Skip to content

Include non-media (baseline) analyses in incremental outcome processor. #1504

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
copybara-service wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Include non-media (baseline) analyses in incremental outcome processor. #1504

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
118 changes: 85 additions & 33 deletions meridian/schema/processors/marketing_processor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -482,14 +482,15 @@ def _generate_marketing_analyses_for_media_summary_spec(
)

return self._marketing_metrics_to_protos(
media_summary_metrics,
non_media_summary_metrics,
baseline_outcome,
secondary_non_revenue_kpi_metrics,
secondary_non_revenue_kpi_non_media_metrics,
response_curves,
marketing_analysis_spec,
date_intervals,
metrics=media_summary_metrics,
non_media_metrics=non_media_summary_metrics,
baseline_outcome=baseline_outcome,
secondary_non_revenue_kpi_baseline_outcome=None,
secondary_non_revenue_kpi_metrics=secondary_non_revenue_kpi_metrics,
secondary_non_revenue_kpi_non_media_metrics=secondary_non_revenue_kpi_non_media_metrics,
response_curves=response_curves,
marketing_analysis_spec=marketing_analysis_spec,
date_intervals=date_intervals,
)

def _generate_media_and_non_media_summary_metrics(
Expand Down Expand Up @@ -556,6 +557,7 @@ def _generate_marketing_analyses_for_incremental_outcome_spec(
if incremental_outcome_spec is None:
return []

selected_times = resolver.resolve_to_enumerated_selected_times()
compute_incremental_outcome = functools.partial(
self._incremental_outcome_dataset,
resolver=resolver,
Expand All @@ -569,14 +571,34 @@ def _generate_marketing_analyses_for_incremental_outcome_spec(
# This contains either a revenue-based KPI or a non-revenue KPI analysis.
incremental_outcome = compute_incremental_outcome(use_kpi=self._kpi_only)

# Baseline outcome is not computed for new data.
if incremental_outcome_spec.new_data is not None:
baseline_outcome = None
else:
baseline_outcome = self._analyzer.baseline_summary_metrics(
confidence_level=marketing_analysis_spec.confidence_level,
aggregate_times=incremental_outcome_spec.aggregate_times,
selected_times=selected_times,
).sel(distribution=constants.POSTERIOR)

secondary_non_revenue_kpi_metrics = None
secondary_non_revenue_kpi_baseline_outcome = None
# If the input data KPI type is "non-revenue", and we calculated its
# revenue-based KPI outcomes above, then we should also compute its
# non-revenue KPI outcomes.
if not self._revenue_kpi_type and not self._kpi_only:
secondary_non_revenue_kpi_metrics = compute_incremental_outcome(
use_kpi=True
)
if incremental_outcome_spec.new_data is None:
secondary_non_revenue_kpi_baseline_outcome = (
self._analyzer.baseline_summary_metrics(
confidence_level=marketing_analysis_spec.confidence_level,
aggregate_times=incremental_outcome_spec.aggregate_times,
selected_times=selected_times,
use_kpi=True,
).sel(distribution=constants.POSTERIOR)
)

date_intervals = self._build_time_intervals(
aggregate_times=incremental_outcome_spec.aggregate_times,
Expand All @@ -586,7 +608,8 @@ def _generate_marketing_analyses_for_incremental_outcome_spec(
return self._marketing_metrics_to_protos(
metrics=incremental_outcome,
non_media_metrics=None,
baseline_outcome=None,
baseline_outcome=baseline_outcome,
secondary_non_revenue_kpi_baseline_outcome=secondary_non_revenue_kpi_baseline_outcome,
secondary_non_revenue_kpi_metrics=secondary_non_revenue_kpi_metrics,
secondary_non_revenue_kpi_non_media_metrics=None,
response_curves=None,
Expand Down Expand Up @@ -620,9 +643,11 @@ def _build_time_intervals(

def _marketing_metrics_to_protos(
self,
*,
metrics: xr.Dataset,
non_media_metrics: xr.Dataset | None,
baseline_outcome: xr.Dataset | None,
secondary_non_revenue_kpi_baseline_outcome: xr.Dataset | None,
secondary_non_revenue_kpi_metrics: xr.Dataset | None,
secondary_non_revenue_kpi_non_media_metrics: xr.Dataset | None,
response_curves: xr.Dataset | None,
Expand Down Expand Up @@ -682,19 +707,21 @@ def _marketing_metrics_to_protos(
):
non_media_analyses.append(channel_analysis)

