Exec – Common Journeys (Focus+Context)
What This Chart Shows
This chart shows the most frequent paths that lead to your selected KPI (conversion/outcome) in a focus + context layout. One focused journey is shown in full detail at the top (steps over time, edge strength, lags, and metadata). The rest appear as context rows below; clicking a context row promotes it to the focus row. It displays what the main paths to conversion are, how strong the steps are, and how long they take.
Only journeys that touch the selected KPI (or the chosen outcome set) are included; if none exist, a tiered fallback (e.g. selected series → KPI group → all outcomes) is used and a badge explains what you're seeing.
Key Questions This Chart Helps Answer
- What are the most common paths that lead to our selected KPI or conversion?
- How strong is each step (TE) and what is the lag between steps (days)?
- Which journey should we focus on first, and how does it compare to others (context rows)?
- When we change KPI or filters, do we see different journeys or a fallback message?
Axes, Metrics, and Units
Element | Description |
|---|---|
| X-axis | Days elapsed. Time along the journey from first step (0) to last step. Same scale for focus and context rows. |
| Y (implicit) | Rows. Top row = focused journey; rows below = context journeys (compact). No tick labels; divider and "Click any row to expand" separate focus from context. |
| Focus row – nodes | Steps in the journey (circles). Position = day; label above = step name (wrapped). Hover: step name, day. |
| Focus row – edges | Lines between steps. Thickness = causal strength (e.g. TE). Color = blue gradient (focus). Inline annotation: "TE X.XXX · Yd" (lag in days). Hover: step A → step B, TE strength, lag (days). |
| Focus row – metadata panel | Right side: journey name, support, instances, steps, avg TE, avg strength, avg lag, duration, p-value, confidence, temporal consistency, edge coverage. |
| Context rows | Compact journey lines and small nodes; thinner edges (slate/gray). Hover: journey name, step, support, "Click to expand." Right: truncated journey name. |
| Badges | When a fallback is used: e.g. "No discovered journeys involve the selected series. Showing journeys for the selected KPI instead." or "Showing strongest direct causal connections instead." etc. |
Control Options Reference
Control | Meaning |
|---|---|
| Outcome group / KPI | Which KPI or outcome set is used. Only journeys that touch this outcome (e.g. end at or pass through target series) are shown when possible. |
| KPI drilldown | When set, narrow to specific series within the KPI group first; fallback to full KPI group then all outcomes. |
| Date range | Time window for attribution. |
| Event categories | Filter by specific event categories (include/exclude). |
| Funnel stages | Filter by specific funnel stages (include/exclude). |
| Search terms | Text search on lever/series labels. |
How to Interpret the Results
- Focus row: The journey you're inspecting. Thicker edges = stronger causal link (higher TE); lag (Xd) = days between steps. Use the metadata panel for support, instances, avg TE, avg lag, and confidence.
- Context rows: Other top journeys. Click a row to move it to the focus row and see full detail. Order is by impact.
- Days elapsed: X position is model time (e.g. cumulative lag or normalized span), not necessarily calendar dates. Use for relative timing and sequence.
- Badges: If a yellow badge appears, a fallback is active (e.g. no journeys for selected series, so showing KPI-level or all-outcome journeys, or CSR direct links instead of multi-step patterns). Read the badge to know what is shown.
- Empty state: "No customer journeys detected yet" means no patterns/chains/CSR data passed the filters; run or check the pipeline and Edge Bridge.
Practical Applications for Marketers
Application | How to use this chart |
|---|---|
| Journey discovery | See the main paths to conversion and their strength and timing. |
| Optimization | Use focus + metadata (support, TE, lag) to decide which journey to improve or replicate. |
| Storytelling | Walk one journey in focus and compare to context; use for exec or planning. |
| KPI sensitivity | Change KPI or drilldown and see how the set of journeys and fallback messages change. |
| Search | Use search terms to restrict to journeys that mention certain channels or steps. |
Common Mistakes & Misinterpretations
Mistake | Why it's a problem | How to avoid |
|---|---|---|
| Treating "days" as calendar dates | X-axis is days elapsed in the model (cumulative lag/span), not literal calendar. | Use for sequence and relative timing, not "on which date." |
| Ignoring badges | When a fallback is used, the set of journeys (e.g. KPI-wide or CSR direct) may not match what you selected. | Read the yellow badge to see which fallback is active. |
| Assuming all journeys are multi-step patterns | If only CSR fallback runs, you see direct causal links (e.g. 2-step), not mined multi-step patterns. | Check badge: "Showing strongest direct causal connections instead." |
| Comparing focus and context without clicking | Context rows are summarized; full metrics are in the focus row. | Click a context row to promote it and compare using the focus metadata. |
| Treating TE as revenue or conversions | TE is causal strength (information flow) in the model, not dollars or conversion count. | Use for relative strength and ordering; use other charts for revenue/outcomes. |
Caveats & Considerations
- Empty state: If no journeys are found from any source, the chart shows "No customer journeys detected yet" and suggests running the pipeline to populate chain or pattern families on the Edge bridge.
- KPI filter: Only journeys that touch the selected KPI (target series set) are shown when possible. If none do, tiered fallback (drilldown → KPI group → all outcomes) is used and a badge explains.
- Tiered fallback: When the exact selection has no journeys, the chart may show journeys for the KPI group or all outcomes, or CSR direct links, with a badge. Read the badge to interpret.
- Up to 24 journeys: At most 24 journeys are shown (context + focus); the rest are not in the view.
- Click-to-focus: Clicking a context row determines which journey is in the focus row; if unset or invalid, the highest-support journey is focused.
- Model-derived: Steps, TE, and lags come from the causal/TCEG pipeline; use for direction and comparison, not as a guarantee of real-world timing or causation.
- Search: When search terms are applied, only journeys whose step labels match are shown; this can reduce the list or leave none.