Daniel Schmidt’s Post

While KPI trees, journey maps, and growth loops are valuable today for strategic alignment and data insights, their real power will be giving AI agents business context. To illustrate, imagine two AI agents who are designed to help with executive decision-making. AI agent 1 says: “ARR growth declined 7% this quarter.  Based on the most common feature requests, consider adding more advanced  analytics features to make your product more competitive.” AI agent 2 says: "ARR growth declined 7% this quarter. The root problem is that enterprise clients at the “expansion stage” are missing specific third-party integrations. Consider asking the Platform Squad to evaluate building an API for these services.” I think it's fair to say that agent 2's suggestion is potentially useful while agent 1's insight is too generic. In this scenario, both agents have access to company data: - Strategy docs - Metric data - Customer feedback - Code version history - Project management history - Public data (e.g., earning reports) But agent 2 has access to business context: - KPI trees - Customer journey maps - Growth loops - Mappings between teams/projects/bets and business goals. For AI agents to go beyond speeding up execution, business context will make the difference.

I'm surprised (and excited) to see companies using DoubleLoop as a sales tool. E.g., imagine your company is selling to public company X. With our map builder AI, we can convert your prospect's publicly disclosed earning report into a business driver map populated with a real data. The business driver map shows the areas of their business where they need to make improvements. With that, you can visually show how your product will influence their most critical business drivers. Your prospect can show their executive team how your service will influence the metrics that determine their compensation. If you want to try this out, ping me.

When doing workshops with companies, it's interesting when two people think the same metric should go in opposite directions. For example, take a metric like "User actions per session." On the one hand, making that number go up could be a positive sign of engagement. On the other hand, making that number go down could mean that users are accomplishing their goal more efficiently. Another example is "Average ARR per new customer." On the one hand, when customers pay more that's good. On the other hand, lowering the price might be an effective way to grow ARR through attracting more customers. In these ambiguous situations, when you have the benefit of historical data to analyze, how do you make a determination whether moving a metric up or down is better for the business? I'm in particular looking for data science answers, but other thoughts are welcome too.

When PMs hear product ideas from others in their company, sometimes we have a gut reaction of "ooh, interesting idea!" or "this is not worth seriously considering." What factors would lead you to have one reaction over the other? For example, factors might include feasibility, alignment with strategy, alignment with user behavior patterns, or "we're tried this before and it didn't work." What else?

Complete business context is in the mind of no one, and this will never change. The executive team, seeing from a high altitude, sees the big picture of the market and business performance, but they don't know the truth on the ground. Team members doing the hands-on work know the terrain in their area, but they have less ability to see the big picture or what's happening in all the other areas. The data is too overwhelming for humans to bridge this gap. So at DoubleLoop, we're making business context portable (as shown in the image) to enable AI to bridge the gap for us. The implications of AI based on portable business context are deep: * Enable every team member to see how their work fits into the big picture. * Enable executive-level strategy to keep in sync with work. * Enable data science to reveal the highest-leverage bets, and forecast their impact. We've already created a custom GPT that takes DoubleLoop data and provides immediate strategy insights/recommendations, based on both industry best practices and data analysis of companies' KPI graphs, journey maps, strategy, work data, metric data, and customer data. But the part I'm really excited about is how portable business context opens up the door to a two-sided market place that connects companies with AI strategy advisor agents that go beyond what we've imagined . Would you want to build your own GPT that analyzes portable business context?

Marc Fonteijn makes a key point that often metrics should be modeled with cyclical or networked graphs instead of linear trees. For example, the North Star Framework and growth loops are some of the most popular frameworks in DoubleLoop and they're not linear trees. See a bunch of examples here: https://lnkd.in/ghSCjm7v The challenge is that, when you're operationalizing your metric graph, it gets way more complex when they're not linear trees. We're feeling this pain right now in building our metric-integrated roadmap tool where initiatives can target any metric in a complex graph. With trees, it's easy to roll-up initiatives at different altitudes; with graphs, these rollups are trickier. But I think we've rounded the corner on it and will have something neat to share soon...

Is there anything more painful in product management than the constant toggling between "the how" and the "the why"? Within minutes, PMs switch between analyzing implementation details with the eng team and explaining the business context to stakeholders. The same PM can be criticized by some as "too in the weeds" and by others as "not technical enough." Switching between writing engineering tickets and communicating strategy requires playing different language games with different rules, conventions, and pitfalls. The people who are good at toggling between "the how" and "the why" spend endless hours doing translation, and their translations constantly break every time stuff changes.

Amazing post from Timo Dechau on a map for measuring data platforms. If you would like to implement this map in DoubleLoop to make it living and breathing with your data, hit me up.

We've oscillated on whether we want to take an anti-OKR stance at DoubleLoop or embrace OKRs while trying to help improve how they're executed. But Tom Critchlow is making a compelling case here for why we need a post-OKR world...

We’re introducing a new series of free data-led Strategy Workshops. The failure of classic strategic planning is that it’s too disconnected from the real data and the work of teams on the ground. Strategy is made in slides and is out of date the moment it’s made. DoubleLoop is a platform to build strategy that stays in sync. But it requires a commitment to get your data into the platform (otherwise we’re just doing strategy theater). That’s why I’m offering free, personalized strategy workshops for companies who connect their data sources to DoubleLoop. I’ve run over 50 of these over the last few months and they’re valuable for you because we quickly surface insights about your strategy and help you gain clarity - and they’re valuable for me because they help me stay connected to real customer problems and approaches (crucial for building a startup!) Step 1: sign up here: https://lnkd.in/gvYP4nHm Step 2: connect your data sources Step 3: have a free personalized 1 hour strategy workshop with me (and invite whoever you'd like)