Sustained is as much about transparency as it is about education.
While we regularly talk about discoveries we have found whilst creating the app, we are yet to discuss the inner workings of Sustained.
This post gives an introduction to some of the values we hold ourselves to, and what to expect from us, as we continue to grow.
One of the main pillars on which Sustained is built is trust.
Trust that is based on objective reason, but which builds connection with people and their subjective beliefs.
In order to build and maintain such trust we must be transparent: we are happy to fully disclose the decision process behind any metric or aggregated value we provide. We also communicate openly and clearly on changes in our methods and approaches and the reasons behind them.
Sustained must also be impartial. We do not decide for our customers or shape their preferences, we simply inform their decisions and help them navigate a complex space full of tradeoffs.
Trust requires continually demonstrated performance, transparency of policies and decision-making, auditability and tracing of data sources.
It also requires a formulated, reasonable and actionable ethical stance.
Sustained treasures scientific method and results of scientific discovery, and we back up our interpretation of the actions companies take and effects they have on the environment, with established, reasonably verified scientific evidence.
We also recognize that science never states anything as a dogma, and continuously, critically tests even the most established theories.
For example: it’s been common knowledge for decades that cholesterol causes cardio-vascular diseases and numerous measures have been taken to reduce LDL-C (aka ‘bad cholesterol’) in dietary products. If Sustained existed at a time, we could easily have had ‘Increases probability of cardio-vascular diseases’ as an info badge for LDL-C-containing products.
In 2018 a meta-research came out challenging this common knowledge. Unless it’s disproved, the right thing to do in such a hypothetical situation would be to remove ‘Increases probability of cardio-vascular diseases’ as an info badge for cholesterol as an ingredient.
We must be prepared for changes in base theories, claims and other data inputs, and technologies that we select to implement the model allow us to recompute effects holistically if such events occur.
While we shouldn’t ignore popular demand, we must put diligence and scrutiny before hype and reason, only taking public action (i.e. issuing a new Sustained info badge or a product score component) when we’ve reached a comparable degree of confidence to our longer evaluated domains.
In particular, we actively filter out unfair marketing practices that can be both trivially true and misleading at the same time. For example, let’s consider the ‘gluten-free’ movement that is fair to assume to be popular enough. It would be a non-starter to even consider tracking that effect for anything other than products having grain in their (sub)ingredients. So, in simple terms we must say firm ‘not on Sustained.com!’ to gluten-free salt, sugar, meat etc.
Another example would be the recent popularity of ‘paraben-free’ cosmetic products for the rumored connection of parabens used as preservatives with cancer. It has been shown that no such connection exists, moreover it has been comprehensively shown that paraben alternatives are at least not safer, but in many cases - significantly more toxic. We may still track ‘paraben-free’ as a product property, but would combine it with educational efforts to present a balanced, evidence-backed view of preservatives in cosmetics.
As we continue to grow and examine the real-life ethical dilemmas about the data we collect, we will undoubtedly find mistakes that we made along the way.
However, we are dedicated to providing you with the most accurate and up-to-date information, even and especially when that means we need to change.