Marginal Gain Dashboard¶
A monitoring surface — instantiates Plateau Detection and Switching
Puts marginal response, its trend, its cost, and its confidence on one shared surface so a plateau is seen by decision-makers rather than argued from anecdote.
Marginal Gain Dashboard is the visibility instrument. It does not compute the plateau — it displays it, taking the marginal-response signal and rendering it as a standing, shared surface: the margin, its trend, what the next unit costs, how confident the reading is, and what continuing is forgoing. Its distinguishing job is legibility and persistence. A plateau argued in a meeting is anecdote that fades; a plateau that has been sitting red on a dashboard for three weeks, in front of the people who control the budget, is a fact that demands a decision. It also keeps watching after a switch, so a team can see at a glance whether the new path actually restored the margin.
Example¶
A city's recycling office runs neighborhood outreach — mailers, door-knocks, events — to lift the diversion rate. A Marginal Gain Dashboard shows, per neighborhood, the marginal tons diverted per outreach dollar rather than raw tonnage. In the three earliest-adopting neighborhoods the marginal-tons line has been flat for a quarter and shaded amber, with a side panel reading "≈$0 next-dollar yield here vs. ≈$0.40/ton in the three newest neighborhoods" — the opportunity-cost view. Because the flat margin is visible and time-stamped, the quarterly budget meeting no longer debates whether the mature neighborhoods have "plateaued"; they can see it, and they can see where the same dollar would buy more. When outreach is later shifted, the dashboard's post-switch panel tracks whether the reallocated dollars actually move the newer neighborhoods' margin.
How it works¶
- Display the margin, not the total. The headline is marginal response and its trend, so a healthy cumulative total can't hide a dead margin.
- Attach cost and an opportunity-cost comparison. Each path's next-unit yield sits next to what the same unit would yield on the best alternative path, making "staying" visibly cost something.
- Persist and time-stamp. The signal stays on screen with its history, so a plateau accrues visible tenure instead of being re-litigated each meeting.
- Keep watching after a switch. A post-switch panel tracks whether the new path's margin recovers, closing the loop.
Tuning parameters¶
- Metric proximity — how close the displayed response is to the real outcome. Push it toward the outcome and it resists vanity; pull it toward easy-to-log activity and it flatters.
- Refresh cadence — live, daily, or per-review. Faster surfaces plateaus sooner but shows more noise as if it were signal.
- Confidence display — whether each reading carries an error band. Bands slow premature calls; their absence makes a shaky number look like a verdict.
- Opportunity-cost baseline — what "the best alternative" is measured against — a specific rival path, the portfolio average, or a hurdle rate.
When it helps, and when it misleads¶
Its strength is converting a plateau from a matter of opinion into a shared, standing fact with a price attached — which is often the difference between a plateau that is noticed and one that is acted on.
Its danger is that a dashboard is only as honest as its metric. Point it at an easy-to-move vanity number instead of the real response and it will show reassuring gains while the outcome stalls.[n1] A glossy surface also lends false confidence to a noisy reading, and the classic misuse is decorative: assembling the dashboard to ratify a decision already taken rather than to inform one — filtering to the window and metric that show what leadership wants. The discipline is to fix the metric and window before reading the chart, and to display confidence, not just a number.
How it implements the components¶
Marginal Gain Dashboard fills the visibility-and-tracking subset — it makes signals legible, it doesn't generate the underlying detection logic:
marginal_response_metric— the surface's headline; it renders and trends the margin (consuming the computation from the detector).opportunity_cost_signal— the panel that shows what continuing the plateaued path forgoes relative to the best alternative.post_switch_monitoring— the panel that keeps watching whether a switched-to path restores the margin.
It does not scope the pathway or set the plateau floor (input_pathway_scope, plateau_threshold → Diminishing Returns Detection), and it does not enforce a stop or pick the alternative (escalation_stop_rule, alternative_strategy → Escalation Stop Workflow, Strategy Switch Decision Tree).
Related¶
- Instantiates: Plateau Detection and Switching — it is the surface that makes plateau evidence visible to the people who decide.
- Consumes: Diminishing Returns Detection supplies the marginal metric the dashboard displays.
- Sibling mechanisms: Diminishing Returns Detection · Controlled Experiment After Plateau · Escalation Stop Workflow · Plateau Review Cadence · Ad Fatigue Switching · Training Plateau Adjustment · Product Growth Plateau Response · Process Redesign After Plateau · Saturation-Aware Resource Allocation · Strategy Switch Decision Tree · Expert-Governed Modality Change
Editorial Notes¶
Form Classification¶
Form family: Monitoring, Sensing & Alerting
Rationale: Marginal Gain Dashboard operates as an ongoing sensing arrangement that repeatedly observes actual state and surfaces changes or alerts because it puts marginal response, its trend, its cost, and its confidence on one shared surface so a plateau is seen by decision-makers rather than argued from anecdote.
Independent corroboration: The frozen evidence defines Marginal Gain Dashboard as 'Puts marginal response, its trend, its cost, and its confidence on one shared surface so a plateau is seen by decision-makers rather than argued from anecdote', so its operative form is Monitoring, Sensing & Alerting.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Economics & Finance
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: The focus on incremental response and declining return is grounded in marginal economic analysis.
Related originating lineages:
- Data Science & Analytics — Trend estimation, confidence display, and instrumentation materially supply the dashboard.
- Organizational & Management Science — Decision-facing monitoring of incremental program response belongs to performance management.
- Statistics & Experimental Design — Trend estimation and confidence intervals materially shape whether an apparent plateau is real.
Review resolution: Light authoritative research supports economics_finance as the primary provenance: The focus on incremental response and declining return is grounded in marginal economic analysis. The Green Book grounds appraisal in marginal social value, incremental costs and benefits, and decision-relevant monitoring. The competing reviewed lineage (organizational_management) and other formative traditions remain explicit alternates rather than being erased or confused with downstream applicability. origin_mode=cross_disciplinary_synthesis records the relationship among those origin traditions, while domain_reach=multi_domain separately records how broadly the generalized mechanism can be applied.
Attribution caveat: The dashboard implementation is an encyclopedia synthesis of marginal analysis, statistics, and management reporting. The exact dashboard is an encyclopedia synthesis around marginal-return monitoring.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Researched adjudication after independent review; medium confidence.
Sources consulted:
- https://www.gov.uk/government/publications/the-green-book-appraisal-and-evaluation-in-central-government/the-green-book-2026 — The Green Book grounds appraisal in marginal social value, incremental costs and benefits, and decision-relevant monitoring.
Notes¶
[n1] Vanity metrics — numbers that rise reassuringly but aren't causally tied to the outcome (raw impressions, cumulative signups). A dashboard built on them shows gains while the real response plateaus; the corrective is choosing a response metric close to the intended outcome. ↩