Three Secrets for Delivering Reliable Service in an Unreliable World

"Reliability is particularly challenging and important as viewers stream live events like sports and news. The impact of any service interruption or degradation here is magnified." -Greg Leja, Managing Director leading Accenture Digital Video in North America

If content is King, reliability is Prime Minister: less glamorous, but the one holding the power. Recent Accenture research indicates that reliability impacts the entertainment economy profoundly. Customers say they rate a reliable service as the most important factor in determining the trust they place in OTT video providers.

Outages, failed streams, app crashes, random error codes – anything that stops the viewing experience from working well – create almost-instant customer frustration while eroding the value of the service’s brand. Negative app store reviews for OTT premium video services show errors as the leading prompt for users’ negative comments, accounting for 46% of bad ratings. That’s more than three times the next most cited category, value (price), at 14 percent. Only seven percent complained about a lack of new features.

Reliability is particularly challenging and important as viewers stream live events like sports and news. The impact of any service interruption or degradation here is magnified. Break the live action flow, damage the experience and anger the customer. And, sadly, you’re unlikely to get any sympathy from viewers about how running your service over unmanaged networks takes control out of your hands. Numerous factors challenge your ability to consistently catch and handle interruptions in the viewer experience, but paying customers don’t care. They just expect everything to work.

So how do you achieve the reliability you need to win, in a market that demands constant evolution, while serving customers who will not forgive a sub-optimal experience? Focus in three areas is separating the winners from the rest: value-focused analytics, architecture and operating models.


Analytics are generally valued, but there’s a lot of confusion about where to focus invest and the value of such investments. Without clarity, companies see only limited value from analytics.

Leading companies’ products are informed by data, and their organizations are accountable … and rewarded for … meeting metrics that drive product value and impact. To ensure a reliable service, business and technology team KPIs must align, to focus efforts on making a material difference to your business.

To find that value, start by respecting the data. Make its collection across the product lifecycle a priority. Treat data issues as critical production matters and use commodity cloud technologies to leapfrog legacy issues and minimize cost. Let data be shared and accessible across product-engineering-operational silos. Use it to understand what individual customers value, to assess platform stability, and to act accordingly.


Reliability of service means that your video product is constantly available, videos start without delay, channels change quickly, customers are swiftly authenticated, and transactions are processed smoothly.

As an industry we are gradually transitioning to a cloud native architecture, one that has modularity and enables distributed and rapid feature delivery and enhancement. However, this shift is challenging accepted architecture patterns, making broadcast quality resilience and scale, a far from a foregone conclusion.

Microservices are a way to keep up with demand and deliver to expectations. With their fine-grained monitoring and targeted control, throttling up and down services is possible, without a cascading impact across the application. However, microservices are not a cure-all. A poorly applied microservice architecture will create snowballing complexity and impede scaling performance and costs.

Architecting for the best OTT video experience represents a fundamental rethink of

traditional ways to do business. Think lightweight, resilient, decoupled, API-driven, with an open architecture. This approach best enables agility for the new supply and demand routine, supporting personalized services and experiences across devices, applications and unmanaged networks.


Leading companies are creating a culture of innovation and continuous improvement that eliminates silos from product, development, marketing, and operations. To operate with greater agility and deliver at speed, you’ll need to adopt the same approach that characterizes the platform- and data-centric models deployed by the digital natives.

That starts with breaking down silos. I mentioned it above when dealing with data, but it’s critical for every aspect of your organisation. Traditional, hierarchical models will hold back performance. Build a new one with a governance model that integrates product, delivery and operations teams, uses data to create a healthy tension between product owners and engineering, puts the right people into the right roles and defines how everyone will work together.

You’ll need to grow skills and talent. Recruit team members who are future-focused, data-oriented and display strong decision-making. Once you’ve found them, keep them … and ensure they deliver what you need … with aligned incentives and accountability. Constructive feedback, incentive models, and appraisals are all essential to reinforce your new model.

It is, perhaps, unfair that the reliability customers take for granted is one of the most challenging things for a modern video provider to achieve. But it’s the reality of an OTT world. Those who turn data insights into action, build architectures to deliver those actions, then support it all with the right operating model are the companies most likely to win.

Greg Leja is a Managing Director leading Accenture Digital Video in North America. In this role Greg is focused on helping clients deploy and manage next generation digital services, driving video transformation for some of the world’s leading brands.