The User Experience Paradox

00:04 Introduction
00:42 Lokad is still very much a 2D environment. When is that going to change?
02:11 Perhaps you should share your vision on how those user interfaces will look like in the future? Do you have an example?
04:01 How about companies? How can they trust these pieces of software if they are never going to interact with them?
05:46 When it comes to demand forecasting softwares, what can we expect from those user interfaces for the future?
07:58 If people aren’t included in these forecasts, how can we actually trust the results?
10:56 Could we not share those metrics with other companies?
13:46 In the future, can we actually have supply chains working on full auto-pilot?
16:18 Is there anything else to do, other than let people write a bit of code?


Supply Chain Management (SCM) systems feature complex user interfaces. Among them, demand forecasting subsystems are not only complex but complicated as well. Better user inferfaces are needed to tackle this complexity.

The future is already here when it comes to user interfaces, but that future is not evenly distributed. In this episode of LokadTV, we talk about the evolution of user experience and what supply chain user interfaces of the future will likely look like. We explain how this software can work without people in the forecasting loop and how companies can trust the results.

The vision of future user interfaces is often influenced by movies such as Minority Report, which shows Tom Cruise working with a futuristic looking 3D user interface. However, this Hollywood perspective is far cry from reality. We don’t want to crush any dreams, but we’re sad to say that 3D user interfaces won’t be making their way over any time soon. We discuss in more detail just why these user interfaces simply do not work in practice.

The best user interfaces are in fact ambient and invisible, and they are often found in the places where you’d least expect them - for example the anti-spam filter in your email inbox. This filter depends on powerful, unusual metrics that work on intelligently removing outliers - which is also what successful forecasting methods for supply chain should do. Instead of relying on mean averages, intelligently removing outliers requires a use of cross entropy and machine learning, which is still alien for many forecasting softwares.

To wrap things up, we see how better programming environments are in fact likely to become the user interfaces of the future. We also debate how supply chain software interfaces are arguably quite “dry” and how they can be rendered more dynamic.