Use AI to create a high-performance culture
AI can help you instrument and scale a high-performance culture across your team. It's something that can help today. Here's how.
How do you build a high-performance culture on your team, and how does AI help?
The number one trait of building a culture that breeds average performance is celebrating work instead of outcomes. The place you see this is in your team's self-reviews and how managers grade those self-reviews.
Your employees celebrate their work and the hours they logged, and managers grade them high because they did a lot of stuff.
Work doesn't matter; IMPACT matters.
As someone who manages hundreds and hundreds of people, spot-checking reviews is one of the best ways to identify this trend happening across teams.
What do I mean by impact? Not everyone has a metric to track against.
I agree with this, but every marketer can articulate the outcome/impact of their work. It's a critical skill to learn as it helps you understand why what you do is important.
Here are 3 outcomes that are pretty translatable across most companies:
a. Metric Improved – A quantifiable improvement in performance (e.g., revenue, deals, conversion rates, efficiency, adoption).
b. Customer Experience Improved – A clear enhancement to the customer journey (e.g., smoother onboarding, better product usability, improved engagement).
c. Business Operation Improved – A meaningful internal process change (e.g., streamlined workflows, better collaboration, automation).
These are certainly open for debate, but I suspect 1 and 2 are pretty standard, and number 3 is more tricky.
AI can help you instrument this and scale it
AI is going to be an incredible performance coach. Today, you can have AI track goals, provide status summaries, and provide the first draft of a person's review. You'd still do the final pass and edits, but AI can help you move towards outcomes vs. 'doing the work'.
Here's an example prompt that can start to grade your self review or employees self review on outcomes:
PROMPT
Evaluate the following self-review based on how effectively it ties work to business outcomes. Categorize each example into one (or more) of the following buckets:
a. Metric Improved – A quantifiable improvement in performance (e.g., revenue, deals, conversion rates, efficiency, adoption).
b. Customer Experience Improved – A clear enhancement to the customer journey (e.g., smoother onboarding, better product usability, improved engagement).
c. Business Operation Improved – A meaningful internal process change (e.g., streamlined workflows, better collaboration, automation). It must be clear how that business operating getting better has been impactful for the business. Don’t assume you know what the improvement was, it has to be clearly stated. An example would be is improving a sales enablement process that lead to sales being able to do more demos. The user must state very clear what the business impact was of this type of work.
Assessment Criteria:
1. Count the total number of examples that fit into at least one bucket.
2. Count the number of statements that do not fit into any bucket (i.e., vague, task-based, or lacking a clear outcome).
3. Calculate the percentage of the self-review that lacks a clear business outcome.
4. Provide feedback:
If most of the review ties to business outcomes → ✅ Strong review
If a significant portion is task-focused/vague → 🔄 Needs improvement with guidance on how to reframe
Format:
Create a Table with columns for each statement and another column for what category the statement fitted into e.g. Metric, Customer Experience, Business
Of course the way you improve this is tie it back to the deliverables/goals for that person e.g. what the person said they'd do during this period of time.
This prompt won't fix bad managers. But it will EXPOSE them.
AI can track outcomes
AI can measure impact
AI can spot patterns
But it only works if you build a culture obsessed with outcomes, not activity
AND
Build consistent processes across your team and internal documentation of those processes.
Once you do that, AI can help to start helping you to instrument a high performance culture by first weeding out people who aren't articulating the impact of their work.
The future belongs to teams who:
Track outcomes religiously
Grade impact objectively
Automate the busy work
You can easily build upon this use case to have AI provide recommendations on how that person can do better. Coming soon.
Until Next Time,
Happy AI’fying,
Kieran