How to select the best ideas for development
Author of the article: Dmitry Kurdyumov
Participated in Agile transformations in the largest companies in Russia (Alfa Bank, MTS, X5 retail group), with international experience in a startup abroad
In any Agile product company, there are many challenges associated with managing the flow of ideas and tasks. Without a clear system for selecting and prioritizing ideas, teams can face duplication of work, overload, and chaos in the development process. This leads to new ideas being thrown at the team spontaneously, without sufficient elaboration and coordination between stakeholders, which ultimately affects the fact that a lot of work is being done that does not bring the desired effect.
Case of one company
In one of the companies I worked with, the problem was that many raw ideas were coming to teams without a quality Discovery process that would allow tasks to be analyzed in terms of value. There was also no single platform for discussing and selecting tasks, and any stakeholder could bring a task into work, even if it was not sufficiently developed or was not a priority. This led to the fact that the initiative was often “urgent”, bypassing the main processes, and moving already planned tasks forward. Stakeholders did not have a single tool for synchronizing their ideas and coordinating with product managers.
Step one: general meetings to select tasks
The first step towards solving the problem was to introduce regular meetings with all stakeholders and product managers, where ideas were discussed and those that were to be included in the work were selected. We implemented the principle of a pull system: as soon as two tasks were completed by the team, two new ideas could be included in the work. This allowed us to synchronize the actions of all participants in the process and improve control over priorities.
To prioritize tasks, we started using the methodology RICE (Reach, Impact, Confidence, Effort). This tool helped evaluate ideas more objectively, based on data rather than subjective opinions. However, this solution did not solve the problem of developing ideas before they were sent to the team. Meetings helped organize tasks, but the Discovery process remained weak, with decisions about task selection based on minimal data.
Step Two: Create a System Discovery Process
To overcome the challenge of insufficient idea development, we set about organizing the Discovery process, which became the next key step. We created a special platform for discussing ideas, where all stakeholders could submit their proposals for further development. In order to make the process systematic and conscious, each idea had to be prepared according to certain criteria:
What customer or user segment is this idea for?
What task or problem does the proposed idea solve for the client?
What metrics will be impacted by implementing this idea?
Supporting materials proving the viability of the idea.
A limit was set on the number of ideas each stakeholder could contribute—no more than one or two per meeting. This encouraged more thorough preparation for idea presentation and reduced the likelihood of minor or poorly developed suggestions being included.
At the meetings, each stakeholder had three minutes to defend their idea, explaining its value according to the specified criteria. If an idea was not defended or was not prepared properly, it was postponed or excluded from consideration entirely. This helped eliminate unprepared or low-priority ideas from the process.
Step Three: Conducting Discovery Sprints
After pitching and prioritizing ideas, the next stage was the Discovery Sprint. In each sprint, the team focused on exploring one or more prioritized ideas to determine their viability. The key element of each sprint was its targetwhich set the focus for the team. The sprints involved product managers, analysts, designers and researchers who tested hypotheses, conducted customer development interviews with users and assessed the idea's economics over the course of two weeks.
The main objective of the Discovery Sprint was not just to check the technical feasibility of implementing the idea, but also to assess how it meets the needs of users and whether it can affect business metrics. It was important to confirm or disprove the hypotheses behind the idea. We implemented a clear definition of ready For each idea, the criteria that had to be met in order to consider the idea explored.
At the end of each sprint a meeting was held results overviewwhere product managers and the team would draw conclusions about the relevance of ideas and decide whether to take them into work or reject them. Sometimes, an additional sprint was needed to deeply explore ideas.
Step Four: Process Visualization and Task Management
To better control and coordinate the process, we implemented visualization of work in the Task Tracker. We created a board where each product had its own column (swimlane), and each idea went through all stages: from pitching to the final assessment. Such visualization allowed us to see the entire process and control the flow of ideas in accordance with the performance of the teams.
An important aspect of using the board was the limitation on the number of ideas coming to the teams. This helped to align the flow of tasks with the development capabilities and avoid overload. Visualization also became a powerful communication tool between stakeholders and teams, allowing everyone to see the current state of the process.
Results and Conclusions
As a result of implementing the systemic Discovery process, we were able to significantly improve the quality of decisions made and reduce the number of tasks that were sent to the team without proper development. This not only increased the productivity of the teams, but also improved the interaction between stakeholders, making the process more transparent and predictable.
Systematization and visualization of the Discovery process turned out to be extremely effective. It allowed us not only to improve the work of teams, but also to provide more predictable results for the business. Regular pitching, control over the number of tasks and clear criteria for the readiness of the idea made the development process more manageable and effective.
If you liked the article, I invite you to my telegram channelwhere I write even more useful information about building effective processes.
You can gain more relevant development management skills within the framework of practical online courses from industry experts.
