Demo gallery¶
These demos are deliberately plain-language and shareable. They answer: what can I show a colleague in one minute?
Every demo can write JSON, Markdown, SVG/PNG charts, and a social-share card. Flagship demos also have ready-made downloadable share packages.
Fastest beginner examples¶
School attendance after a snow closure¶
- what goes in: One attendance series, a nearby-school control series, and the closure date.
- why it matters: This is a direct before/after question that educators can understand without learning causal-inference jargon first.
- bundled example file:
demo_school_closure_attendance.csv
Retail foot traffic after a store redesign¶
- what goes in: One treated store, several comparison stores, and the redesign date.
- why it matters: It shows the panel path on a business problem that looks nothing like a canonical policy case.
- bundled example file:
demo_retail_foot_traffic.csv
Website conversions after a landing-page redesign¶
- what goes in: One conversion series, peer conversions, search interest, and the redesign date.
- why it matters: It is the plainest possible product-growth explanation of an impact workflow.
- bundled example file:
demo_website_redesign_conversions.csv
Flagship shareable demos¶
Use these first if you want a chart that is easy to explain outside the causal-inference community.
City traffic intervention¶
CSV in, report out: one treated city after a transport intervention.
- question: Did the treated city diverge from its donor-pool counterfactual after the intervention?
- domain:
transport - social angle: Show a colleague how one transport change turned into a city-level counterfactual chart in one command.
- intervention:
2024-03-06 - family:
panel
python -m tscfbench demo city-traffic
python -m tscfbench make-share-package --demo-id city-traffic
- sample download: city-traffic share package
Takeaway: A single CSV can become a treated-vs-counterfactual chart, a placebo report, and a share card.
Product launch signups¶
One product launch, one treated metric, two control series.
- question: How many extra daily signups appeared after the feature launch?
- domain:
product - social angle: A launch story that product and growth teams can understand instantly.
- intervention:
2024-04-23 - family:
impact
python -m tscfbench demo product-launch
python -m tscfbench make-share-package --demo-id product-launch
Takeaway: Event-style counterfactual analysis is easier to explain when the output starts with a chart, not a wall of JSON.
Heatwave and ER visits¶
Medicine / public health demo using one hospital metric before and after an extreme-heat event.
- question: How many excess ER visits appeared during the heatwave window?
- domain:
medicine - social angle: A clean excess-visits chart is much easier to share than a methods-heavy policy benchmark.
- intervention:
2024-07-14 - family:
impact
python -m tscfbench demo heatwave-health
python -m tscfbench make-share-package --demo-id heatwave-health
- sample download: heatwave-health share package
Takeaway: Cross-disciplinary demos make the package legible to scientists who do not start from synthetic-control jargon.
Climate shock and grid demand¶
Climate + energy demo: one treated grid region during a heat-driven demand spike.
- question: How much extra demand hit the treated grid during the climate shock window?
- domain:
climate - social angle: A climate-grid story that works for energy researchers, utilities, and LinkedIn audiences.
- intervention:
2024-08-11 - family:
panel
python -m tscfbench demo climate-grid
python -m tscfbench make-share-package --demo-id climate-grid
- sample download: climate-grid share package
Takeaway: Climate-and-energy narratives are easier to share when the output is a treated-vs-counterfactual demand chart with donor contributions.
Hospital surge during a respiratory outbreak¶
Medicine demo: one hospital-system metric before and after a respiratory outbreak wave.
- question: How much ICU occupancy rose above counterfactual during the outbreak surge?
- domain:
medicine - social angle: An outbreak-surge case that reads like a hospital operations story, not a methods lecture.
- intervention:
2024-01-17 - family:
impact
python -m tscfbench demo hospital-surge
python -m tscfbench make-share-package --demo-id hospital-surge
- sample download: hospital-surge share package
Takeaway: Medicine demos become easier to trust when the workflow writes a chart, a report, and a compact share package by default.
Repo breakout after a launch¶
Internet-native public demo for launch attention, GH stars, or repo adoption.
- question: Was the repo's breakout real, or was it already on that path?
- domain:
open_source - social angle: The most shareable demo: was the launch a real breakout or just trend continuation?
- intervention:
2025-12-18 - family:
impact
python -m tscfbench demo repo-breakout
python -m tscfbench make-share-package --demo-id repo-breakout
- sample download: repo-breakout share package
Takeaway: A repo-breakout share card makes the package legible to internet audiences who may never read a benchmark appendix.
Detector downtime after a solar storm¶
Physics demo: one detector uptime series before and after a solar-storm event, with a reference detector and solar proxy as controls.
