When I first started talking about predictive and proactive analytics about a year ago, I talked about the promise of using artificial intelligence and machine learning (AI/ML) to predict when the blue screen of death might be close, telling the user what to do about it, and being able to prevent the problem from occurring in the first place.
As an industry, we’re moving predictive and proactive analytics from something that was a science project in the lab to the mainstream, from being something software developers care about to something that responds to users’ needs and improving the user experience (UX).
That’s obviously a good idea. What’s changed over the past 12 months or so, because of Covid-19, is the broader context. Proactive problem resolution for remote workforces is a “must”, not a “nice to have”.
We now all work in hybrid workspaces. The IT support desk, or even a friend or colleague in the next cubicle, is no longer there. If you’re offline, you’re offline.
The ability to resolve problems even before they happen has become more important and more critical than ever. Being able to predict failures and be proactive about resolving them ahead of failure is what makes the remote workforce possible. Without predictive and proactive analytics solutions, if your device goes down, you’re sitting at home, you’re no longer productive, and you have no means of fixing your problem.
The absence of a device failure is not actually enough. The reasons why the failure was likely, and the data surrounding any particular user’s situation, are as important. The whole experience should be part of a much-larger software ecosystem built around the user.
I think that’s what has brought predictive and proactive analytics into the mainstream: Users now do care and understand that advanced software can systematically reduce these performance risks and constraints. Anyone who has worked away from their office in the last 12 months (and that’s probably most of us) who has experienced a frozen screen, or performance degradation, will understand this only too well.
Going through this process of being at the cutting edge of innovation, defining over time what the customer problems might be, and how to resolve those problems, we suddenly went from backroom testing and prototypes to a commercial solution.
As a result, it’s no longer a discussion with customers about a solution that might work. It’s now a discussion about how proactive and predictive analytics can help the user, and how enterprises can provide better services inside their organizations, to employees and to their customers in turn.
It’s an incredible evolution from a hackathon-type project to a product that is being taken to the market, and being adopted by users.
This evolution happened from the customers’ mindsets. It’s about the relevance of the problems that customers want solving.
Users constantly come to us seeking more reliability, better compatibility, fewer issues, fewer crashes, less complexity in applications, more-reliable firmware. This is not a new problem. It’s like a race — the faster you become, the faster you have to be.
Customers define the problem well, and it’s not up to us technology solutions providers to second-guess problems. It’s our job to fix the problems. That’s changing as well. We are moving away from the model of analyzing the crash, to stopping the crash before it even happens.
What’s exciting is working out how to stop the problems before they occur, and to put those solutions back into software so that the problem becomes fixed forever across every device.
New partnerships for a new world
More than anything, we expanded our circle of partners. We realized that more data, more knowledge, and more technology provide better results for customers.
We work very closely with partners to get better predictive and proactive models for data, analytics, faster processing, and to look further into the future.
When done properly, I think this becomes an incredibly powerful exercise.
In a work-from-home, study-from-home, play-from-home world, predictive and proactive analytics is crucial.
The last 12 months have accelerated the industry-wide effort to make devices more resilient and operate better without the intervention of some kind of support, whether it’s from your employer, your school or your parent (or child).
I think it’s changed the tech industry for the better. The new environment changed the industry over a very short period of time, in the way tech industry leaders work together to improve the user experience and turn this new normal to the customers’ advantage.
It’s worth making clear here that we take the responsibilities around data very seriously. In creating predictive and proactive analytics solutions for customers, we are very mindful of the different data types we’re working with, specifically using only and exclusively anonymous device telemetry based on user opt-in. And we’re working here with users to make their devices perform and operate better, with their permission.
We’re maximizing the value that’s already been acquired and paid for, that’s already embedded in the device and software used by the users.
All of this is unifying the industry around the user experience, which I think is a very cool thing. Making things as good as technically possible makes me proud to be part of this industry.
Always on, always connected, always usable, always performing
What next? I think what we’re trying to do is actually to make all of this invisible. Right now, it’s still a technology that’s a process to resolve problems.
“Smarter PC” does not just mean faster performance. It means devices reacting in a natural and expected way. It’s about not having to go into settings, about remembering where I am or what work mode I’m in, about whether a device is on my lap or on my desk, and whether its cooling system is coping.
I think all this needs to become invisible to the end-user.
Users shouldn’t be bothered by how it works, they should just know that it does work.
That would be a great success, built around partners in the industry improving the customer experience to the nth level.
That means that when we ask a user when they last experienced the blue screen of death, they don’t know what it is because they’ve never seen it.