← tanbedawy.workbook
Case Study · MENA Ride-Hailing Leader · Analytics Product 06 / 07

Real-Time Operations

Intraday marketplace visibility for Egypt and Pakistan: 15-minute-granularity views on live tables refreshing roughly every five minutes, with day-over-day comparison overlays from sessions all the way down to utilization.

Near-Real-Time BILive PipelinesTableau SQLOps Analytics2 Countries
product scorecard — structural factsproduction
15min
finest time granularity in the views
~5min
refresh cycle of the live source tables
14+
metrics in the grid, sessions to utilization
2
countries — Egypt & Pakistan builds

Note on numbers: no figure from inside a client dashboard is presented here as an achievement. Published screenshots have absolute figures blurred.

01The Business Problem

Ops teams managed an intraday business with yesterday's report. A supply gap at 2pm, a demand spike after rain, a promo misfiring at launch — by the time the daily dashboard showed it, the moment to react was gone. The marketplace needed operational visibility: what is happening right now, and is it normal for this hour?

The goal: a live view where "is right now normal?" is answered by overlaying today's curve on comparable days — down to 15-minute slices.

02My Role

Sole analyst on the product: the SQL against live tables, the 15-minute/hourly view architecture, the day-over-day comparison design, and the in-dashboard timezone documentation that kept a GMT pipeline readable for local ops teams.

03What I Built

Real-Time Egypt (V2)

A grid of metric trends — sessions, bookings, trips, missed bookings, C/R, fulfillment, active users, promo and peak trips, peak factor — plus a full supply block: available and active captains, available/active/busy hours, cash-collection hours, and a full-width utilization view. Three comparison days overlaid per chart.

Real-Time Pakistan

The same architecture deployed for a second market, with four comparison days overlaid and its own supply section — proving the design portable across markets rather than a one-off build.

15-Min / Hourly Toggle

One switch moves every chart between 15-minute slices (for incident forensics) and hourly aggregation (for shape reading) — same layout, same metrics, two operating altitudes.

Timezone & Partial-Day Handling

GMT-to-local conversion documented inside the dashboard, and today's partial curve simply ends at "now" against complete comparison days — no misleading extrapolation.

04Signature Build Details

Comparison overlays as the baseline: every chart draws today against same-weekday reference days — anomaly detection by eye, calibrated to the hour.
Supply depth most live dashboards skip: available vs active captains, and available vs active vs busy vs cash-collection hours, each defined in-view — ending ambiguity about what "available" means.
Utilization as the bottom line: the full-width closing chart divides busy time by available time — one line telling ops whether supply and demand are actually meeting.
Built on a live pipeline: source tables refresh roughly every five minutes; the views are designed to be re-read all day, not exported once.

05KPIs Defined & Governed

SessionsBookingsFulfillment Rate C/RMissed BookingsPeak Factor Available / Active HoursUtilization %

06Decisions Enabled

Ops responded to supply gaps and demand spikes within the hour, not the next day.
Launch-day behavior of promos and pricing changes was visible in 15-minute slices.
The same product served two markets, giving regional ops a shared operating picture.
SCREENSHOT SLOT
Real-Time Egypt — 15-minute view
(blur all absolute figures)assets/rt-egypt-15min.png

07Skills Demonstrated

Near-real-time pipelinesIntraday analytics design Comparison-overlay UXTimezone handling Partial-data handlingMulti-market deployment Advanced SQL
Mohamed Eltanbedawy · Business & Data Analyst — BI & Analytics
mohamed.tanbedawy@gmail.com · linkedin.com/in/tanbedawy