How can we monitor whether our services *actually work* for our customers? At scale.
How can we better understand how our customers deal with the coordinations and breakdowns?
Netflix tackles this with chaos engineering medium.com/netflix-techbl…
@jeffsussna shows a path, with Promise-Driven Design
medium.com/@jeffsussna/vi…
vimeo.com/326447998
There’s no way for them to log what the knock-on effects were. And even is they could, the product team probably wouldn’t be able to fix it.
How could we bring Quality Circles into services that straddle the physical and the digital, where value is co-created through use, where the context of use is unpredictable, errors vary wildly and *the customers are our colleagues*.
I suspect the service quality problem is deeply intertwined with how we create services.