If you as a product owner want to make a true positive impact on customer satisfaction or your NPS score, you want to be obsessed with real-time data-driven insights. Once upon a time, you had to wait on a weekly report that was two weeks behind the curve by the time you received it (if you were lucky), and would never contain technical data, such as response times. In this series of blog posts, we examine how these kinds of outdated IT processes are becoming a thing of the past.
Release, test and optimize experiments at will
In our last two posts, we looked at how testing in production on real customers and being able to control that process by adding fine-grained customer segmentation makes releasing a KPI-driven business decision. In this post, we’ll look at how to add real-time data to that process and make experimentation a fast-to-set-up, smart, hands-off process.
With long reporting and experimentation cycles, how can you improve the product?
It’s your job to drive the product forward in terms of customer experience and revenue, but how can you do that without having fast access to relevant data? In many cases, your data sits parked somewhere waiting for an analyst to dig it up, so that by the time it comes to you, it’s outdated. On top of that, if you want to run experiments to improve the product, that can mean logging tickets and waiting till the end of a bi-weekly sprint to see the results.
Best practice: smart release strategies integrated with tech and business analytics
Luckily, new technologies are making it easy to integrate with analytics tools like Google Analytics, Omniture, Data Dog, New Relic, and more, so you’re able to make decisions using real-time KPIs. Or better: that happens in a hands-off automated process. That happens by integrating data sources with the business requirements for your release through release policies. Here’s how it works:
- Release policies are rules to govern a release and allow you to specify what business requirements you would like a new release to achieve
- The release policy is a set of instructions that tells a release orchestration technology how to segment customer traffic to meet your specified business requirements (think: release only to 5% of logged-in Android users in Germany)
- Because it “knows” your business requirements, the release policy reads metrics from monitoring and analytics tools, so it knows how your release are performing across the board
- That process easily ties IT metrics to business outcomes.
- Experimentation now becomes smart and hands-off: you no longer have to wait for the outcome of an experiment and make a change request to see it implemented, but a release solution integrated with the APIs in tools such as Optimizely and Adobe Test and Target can do that for you.
- What’s more, you can take the results of a set of AB tests and create a higher-level release strategy out of a number of policies to automatically and continuously optimize revenue. In this way, you can automate the roll out of winning experiments to the entire customer base!
Automatic releasing and experiments sound great! But…what if something goes wrong?
Ah if there’s one thing that everyone in IT has in the back of their minds, it’s that there will always be broken code. Since enabling Friday releases is our mission, we will have an answer to that too in the next instalment of our series (hint: automated rollback). Or if you would like to read the whole story now, download our whitepaper 📚 ‘Free from the Department of ‘No’: Release Features, Test and Optimize Revenue Without IT.
Vamp is specifically designed to give product owners direct control over releasing, testing and optimizing at will
If you would like to see how Vamp Cloud-Native Release Orchestration gives you real-time control over releasing, on a Friday, Sunday, or any time of day or night, without having to rely on IT, explore Vamp.io or book a guided product tour.