Experimentation Platforms Speed Innovation for Development Teams

According to a new report from Forrester Research Inc., experimentation platforms are now considered the new normal for application development and delivery (AD&D) teams, allowing faster, more focused innovation. Software teams from top-performing companies rely on experimentation platforms to push the right code faster, with more control, and greater insight.

Since the Agile Manifesto was first published nearly 20 years ago, many organizations have adopted the practice to increase velocity and reduce delivery risk. Today digital experience is becoming the differentiator for every business, and software development teams are more critical than ever to the success of these experiences. It’s no longer enough to just deliver software on time with few bugs. As Forrester analysts Chris Condo and Bill Seguin outline in this report, this shift means that “the pressure increases on AD&D teams to produce winning results faster — with fewer mistakes.”

Leading application development teams are now going beyond Agile and DevOps, and introducing experimentation into their product development process to “push the right code faster, with more control, and greater insight.”

The January 2019 Forrester report, AD&D Pros: Stop Guessing and Start Experimenting (download the complimentary report), shares three main takeaways, which we’ve paraphrased below:

• Without experimentation, companies can’t innovate
  Unless companies are able to experiment live with real users, there is no innovation, just guesswork.
• Experimenting in Production Speeds Innovation
  Experimentation platforms provide a convenient and safe way for AD&D teams to test hypotheses in production, with real users and real data, replacing internal opinions and disputes with actual behavioral data.
• Experimenting In Production is the New Normal
  “Continuous discovery”—getting new ideas in front of users quickly—is becoming common practice among a select few innovators, and experimentation platforms let teams test real-world scenarios with precise control.

Top Performers Utilize A/B Testing, Feature Flagging, and Direct User Testing

Experimentation platforms can determine winning tests in a matter of hours or days, allowing teams to quickly launch and test new ideas. The three main experimentation capabilities that every organization needs in their product development process are feature flagging, for controlling the release of new features; A/B testing, for measuring the impact of changes to the user experience; and direct user testing (DUT), for collecting qualitative feedback from real users.

Each of these capabilities offers different benefits at different points in the product lifecycle. Feature flags are now being widely implemented by organizations that are using continuous delivery to move away from traditional release cycles. They enable developers to decouple feature release from feature deployment, enabling developers to focus on deploying code behind a flag, and giving business teams the ability to turn on features in production. Feature flags also enable single-trunk development, and improved risk mitigation through progressive release strategies.

A/B testing, while traditionally a strategy used by marketers and front-end design teams, is now being widely adopted by development teams as well, who are able to prove the impact of their work on business metrics through server-side A/B testing with statistically rigorous measurement. Server-side testing is particularly useful for development teams who release changes to the core business logic of an application, such as an algorithm, price, or new feature. The method enables these teams to quickly iterate on underperforming features and concepts to deliver winning digital experiences.

Only Optimizely Offers a Powerful Combination of Feature Flagging & A/B Testing

While each type of tool provides value by itself, top-performing companies use all three to maximize the impact of their efforts. By offering both feature flagging and A/B testing in one platform, Optimizely streamlines the flag management and experimentation processes, and allows teams to back every decision with rigorous analytics. By combining the powerful functionality of multiple categories of experimentation platforms, Optimizely empowers forward-thinking companies to do their best work.