The AI Companion for Key Results
A powerful application of artificial intelligence is to serve as an intelligent companion for teams and individuals responsible for achieving Key Results (KRs). The core objective is to leverage AI to help make more effective, real-time decisions regarding the progress and strategy for each KR.
The AI-Assisted Workflow
Consider a KR owner who, along with their team, has established an initial plan to progress their Key Result. An AI companion could integrate into this workflow in several key ways:
Proactive Planning & Analysis: The AI agent can assist in analysing the initial plan, identifying potential risks and problems, and thinking through weekly progress goals.
Dynamic Adaptation: As the week unfolds, the assumptions made during planning are tested against real-world data and events. When these assumptions are not validated in practice, the AI can help the team adjust the plan to maintain momentum and continue progressing.
A Multi-Faceted Helper: The AI's role is to act as a consistent partner to fine-tune plans, generate new ideas when a team is stuck, and analyse complex data—such as statistics, reports, and other documents—to uncover insights.
Guiding Philosophy: Continuous Validation
Instead of rigidly adhering to a plan that is no longer working, the team is encouraged to continuously test assumptions and explore new strategies with the help of AI.
This cycle of testing, validating, and adjusting is where the AI provides its greatest value. It facilitates a constant loop of learning and optimization, helping the team navigate from their initial assumptions to the most effective solution for their specific situation.
The ultimate goal is not to replace the KR owner but to augment their capabilities, thereby significantly increasing the probability of successfully reaching the final destination.
Example 1: AI Companion in a Public Health Programme
Imagine a public health department with an Objective stated as:
Objective: To increase the vaccination coverage in rural municipalities to strengthen community health and reduce preventable diseases.
One of the Key Results (KRs) under this Objective is:
KR - To increase the vaccination coverage of children under five years old from 72% to 90%.
The department’s KR owner, a regional coordinator, manages dozens of health units across multiple municipalities. Her team has created an initial action plan that includes outreach campaigns, coordination with local schools, and mobile vaccination units.
Here is how an AI companion supports the KR workflow from planning to execution:
1. Proactive Planning & Analysis
When the coordinator uploads the initial action plan, the AI companion analyses it using historical vaccination data, demographic patterns, and logistical records.
It identifies potential bottlenecks such as:
rural areas with historically low vaccination rates,
seasonal accessibility issues (e.g., rainy-season road closures), and
shortages in specific vaccines at distribution points.
The AI proposes micro-goals for the week, such as
Increase coverage in Municipality X by 2% through school-based vaccination drives.
It also forecasts the expected weekly progress if all assumptions hold, helping the team visualise realistic milestones.
2. Dynamic Adaptation During Execution
Midway through the month, new data show that vaccine delivery delays are affecting three municipalities. The AI flags these deviations automatically and runs scenario simulations.
It suggests reassigning mobile units from nearby areas operating above target and recommends a revised micro-plan that keeps the overall trajectory toward the KR intact.
The AI’s recommendations are not directives but decision-support insights, allowing the KR owner to validate, adjust, or reject suggestions based on local context.
Beyond analytics, the AI companion:
Reads and summarises lengthy field reports from health units to highlight patterns (e.g., parent hesitancy due to misinformation).
Synthesises insights from national-level vaccination data, identifying that similar campaigns succeeded elsewhere when supported by local radio communication.
Generates new ideas when progress stagnates, such as testing reminder SMS/WhatsApp campaigns targeting families registered in the public health database.
This ongoing dialogue with the AI becomes a form of continuous learning, where the team explores and validates hypotheses in real time.
For example, each week, the AI companion may prompt the team to revisit their assumptions:
Did the communication campaign increase attendance as expected?
Are transportation constraints still the main limiting factor?
What new evidence emerged that changes our understanding of the problem?
Instead of rigidly following the initial plan, the team—supported by AI—adopts a first principles approach: questioning, testing, and refining their strategy based on evidence, not tradition or routine.
3. The Outcome
By the end of the quarter, vaccination coverage in the region rose from 72% to 88%, nearly reaching the target.
More importantly, the team developed a data-driven mindset. They no longer waited for quarterly reports to adjust plans; they made informed decisions weekly, guided by evidence and aided by AI analysis.
