promise-keepingPhoto by Jonathan Velasquez on, under a CC0 licence,

It is a busy day at work. Suddenly a colleague jumps into your office asking you to quickly review some slides for an upcoming presentation. “Could you please do me this favour by tomorrow?” she asks. Well, this task is not really related to your work, but you promise her to do the review. Now imagine the identical situation, but this time your colleague asks for the favour via e-mail. How would this change in communication channel affect your response, and, in case you commit yourself, your propensity to actually keep your promise?

Given the huge prevalence of modern communication channels (e.g. phone, e-mail, instant messaging) in globally and digitalised corporations, our paper aims to answer this question. More precisely, we investigate how different channels of communication shape how people behave when making and keeping promises.

In a laboratory experiment, college students were asked for their willingness to promise to participate in a short online survey for a scientific purpose in the next 24 hours. In case a participant agreed to take part in the survey, he/she received the information needed to access the online survey platform in the next 24 hours.

Thus, two variables of interest are at hand. First: whether or not the student promises to take part in the online survey. Second: whether or not the promise is actually kept. No monetary incentives are involved to motivate participation as participants were already paid for an unrelated experiment task.

To differentiate the effects of different communication channels, we implemented five different conditions to ask for the promise: 1) face to face, 2) phone (via Skype call), 3) chat (via Skype chat), office email (within laboratory) and 5) home email (outside laboratory). Each student only participated in one of the treatments (between-subject design). In each condition, the same research assistant approached the students verbally or in written form with the identical protocol requesting the promise.

From a more theoretical perspective, the channels of communication we employed have different features that may affect promise making and keeping.

Technically, face-to-face and phone communication are synchronous, whereas email is recordable. Chat-communication features both, synchronicity and recordability. So called telltale cues, e.g., blushing or tone of voice, are only prevalent in face-to-face- and phone communication.

Previous research also categorised the employed channels by different degrees of social closeness, with face to face as the closest and home email as the most distant form. Three key aspects of social closeness are prominently discussed, which may explain more pro-social behaviour with increasing social closeness: higher degree of identification, less anonymity, and stronger social norm activation.

Table 1. Characteristics of the different communication channels


Notes: With respect to the aspects of social distance, 1 stands for the lowest degree and 5 for the highest degree of a respective aspect. Check marks represent the presence of a technical feature.

Turning to the results, we find that 88 per cent of the participants make the promise in the face-to-face condition. Almost identical shares (85 per cent) are found in the phone and 78 per cent under the chat condition. However, these proportions drop with statistical significance to 67 and 53 per cent under asynchronous office email and home email, respectively.

But how many of the promises made are actually kept? Surprisingly, promise keeping rates are pretty low. They range from 36 per cent in the chat and home email conditions to 51 per cent in the phone condition. Between the communication channels, no statistically significant difference can be found with regards to promise keeping. Recordability concerns do also not contribute to the achievement of higher promise-keeping rates.

Table 2. Descriptive statistics


The study shows how “oral” versus “written” concerns do not lie at the core of the question. Instead, the synchronicity versus asynchronicity property of the communication channel seems to be the key element in promise making. When people are approached through a synchronous and interactive communication channel (face to face, phone, chat) around 85 per cent make a promise, while only about 60 per cent do so when communication takes place asynchronously and computer-mediated (office email and home email).

Synchronicity may thereby lead to a rather automatic than reflective handling of the request to make a promise. The automatic response may be more pro-social in contrast to a reflective assessment of the request, which might lead to an anticipation of the resulting effort and a possible refusal of the request.

Thus, also in terms of promise-making and keeping, the widespread implementation of chats rooms such as Slack and Chatter, or the transition to email communication seem to be justified. It reduces transaction costs and does not influence outcomes in the sense of promise keeping. But still, when you are asked for a favour personally, it just feels better, right?



julian_conradsJulian Conrads is currently consultant at CONCERN GmbH, which is specialized in the development of CSR and sustainability strategies. He received his Ph.D. in business administration from the University of Cologne to which he is still affiliated. His research and consulting interests are behavioural business ethics, compliance culture, and sustainability management. (Email:

tommaso-reggianiTommaso Reggiani  is currently assistant professor of economics at LUMSA University, Rome – Palermo campus, and research affiliate at IZA-Bonn. He received his Ph.D. in Economics from the University of Bologna and he held research positions  at  the University of Cologne and Toulouse School of Economics. His research interests include behavioural labour economics, economics of education, experimental economics as well as microeconometrics. (Email: