:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
Python ab-testingai-analyticsanalyticscdpdata-warehouseexperimentsfeature-flagsjavascriptproduct-analyticspythonreactsession-replaysurveystypescriptweb-analytics
import { LemonInput } from '@posthog/lemon-ui'
import { LemonLabel } from 'lib/lemon-ui/LemonLabel'
import { Tooltip } from 'lib/lemon-ui/Tooltip'
import {
ExperimentMeanMetric,
ExperimentMetric,
ExperimentWarehouseNativeMathType,
MathType,
isExperimentMeanMetric,
} from '~/queries/schema/schema-general'
import { ExperimentMetricMathType } from '~/types'
/**
* Threshold turns a per-user accumulated value into a binary outcome ("did this user reach N"),
* so it only makes sense for math types that sum/count per user.
*/
const THRESHOLD_ENABLED_MATH_TYPES = new Set<MathType | ExperimentWarehouseNativeMathType>([
ExperimentMetricMathType.Sum,
ExperimentMetricMathType.TotalCount,
])
export function isThresholdAvailableForMath(math: MathType | ExperimentWarehouseNativeMathType | undefined): boolean {
return math !== undefined && THRESHOLD_ENABLED_MATH_TYPES.has(math)
}
/** Whether the metric has a valid threshold applied, so the header should show a cue.
* A 0 threshold is treated as unset (an always-true proportion is meaningless). */
export const isMetricThresholdSet = (metric: ExperimentMetric): boolean =>
isExperimentMeanMetric(metric) && !!metric.threshold
export const isMetricThresholdCueVisible = (metric: ExperimentMetric): metric is ExperimentMeanMetric =>
isExperimentMeanMetric(metric) && !!metric.threshold && isThresholdAvailableForMath(metric.source.math)
export interface ExperimentMetricThresholdProps {
math: MathType | ExperimentWarehouseNativeMathType | undefined
value: number | undefined
onChange: (value: number | undefined) => void
}
export function ExperimentMetricThreshold({ math, value, onChange }: ExperimentMetricThresholdProps): JSX.Element {
const enabled = isThresholdAvailableForMath(math)
return (
<div className="mt-4">
<LemonLabel
info="Reports the percentage of exposed users whose per-user value reaches or exceeds this threshold. Only available for summed or counted metrics."
className="mb-1"
>
Threshold
</LemonLabel>
<Tooltip title={enabled ? undefined : 'Thresholds are only available for summed or counted metrics'}>
<LemonInput
type="number"
className="max-w-32"
fullWidth={false}
min={0}
step={1}
placeholder="e.g. 100"
value={value}
/**
* Treat empty / cleared (0) / NaN as "no threshold" so dependent fields re-enable.
*/
onChange={(newValue) => onChange(newValue ? newValue : undefined)}
disabled={!enabled}
/>
</Tooltip>
<div className="text-muted text-xs mt-1">
{enabled
? 'For example, reports the percentage of users who reach 100 total words across the experiment.'
: 'Thresholds are only available for summed or counted metrics.'}
</div>
</div>
)
}