The four analytics surfaces every biotech data platform should carry: cohort performance over time, consumer effect signatures, end-to-end conversion, and environment × outcome correlations. Every chart re-derives from live platform data on load.
Yield per plant over time for every genotype with enough harvests to trend. Dashed line is the linear regression for the cohort.
Clone production through consumer logging. Step-conversion % shows where each stage tightens or loses ground.
Pearson r across per-batch aggregates. Advisory signal, not causation.
Which reported effects show up together across consumer logs. Node size tracks frequency, edge weight tracks co-occurrence, color tracks average sentiment.
Each visual re-derives from live platform data every time the page loads. Cohort trajectories come from harvest outcomes joined back to their batch's genotype. The funnel counts each stage in the pipeline. Correlations run Pearson r across per-batch environment aggregates × per-batch outcomes. The effect network clusters consumer-reported effects by co-occurrence. All four are advisory surfaces right now, feeding the same rule-based intelligence engine that powers the rest of the platform. As harvest volume grows and ML models land, these same surfaces tighten without changing shape.