Analysis Overview
Segflow’s analyses help you move from “what happened?” to “what should we do next?” They run inside your Flow Boards, using the relationships you’ve mapped between bets and metrics — so insights stay connected to the decisions they’re meant to support.
Use them to answer:
- What’s most likely driving our KPI?
- Where does the system compound (or leak)?
- Which bets are worth doing, and which are urgent?
- What should we watch to learn faster and act earlier?
Analyses use your Board connections as input.
How it worksA practical way to use the models together.
Recommended sequenceSee how the workflow looks end-to-end.
Example (B2B SaaS)Why Analyses Matter
Reports show outcomes. Roadmaps show effort. Analyses reveal leverage — the mechanics of how work creates impact, where timing matters, and what to do next.
Without this layer:
- KPI movement becomes hard to explain (and hard to repeat).
- Prioritisation becomes opinion-heavy.
- Teams optimise local metrics instead of system outcomes.
- Learning is slow because lagging KPIs react late.
Analyses are decision-support. They don’t “prove causation” by themselves — they help you focus on the most likely levers and validate them with judgment, experiments, and iteration.
How Analyses Work Inside Boards
Each Flow Board is a connected system: bets, metrics, and relationships. When you run an analysis, it doesn’t start from scratch — it uses your Board structure as the input.
Connections define the logic
Your relationships (Bet → Metric, Metric → Metric, and loops) determine what the model can test and how results stay contextual.
Context stays intact
Analyses run within your Board scope: the metrics you chose, the time window, and the links that reflect your strategy.
Insights remain actionable
Results map back to the same nodes and edges you use to plan — so it’s clear what to change, what to monitor, and what to prioritise.
The Analysis Lenses
| Analysis | What it helps you decide |
|---|---|
| Key Metric Driver Analysis | Which connected metrics most likely explain KPI movement (contribution, direction, confidence). |
| Growth Loop Analysis | Whether a loop compounds or decays, where leverage is highest, and where momentum leaks. |
| Leading Indicators Analysis | Which metrics move ahead of your KPI (lead time + reliability) so you can act earlier. |
| Bet Impact Analysis | Which bets have the best leverage per effort (plan now, then measure lift after shipping). |
| Cost of Delay Analysis | Which bets are most urgent to ship first (value lost per day + CD3). |
Start small: one KPI, 5–10 candidate drivers, and a handful of bets is enough to get decision-quality signal.
Quick Guide: Where to Start
| Situation | Recommended analysis |
|---|---|
| Your KPI changed and you need answers | Key Metric Driver Analysis |
| You want to understand compounding dynamics (or leaks) | Growth Loop Analysis |
| You need earlier signals than your KPI | Leading Indicators Analysis |
| Multiple bets are competing for focus | Bet Impact Analysis |
| Multiple bets look valuable, but you need sequencing | Cost of Delay Analysis |
Recommended Sequence
These analyses are most powerful when you treat them as a workflow — not isolated reports. A simple sequence that works well for most teams:
1. Find leverage (what to move)
Run Key Metric Driver Analysis to identify the few metrics that most likely explain KPI movement. Use tiers + contribution + direction to pick the levers you want to influence.
2. Check for compounding (where the system reinforces)
If your Board contains a loop, run Growth Loop Analysis to see whether it compounds (Loop Gain > 1) or decays (Loop Gain < 1), and where leverage is highest inside the loop. This helps you avoid scaling effort into a system that can’t yet “return value.”
3. Pick early signals (what to watch)
Run Leading Indicators Analysis against your KPI (or a top-tier driver) to find credible upstream signals and their lead times. Use these as your early-warning layer and for faster bet validation.
4. Rank bets (what to do)
Connect bets to the levers you care about and run Bet Impact Analysis to compare leverage per effort (and risk-adjusted scores when uncertainty is wide).
5. Sequence bets (what to do first)
Run Cost of Delay Analysis to quantify urgency (value lost per day) and break ties fairly using CD3 (urgency per build time).
This is a loop, not a one-way funnel. After shipping, use leading indicators and measured lift to update confidence — then re-run Key Metric Driver and Growth Loop analyses to see what actually changed.
What Value It Gives Together
- A shared definition of leverage (drivers and loops), not just “important metrics”.
- A clearer bet ranking grounded in expected lift, effort, and uncertainty.
- Better sequencing by urgency, not just impact.
- Earlier learning via leading signals, so you don’t wait for lagging KPIs to react.
- A tighter strategy loop: diagnose → prioritise → ship → measure → refine.
Example: B2B SaaS
Scenario: You’re a product-led B2B SaaS. Your north star is weekly subscription revenue, but it moves slowly. You need faster signals and clearer prioritisation.
How you might use the sequence:
- Key Metric Driver Analysis shows the top tier drivers are churn rate (negative), activation rate (positive), and upgrade rate (positive). You align on a focus: reduce churn and improve activation.
- Growth Loop Analysis shows your “invite loop” (activated teams → invites → new teams → activation) is decaying (Loop Gain < 1), with a weak link at “activated → invites”. You decide to strengthen that link before scaling acquisition.
- Leading Indicators Analysis finds activated teams leads revenue by ~2 weeks and is reliable. You make it your early warning metric and define a playbook for what to investigate if it drops.
- You create bets (onboarding improvements, invite prompts, cancellation flow fixes) and run Bet Impact Analysis to rank them by leverage per effort.
- You run Cost of Delay Analysis and see the churn fix has the highest cost per day (leakage is immediate), so it goes first — even if another bet has similar ROI.
Result: You ship fewer bets, earlier. You learn faster, and you compound improvement by strengthening the system (drivers + loops) instead of chasing isolated metric spikes.
Summary
Use these analyses as a connected workflow: identify leverage, understand compounding, choose early signals, rank bets, and sequence by urgency. When you run them inside a Board and iterate after shipping, strategy becomes easier to explain — and faster to improve.