As Christmas shopping ramps up, you may be getting a lot of emails offering you attractive discounts for a short period only. You may see flash sales or special deals that exhort you to “buy now” to avoid missing out.

These digital “time-limited” offers, as they are called, are actually an old sales tactic.

Those in the game of selling cars, for example, have long used the trick of alluding to that other very interested buyer who’s likely to return and snap up the bargain that’s before you. Telephone salespeople routinely offer deals that must be accepted during the call. Want time to think about it? Too bad.

Online time-limited sales work on the same basis, but with technology taking it to a whole new level. Now retailers can bombard you with offers that are highly customised and super-short – a deal, perhaps, for something you might have been searching online for, and now available at a discount only until midnight.

But for these tactics to work, our research suggests, requires finding a Goldilocks zone between being too pushy and not all. Time needs to be limited to deter you from searching elsewhere for a better deal. But paradoxically you also need enough time to convince yourself that buying is the best decision.

Experimenting with time limits

To find out what makes time-limited offers effective, I and my colleagues Robert Sugden and Mengjie Wang from the University of East Anglia ran experiments to see what leads people to accept or reject such offers.

What we found is that these offers leverage risk-aversion. That is, the more you dislike risk, the more likely it is you will take the bait and buy now.

In our experiments, using university students, we asked participants to complete 30 “price search” tasks. These tasks involved giving participants a “budget” and asking them to buy a product from six different price offers, shown to them sequentially with a few seconds between each. Any unspent money they got to keep.

In half of the tasks they could consider all six offers before making their choice. In the other half, one of the first three offers would be time-limited, lapsing after either four or 12 seconds, which they could only accept before the next offer appeared.

We also varied, when participants accepted a time-limited offer, between showing them no more offers or showing all remaining offers immediately. This was to test if greater feedback (increasing the possibility of regret) reduced the probability of a time-limited offer being chosen.

Participants then did 15 related risk-taking tasks based on their choices in the tasks with time-limited options. This helped us determine what was going on with their choices.

A time paradox

Overall our results point to choosing time-limited options being linked to risk aversion. People generally prefer to secure a certain cake now over the uncertain possibility of a better cake in the future. We really do believe the old proverb that a bird in the hand is worth two in the bush.

But there was a catch – and a big one. Somewhat paradoxically, people also need to think things through to jump on the time-limited offer. Time-limited offers were accepted more when participants had 12 seconds to decide rather than four seconds.

This indicates people need enough time to reflect on the task to decide they are better off going for the “safe” deal.

As we warn in our paper, one should be wary about extrapolating too directly from laboratory behaviour to real markets, but our results suggest time-limited offers do not rely on limits to the consumers’ ability to make a rational decision. When they work it is because they are mechanisms of search deterrence – restricting the consumers’ opportunities to compare available offers – amplified by risk aversion.

So businesses may be shooting themselves in the foot when they create offers that are too short, too pushy. If you’re like most people, you need time to reflect on the risk of not buying. If the offer is too fast and furious, you’re likely to just be turned off.

The research referred to in this blog was supported by the Economic and Social Research Council of the UK (awards no. ES/K002201/1 and ES/P008976/1); by the Centre for Behavioural and Experimental Social Science (CBESS) at the University of East Anglia (UEA); and by funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme, grant agreement no. 670103.