Implementation Issues with Data Trust

The last blog post presented the WRAP (Waste & Resources Action Programme) project and their data trust pilot. Unfortunately, this food waste data trust never saw the light of day, which brought to light the challenges and difficulties of operationalizing certain data trusts. In the context of this specific project, it is important to note that contextual elements prevented the trust from going ahead (e.g. the Brexit and the inability to conduct enough interviews to build the trust, among others). However, there are other difficulties that can be applied to the implementation of data trusts in general.

The first issue (and one of the most important in my eyes) that can be encountered while carrying out a data trust is the lack of common objective or common view between all stakeholders: all parties must agree on the purpose and problem that the trust is meant to address. Everyone must have the same view of the problem…or agree on the fact that there is even a problem to solve. Each party must therefore work with the same vision: a lack in this regard can have a devastating effect on the viability of the data trust.

Another difficulty can be a lack of strong incentives for data trusts to be a suitable approach: because there are different stakeholders involved in the operationalization of data trusts, they sometimes have divergent incentives, which can be troublesome. In the example of the WRAP project, manufacturers and retailers had fears about the commercial confidentiality or the reputational harm in sharing the data, which discouraged them from providing complete and comprehensive data.

Additional complications can be related to the ownership of the data contained in the trust: several schools of thought are of the opinion that the subject-matter of a trust must be definable as a property right for the trust to be feasible and usable.

Finally, another challenge is the difficulty to ascertain what data relates to which identifiable individual: data can relate to more than one person at a certain time, but can also vary over time, which can be tricky.

There are of course other implementation issues with data trusts, like data governance, but these are the most problematic in my opinion.

Ce contenu a été mis à jour le 1 avril 2021 à 21 h 38 min.