River Logiceode docs

Read results

Read what a solved scenario says: the scoreboard, each KPI and its change against Base, how sure each number is, and where it came from.

Needs a model openNeeds a solved scenario

Where: Models→a model→Scenarios→Base

A solved scenario's screen opens beside its conversation with RIA: in the rail, under the open model, click Base or a what-if (or click its row on Scenarios). The screen leads with a scoreboard, then a tile per KPI with its change against Base. Click any tile to see how the number was worked out.

Read the scoreboard

Under the heading (the what-if's name on model, or on Base the model's name and Live model):

  • Base: the baseline's value for your goal. A model that minimizes cost shows a positive cost. A dash means Base hasn't been solved; Cost unknown means the prices are placeholders. A figure too long for the card is shortened; hover for all of it.
  • Scenarios solved: n of total. A scenario solved before the model last changed doesn't count; n to solve again under the count says how many there are. See Keep results current.

When the checks on Base's figure don't all agree with it, the figure is still shown, in amber, with a line under it naming the check, such as Plan doesn't add up to this, P&L differs from this, Up to n% better possible or Not sure what this measures; Doesn't check out when several disagree. A run that reported errors turns it red: This run hit errors. Don't rely on this total. Hover the line for each check's sentence.

The same figure leads the model's record; see Read a model's record.

Read the KPI tiles

KPIs shows the model's KPIs for the scenario on screen. Each tile has the KPI's name, its change against Base, and its value.

You seeMeans
+$420 vs base, +3.2 points vs base, +12.00 unit vs baseThe change against Base, with an arrow. No change, no line. Green means up and red means down, even for a cost, except on the KPI that tracks your goal, where green means better (on a cost model, lower).
$1.24B, $1.24M, $12,345Large amounts shortened. Hover for the full figure.
Verified"Checked against the solved plan"
Estimated"Worked out from other figures, not checked"
Rests on a placeholder / Rests on a limit not yet givenThe number depends on a stand-in, or on a limit you haven't given yet.
Not computed, or a sentenceThe KPI has no value for this scenario. Hover for why.
Partial coverage (triangle)The KPI doesn't cover everything. Hover for what's missing.
All types (layers)In a workspace filtered to one type: this KPI has no per-type figure, so it shows all types.
  • Captions turns on a line under each value saying what it means and, where it has one, Source: where the figure comes from. It's off by default and remembered.
  • Refresh figures works every KPI out again from your last solve, without solving again: "Recalculate every KPI from your last solve. The model is not run again." It reports how many scenarios' figures changed. It shows to owners and contributors once the model has KPIs.
  • Tiles show 6 at a time. View more (n more) and View less page through the rest.
  • Drag a tile onto another to reorder. The order is saved for the model, for everyone.
  • Add KPI with RIA and Add custom KPI, after the tiles, ask RIA for a KPI; editing and removing are on the tiles. See KPIs and custom metrics.

A new model reads Selecting KPIs for this model while RIA picks them. Reload fetches them if they already exist.

Prove a number

Click a KPI tile. Prove this number opens with the KPI, its value against Base, and how sure eode is. If the number hasn't been traced before, eode starts tracing it: "Tracing this number back to the model sources…"

SectionShows
Partial coverageWhat the number leaves out, such as Unpriced: items.
Root-cause summaryWhat produced the number, in a sentence or two.
SourceWhere the figure comes from, in words.
The arithmeticThe formula, when it says more than the summary.
Breakdown (n)The lines that add up to it: Line item, Active, Base, Δ, Source.
What's setting this number (n binding)The limits holding it where it is, what each means, the level it sits at, and what one more unit is worth.

Re-trace works it out again in full, which takes about 30 seconds. After a quick trace, the button reads Run full diagnostic instead. If the trace fails, Retry trace. Press Esc or click outside to close.

Tell how sure a number is

Each tile carries Verified or Estimated. In Prove this number and in Verify, a fuller badge adds Stale (solved before the model last changed) and Unverified. Hover the badge for the reasons.

  • Verified: eode confirmed the value from the solved plan.
  • Estimated: worked out another way. A full trace can upgrade it.

Read a what-if with no plan

When eode proves a what-if's change can't be met, the screen shows a card in place of a result: "No plan: the change can't be met" ("No plan: nothing limits the plan" when nothing holds it in). Under it are the diagnosis saved with the run and up to 5 of the limits in conflict, each with its cause and the fix proposed. There are no figures. Diagnose opens a workspace on it; see Take a result further.

Check an imported workbook

On a model imported from a workbook, BASE DATA and BASE show your workbook's own figures instead of the usual tiles.

  • BASE DATA shows up to 5 of the workbook's headline KPIs, each with the Sheet!Cell it came from, then Imported KPIs, grouped into the totals eode worked out from the model and those Read from workbook.
  • A comparison card puts each KPI from your workbook beside the model's figure, with Δ colored green under 10%, amber under 50%, red beyond. On BASE it is titled Workbook vs solved BASE: same KPI, two sources.

Take a result further

ButtonWhat it does
Open in workspaceOpens the workspace for this scenario. The first press creates it, named model · scenario, with widgets for this scenario; later presses open the same one.
DiagnoseReplaces Open in workspace on a scenario with no plan. Opens a workspace laid out to show what's blocking it and how to fix it, made the same way.
Run with scenarioOn the Workflows for model card: starts a run of that workflow with this scenario's results. See Workflow runs.

Both workspace buttons are off until the scenario has a result; Open in workspace says "Solve this scenario first. Workspace widgets need a result." Every widget you'll find there (the comparison panel, KPI card and grid, narrative, financial statements, where to invest next, the diagnosis widgets, the network map) is described on Widgets.

If a link or RIA takes you to Scenario Analysis has moved to Workspaces, pick Single-scenario review, N-way comparison or Infeasibility diagnosis and click Open workspace.

Limits

  • 6 KPI tiles per page.
  • A full re-trace takes about 30 seconds.
  • BASE DATA shows up to 5 headline KPIs.
  • A no-plan card lists up to 5 limits in conflict.

On this page