- question: How much detector uptime was lost after the solar storm relative to a counterfactual path?
- domain:
physics - social angle: A solar-storm downtime story makes the package legible outside policy and product analytics.
- intervention:
2024-05-18 - family:
impact
python -m tscfbench demo detector-downtime
python -m tscfbench make-share-package --demo-id detector-downtime
- sample download: detector-downtime share package
Takeaway: Physics users do not need to speak synthetic-control jargon to get a chart-first counterfactual workflow.
Regional employment after a minimum-wage change¶
Economics / social-science demo with one treated region and several donor regions.
- question: Did the treated region's employment index diverge after the wage-policy change?
- domain:
economics - social angle: A wage-policy chart is a more broadly legible social-science demo than canonical policy cases alone.
- intervention:
2024-08-04 - family:
panel
python -m tscfbench demo minimum-wage-employment
python -m tscfbench make-share-package --demo-id minimum-wage-employment
- sample download: minimum-wage-employment share package
Takeaway: An economics-style policy question can start with one command and end with a donor-based counterfactual chart plus share package.
Viral attention spike¶
Social-science / public-attention demo with one treated attention index and two peer-topic controls.
- question: Was the public-attention spike a real breakout, or just continuation of the existing trend?
- domain:
social_science - social angle: A viral-attention case turns social-science language into a chart people actually repost.
- intervention:
2025-10-14 - family:
impact
python -m tscfbench demo viral-attention
python -m tscfbench make-share-package --demo-id viral-attention
- sample download: viral-attention share package
Takeaway: Public-attention narratives become much easier to share when the default output is a counterfactual chart instead of a methods appendix.
Full demo catalog¶
Use the demos below when you want a fast domain-first example without starting from canonical policy studies.
City traffic intervention¶
- family:
panel - domain:
transport - question: Did the treated city diverge from its donor-pool counterfactual after the intervention?
- dataset file:
demo_city_traffic.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo city-traffic
Product launch signups¶
- family:
impact - domain:
product - question: How many extra daily signups appeared after the feature launch?
- dataset file:
demo_product_launch.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo product-launch
Heatwave and ER visits¶
- family:
impact - domain:
medicine - question: How many excess ER visits appeared during the heatwave window?
- dataset file:
demo_heatwave_health.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo heatwave-health
Electricity demand after a grid shock¶
- family:
panel - domain:
engineering - question: How much did regional demand move after the grid shock?
- dataset file:
demo_electricity_shock.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo electricity-shock
Climate shock and grid demand¶
- family:
panel - domain:
climate - question: How much extra demand hit the treated grid during the climate shock window?
- dataset file:
demo_climate_grid.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo climate-grid
Hospital surge during a respiratory outbreak¶
- family:
impact - domain:
medicine - question: How much ICU occupancy rose above counterfactual during the outbreak surge?
- dataset file:
demo_hospital_surge.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo hospital-surge
GitHub repo breakout¶
- family:
impact - domain:
open_source - question: Did the launch create a real breakout in daily stars, or was growth already happening?
- dataset file:
demo_github_stars.csv - beginner_friendly:
False - public_interest:
True
python -m tscfbench demo github-stars
Repo breakout after a launch¶
- family:
impact - domain:
open_source - question: Was the repo's breakout real, or was it already on that path?
- dataset file:
demo_repo_breakout.csv - beginner_friendly:
False - public_interest:
True
python -m tscfbench demo repo-breakout
Crypto event study¶
- family:
impact - domain:
markets - question: How much of the BTC move looks event-driven rather than co-movement with the rest of the market?
- dataset file:
demo_crypto_event.csv - beginner_friendly:
False - public_interest:
True
python -m tscfbench demo crypto-event
Detector downtime after a solar storm¶
- family:
impact - domain:
physics - question: How much detector uptime was lost after the solar storm relative to a counterfactual path?
- dataset file:
demo_detector_downtime.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo detector-downtime
Regional employment after a minimum-wage change¶
- family:
panel - domain:
economics - question: Did the treated region's employment index diverge after the wage-policy change?
- dataset file:
demo_minimum_wage_employment.csv - beginner_friendly:
True - public_interest:
False
python -m tscfbench demo minimum-wage-employment
Viral attention spike¶
- family:
impact - domain:
social_science - question: Was the public-attention spike a real breakout, or just continuation of the existing trend?
- dataset file:
demo_viral_attention.csv - beginner_friendly:
False - public_interest:
True
python -m tscfbench demo viral-attention