Skip to main content
We’ve looked at every prioritisation framework in existence: RICE, ICE, PIE, PXL, HIPE, The Value Effort Matrix, Weighted Decision Matrix, Weighted Shortest Job First (WSJF) and more. These don’t work for our customers for the following reasons:
  1. Goals can change throughout the year and customers want to reprioritise campaign ideas based on their current goal
  2. Factors like Impact, Potential and Confidence are all highly subjective metrics and (more often than not) complete unknowns. Teams want to remove this subjectivity.
  3. Teams don’t want to spend manual effort on prioritisation, particuarly in AI world. The biggest challenge is not moving fast enough. Every minute spent on manual prioritisation is time not spent on doing real impactful work.
In our experiences team are, in general, poor at guessing user behaviour and the likely impact of changes. There is no reliable way to generate an estimate of a test’s impact or confidence, short of actually running the test itself. This is the reason AB testing exists in the first place.

​
Real-time scoring

Our campaign hypothesis builder ensures all new campaign ideas follow a best practise format and structure which means every campaign idea in Growth Method has sufficient detail and context to enable automated scoring and work prioritisation. Upon creating a new campaign idea our AI-powered scoring algorithm runs automatically in the background. Our scoring algorithm then generates a score (from 1 to 10) as well as a score explanation. This typically takes around 3 seconds. Scoring currently takes into account 2 factors - ease of implementation, and relevance to the current team goal and outputs the following information.
FactorDescriptionRange
Ease ScoreEase with which the idea can be implemented where the easier and simpler the idea, the higher the score.1-5
Relevance ScoreThe relevance of the idea to the current/active goal. The higher the relevance, the higher the score.1-5
Total ScoreEase and relevance scores combined into a single overall prioritisation score1-10
Additionally the AI generates an explanation for the score that has been given which is visible by hovering over the numeric score in the app. An example explanation is shown below:
This campaign is relatively complex (3/5) to set up as it involves crafting messages, identifying target audiences, and tracking various metrics on LinkedIn. The relevance is high (4/5) as outbound sales strategies can significantly boost lead generation and align well with the current goal of increasing leads.

​
Future improvements

Our scoring algorithm was built to enable continuous improvement over time based on real-world data and customer feedback. Additional weightings that may be added to the scoring algorithm over time are listed below.
WeightExplanation
VersionsRelated to a previous campaign e.g. a v2 or later campaign indicating previous success
CommentsHas the campaign received comments from one or more members of the team?
LikesHas the campaign received likes from one or more members of the team?
TimingHow old is this idea? When was it created and last updated?
IndividualsHistorical win rate of the individual
CategoriesHistorical win rate of the category and/or channel

​
Resources

Additional resources on work prioritisation.