indeed be a successful tool to reduce online spending. As a consequence, our insights seem highly relevant and applicable to TFI´s mission to help people better manage their finances. We look forward to completing the analysis in the final weeks of our project and also to hearing more about the other Research Challenge initiatives!  

Are you curious to know more about our specific outcomes? Keep a close watch to the TFI website and read our report which will be published in March 2018.


  • 1.
  • 2. Soman, D. (2003). The effect of payment transparancy on consumption: Quasi-experiments from the field. Marketing Letters, 14, 173-183.
  • 3. Ariely, D. (2002). Predictably irrational: The hidden forces that shape our decisions. New York, NY: HarperCollins Publishers.
  • 4. Chatterjee, P., & Rose, R.L. (2012). Do payment mechanisms change the way consumers perceive products? Journal of Consumer Research, 38, 1129-1139.
  • 5.