marketing_analysis = marketing_analysis_pb2.MarketingAnalysis(
date_interval=date_interval,
media_analyses=media_analyses,
non_media_analyses=non_media_analyses,
)
if baseline_outcome is not None:
baseline_analysis = self._get_baseline_metrics(
marketing_analysis_spec=marketing_analysis_spec,
baseline_outcome=baseline_outcome,
secondary_baseline_outcome=secondary_non_revenue_kpi_baseline_outcome,
start_date=start_date_str,
is_revenue_type=(not self._kpi_only),
)
marketing_analysis.non_media_analyses.append(baseline_analysis)
non_media_analyses.append(baseline_analysis)

marketing_analysis = marketing_analysis_pb2.MarketingAnalysis(
date_interval=date_interval,
media_analyses=media_analyses,
non_media_analyses=non_media_analyses,
)
marketing_analyses.append(marketing_analysis)

return marketing_analyses
Expand Down Expand Up @@ -826,15 +853,20 @@ def _get_baseline_metrics(
self,
marketing_analysis_spec: MarketingAnalysisSpec,
baseline_outcome: xr.Dataset,
secondary_baseline_outcome: xr.Dataset | None,
start_date: str,
is_revenue_type: bool,
) -> non_media_analysis_pb2.NonMediaAnalysis:
"""Analyzes "baseline" pseudo-channel outcomes over the given time points.

Args:
marketing_analysis_spec: A user input parameter specs for this analysis.
baseline_outcome: A dataset containing the model's baseline summary
metrics.
secondary_baseline_outcome: A dataset containing the model's secondary
baseline summary metrics.
start_date: The date of the analysis.
is_revenue_type: Whether the baseline outcome is revenue type.

Returns:
A `NonMediaAnalysis` representing baseline analysis.
Expand All @@ -855,25 +887,46 @@ def _get_baseline_metrics(
contribution_share, marketing_analysis_spec.confidence_level
),
)
baseline_analysis = non_media_analysis_pb2.NonMediaAnalysis(
outcomes = [
outcome_pb2.Outcome(
contribution=contribution,
kpi_type=kpi_type_pb2.REVENUE
if is_revenue_type
else kpi_type_pb2.NON_REVENUE,
)
]
if secondary_baseline_outcome is not None:
if constants.TIME in secondary_baseline_outcome.coords:
secondary_baseline_outcome = secondary_baseline_outcome.sel(
time=start_date,
)
secondary_incremental_outcome = secondary_baseline_outcome[
constants.BASELINE_OUTCOME
]
# Convert percentage to decimal.
secondary_contribution_share = (
secondary_baseline_outcome[constants.PCT_OF_CONTRIBUTION] / 100
)
secondary_contribution = outcome_pb2.Contribution(
value=common.to_estimate(
secondary_incremental_outcome,
marketing_analysis_spec.confidence_level,
),
share=common.to_estimate(
secondary_contribution_share,
marketing_analysis_spec.confidence_level,
),
)
secondary_outcome = outcome_pb2.Outcome(
contribution=secondary_contribution,
kpi_type=kpi_type_pb2.NON_REVENUE,
)
outcomes.append(secondary_outcome)

return non_media_analysis_pb2.NonMediaAnalysis(
non_media_name=constants.BASELINE,
non_media_outcomes=outcomes,
)
baseline_outcome = outcome_pb2.Outcome(
contribution=contribution,
# Baseline outcome is always revenue-based, unless `revenue_per_kpi`
# is undefined.
# TODO: kpi_type here is synced with what is used inside
# `baseline_summary_metrics()`. Ideally, really, we should inject this
# value into that function rather than re-deriving it here.
kpi_type=(
kpi_type_pb2.KpiType.NON_REVENUE
if self._kpi_only
else kpi_type_pb2.KpiType.REVENUE
),
)
baseline_analysis.non_media_outcomes.append(baseline_outcome)

return baseline_analysis

def _compute_outcome(
self,
Expand Down Expand Up @@ -918,7 +971,6 @@ def _compute_outcome(
confidence_level,
),
)
# Convert percentage to decimal.
if constants.PCT_OF_CONTRIBUTION in data_vars:
contribution_share = (
media_summary_metrics[constants.PCT_OF_CONTRIBUTION] / 100
Expand Down
Loading