What is the impact of your client reviews on e-commerce platforms, like Amazon or eBay, on potential clients? What does it benefit you if you were able to carry your reputation data along to another platform? The following are insights from a professor in the field of ‘Trust in Digital Services’, Timm Teubner, research pioneer on the portability of reputation data.
Professor Timm Teubner is fascinated by questions like, ‘Do combined reviews from different platforms improve a client’s trust? And what influence does ‘portable’ reputation data have on a supplier’s online profile?’ He researches how suppliers on online marketplaces could carry their data along to other platforms.
Before the measures against the corona pandemic made all international visits impossible, I met up with Teubner in Berlin. This podcast is the recording of our interview regarding online reputation and the added value of data portability. Here, I’ll share the key points of our conversation.
Little research on portability
Very little is known about the impact of portability of reputation data. The limited research conducted so far has been done by Teubner, professor in ‘Trust in Digital Services’, in collaboration with the faculty of Economics and Management at the Technical University of Berlin.
Teubner researches how suppliers on online marketplaces could carry their data along to other platforms in order to create a trustworthy profile on several marketplaces. Currently, most reputation data (e.g. reviews, number of successful sales, delivery time) is locked in to the platform where the data is ‘earned’.
Platform experts foresee benefits for all stakeholders (supply, demand, and platform) if this data could flow freely among platforms. That is, if you could take your data along, to make use of it in other marketplaces.
Will there truly be benefits? Teubner tries to unveil the future, and tests whether reputation portability really has any economic impact. When a supplier carries their reputation over to a new platform context, how trustworthy will the perception of it be? To what extent will potential users, guests, or buyers consider reviews from a different source to be valid?
These questions are central to Teubner’s research paper “Unlocking Online Reputation — On the Effectiveness of Cross-Platform Signaling in the Sharing Economy”. In his second paper, “Bring your own stars – The economics of reputation portability”, Teubner and his co-researchers look at the interaction between existent and imported reviews for suppliers on online marketplaces.
The impact of imported scores
It appears from both research papers that imported reputation scores do have a significant effect on trustworthiness as perceived by the client. Simply stated, when a supplier shows reviews earned on another marketplace, it could improve their reputation, which in turn leads to more transactions.
But the imported scores still do not reach the level of trust given to scores obtained on the platform where they are uploaded. Clients perceive so-called ‘on site’-reputation as more trustworthy. However, a profile with reputation data stemming from a different platform is seen as significantly more credible than a profile without any reviews at all.
In short, the results of these first research papers indicate that the portability of reputation data contributes to the trust buyers have in a supplier. This in turn helps suppliers get started faster on new platforms.
Why focus on e-commerce?
It strikes me that Teubner and his colleagues are mainly researching e-commerce platforms. Similarly, start-ups developing reputation data related services are generally focusing on online sales platforms. There are a couple of reasons underlying their choice for e-commerce.
First of all, e-commerce is a big market and is expected to grow even larger in the years to come. We order more and more online, especially during this time of the coronavirus. Even more suppliers are emerging with similar products. And as these sellers cannot stand out with their supply content, they will differentiate themselves through price and … reputation.
Secondly, e-commerce exists in virtually every industry. Still, the service of suppliers is particularly universal. This makes it easy to link a rating of book sales to, for example, a review of bicycle tire sales.
This aspect is much harder within the gig economy. One being an excellent meal deliverer on UberEats, doesn’t imply much about their qualities as a data scientist on a platform like Upwork. The cross-platform reputation effects are much more compatible among e-commerce platforms than other types of platforms.
Finally, reputation scores within e-commerce are relatively easy to measure. The way sellers are being reviewed is practically the same on all marketplaces. The supplier usually has a lot of influence on all these variables: the product, the communication, aftersales and transport.
When service related platforms are evaluated, suppliers are found to have far less control. For example, the food deliverer has neither influence on the quality of the food he delivers, nor the time the restaurant finishes the meal, nor the current traffic conditions. Nonetheless, all of these still influence the review the courier receives.
The right to be forgiven
A few bad reviews can have a severe impact on one’s score. Teubner states, “It strikes me that I come across users with very, very positive reviews in my research. It is not true that mediocre or bad suppliers don’t exist, but suppliers with a bad review have a higher probability of dropping out of a dynamic market over time. They abandon their account, as they can’t get rid of the bad reviews.” In one of his papers, he refers to these skewed ratings as ‘survivorship bias.’ “Suppliers with bad reviews are less visible, receive fewer customers, and disappear from the platform after a while. Though, they often simply start again with a different name or email address”, he says.
Profiles with bad reviews disappear from the platform to start over with a clean slate somewhere else. This may be harder on gig economy platforms, given that a profile is linked to someone’s physical identity. It would be a great idea for these kinds of platforms to implement ‘the right to be forgiven.’ Teubner agrees on this matter, “you could be allowed to delete one review every 3 or 6 months,” he suggests. “Everyone makes mistakes from time to time. There should be room for that as well on online platforms.” On the platform Meeting Review, scores are discounted over a period of five years, which is another way to provide space for learning and experimenting. On gig platform Upwork, scores can only influence a worker’s average for up to two years. Additionally, top rated Upwork freelancers receive a perk to remove one review that negatively influences their appearance on the platform once every three months.
What we need to learn from further research
Every research project raises new questions, and so does Teubner’s. To gain deeper insight into the added value of the portability of data for platform workers, more specific research will be needed. This interview has shed some important initial light on how to discover under which conditions portability of the reputation scores of platform workers could contribute to the empowerment of those workers.
This post is also available in: Dutch