AI can play a proactive role in helping teams analyze their progress and formulate actionable next steps.
‍

AI Prompt 1: Generating Actionable Steps for Accelerating Progress

“Given our current progress update for Key Result '[Increase monthly active users (MAU) from 5,000 to 15,000]' which is currently at [7,000] out of [15,000], what are 3 actionable steps we could take in the next [two weeks] to accelerate progress? Consider typical growth strategies for [SaaS products].”

How to use it: Provide the specific Key Result, its current metric, the target, and the timeframe for actions. Add context about your industry or product type for more relevant suggestions.

Expected AI Output (Examples)

"Launch a targeted in-app re-engagement campaign for dormant users, offering a new feature preview."

"Initiate a viral loop strategy by adding a prominent 'invite a friend' feature with a clear incentive for both parties."

"Analyze user onboarding drop-off points to identify and fix critical friction areas, focusing on the first 24 hours post-signup."

"Run A/B tests on our acquisition landing pages to optimize conversion rates from marketing spend."

AI Prompt 2: Identifying Off-Track KRs and Suggesting Interventions

“Analyze the following weekly progress updates for our OKRs. Identify any Key Results that are significantly off track and suggest potential reasons or interventions. Assume an ideal weekly progress rate of ~8% for quarterly KRs.
‍

• Objective: Launch groundbreaking GenAI feature.
  
  • KR1: Reduce processing time for queries from 5s to 1s. (Current: 3s, Target: 1s. Change in last week: 0.1s)
  • KR2: Achieve 90% accuracy for model predictions. (Current: 75%, Target: 90%. Change in last week: 1%)
  • KR3: Attract 5,000 beta users for testing. (Current: 2,000, Target: 5,000. Change in last week: 100 users)
  • Time elapsed in quarter: 4 weeks out of 12."
    
    

How to use it: Provide multiple KRs with their current progress and ideally the time elapsed in the cycle. You can also specify an "ideal" progress rate for the AI to use as a benchmark.

Expected AI Output (Examples)

KR1 (Processing Time):

Status: Off track. Only 0.1s improvement in the last week when 2s improvement needed over 8 weeks. The rate is too slow.

Potential Reasons: Technical blockers more complex than anticipated, insufficient engineering resources, lack of focused effort on optimization.

Interventions: Conduct a deep-dive technical spike to identify specific bottlenecks; allocate dedicated senior engineering talent; explore external libraries or APIs for performance boosts.

KR2 (Model Accuracy):

Status: Yellow/At risk. Only 1% improvement last week, needs 15% in 8 weeks. Requires more acceleration.

Potential Reasons: Insufficient training data, model architecture limitations, complex edge cases.

Interventions: Acquire/generate more diverse training data; explore alternative model architectures; dedicated data scientist sprint for error analysis and targeted improvements.

‍

KR3 (Beta Users):

Status: Significantly off track. Only 100 new users last week, needs 3,000 more in 8 weeks.

Potential Reasons: Ineffective marketing/outreach strategy, beta program not appealing enough, technical issues with signup process, low awareness.

Interventions: Revamp beta landing page and signup flow; launch targeted social media ads; partner with relevant communities/influencers; offer stronger incentives for sign-up and participation.