HYDRADATA · Power BI · E-Commerce

E-Commerce Growth Dashboard

Connect traffic, conversion, commercial performance and refund exposure in one five-page review sequence.

Portfolio demo | Public sample data | Not client results

Démonstration de portfolio | Données d’exemple publiques | Aucun résultat client

Business question

Where should the next investigation begin?

Management, marketing, commercial and product teams often review different slices of the same business. This portfolio connects those questions without merging session, order, item and refund definitions into one misleading total.

One deliberate sequence

Executive Overview
Marketing & Conversion
Sales Performance
Products & Refund Analysis
Insights & Management Actions

Move from a headline observation to a specialist breakdown, then to a justified human follow-up.

Watch the case

Watch the walkthrough

Follow the business questions from traffic and sales to refunds and the next investigation.

English narration · English and French subtitles · Static screenshot walkthrough

Loading the player connects to YouTube. Playback starts only when you press play.

Five-page walkthrough

A question for every view.

Default exhibition state: 3 April 2014 to 1 April 2015 inclusive, with source, device and product unrestricted. Currency is USD. These are static report captures, not an embedded live report.

Executive Overview

For: Management and functional leads

Question: Are traffic, orders, revenue efficiency and refund exposure moving together, and where should the next review begin?

Executive Overview with revenue, sessions, orders, conversion, revenue per session and refund rate; monthly trends and source and product breakdowns. — The overview brings volume, conversion and revenue efficiency together. Source contribution and product refund exposure provide two routes into the detail.

Possible follow-up: Choose the marketing, commercial or product question that needs a closer review before proposing an action.

Marketing & Conversion

For: Growth and marketing teams

Question: Which source and device segments combine traffic volume with stronger session conversion?

Marketing and Conversion with a session funnel, source volume and efficiency charts, device conversion and a source-by-device matrix. — Read source volume alongside conversion and revenue per session. A high-volume source is not automatically the most efficient source. The funnel shows where session progression narrows.

Possible follow-up: Check segment size, device mix and landing-page context before designing a controlled test. Source association alone cannot establish a campaign effect.

Sales Performance

For: Commercial and sales operations teams

Question: Which products and periods contribute revenue and gross profit, and how does the mix affect margin?

Sales Performance with revenue, orders, average order value, product revenue, gross profit, margin, monthly trends and product rankings. — Order revenue and average order value describe transactions. Product revenue, gross profit and margin describe purchased items. Their calculation grains are not interchangeable.

Possible follow-up: Review material product and period differences at the correct grain. Gross profit is not net profit and does not include every operating cost.

Products & Refund Analysis

For: Product and e-commerce operations teams

Question: Which products combine meaningful sales contribution with elevated refund exposure?

Products and Refund Analysis with product revenue and margin, refund amount and rate by product, and an illustrative product-risk matrix. — Compare refund amount with refund rate and product contribution. A large refund amount can reflect sales volume; the rate adds another perspective. Risk labels use illustrative thresholds, not benchmarks.

Possible follow-up: Review refund reasons, fulfilment and support evidence before asserting a cause. These additional sources are not present in the sample dashboard.

Insights & Management Actions

For: Management and functional leads

Question: Which observations deserve discussion, and what follow-up is justified by the available sample data?

Insights and Management Actions with growth, conversion, funnel, source and product summaries, suggested human review actions and limitations. — The final page restates measures from the preceding views in plain language and separates observations from possible follow-up. Source-linked revenue remains an association, not causal attribution.

Possible follow-up: Assign a human owner to verify context. This is an interpretation layer, not a forecast or an automated decision system.

Definitions before conclusions

Keep the grain and the comparison visible.

Conversion Rate: Orders divided by sessions in the selected context, not customer or unique-visitor conversion.
Revenue per Session: Order-level revenue divided by sessions. Compare efficiency separately from total revenue.
Average Order Value: Order-level revenue divided by orders. Item rows must not duplicate order totals.
Product Revenue / Gross Profit: Item-level revenue and revenue less item cost. Gross margin is gross profit divided by product revenue.
Refund Rate: Refund records divided by the eligible purchased-item base. The denominator is protected against being reduced to refunded items only.
Previous Period: The immediately preceding date window of equal length, retaining other filter context. The selected dates are part of the interpretation.

Direction is not a business verdict

The neutral-grey arrow reports the actual movement. For refunds, an upward arrow means an increase; it is not inverted to suggest improvement. Rate changes are percentage points, while refund-amount comparisons are percentage changes.

Read the separate status

Favourable and Unfavourable apply the illustrative rule that lower refunds are preferable. Neutral means equal values. Missing comparisons show N/A and No comparison; invalid inputs show Check data. These labels do not prove an overall business improvement.

Public sample data

What the evidence can support.

The source is the Maven Fuzzy Factory sample dataset for a fictional online retailer, available through Maven Analytics learning materials. Sessions, orders, purchased items and refunds cover a historical sample beginning in March 2012; refund records extend to April 2015. The displayed selection is narrower than the full model.

Maven Analytics public sample datasets

Analysis limits

Source-linked revenue is association, not multi-touch attribution or incrementality.
No claim of client results, deployment, adoption, revenue uplift, ROI or cost savings.
No prediction, customer lifetime value or validated operating benchmark.
Raw pageview events are outside this report model; the funnel uses a session-level aggregate.
Refund reasons, fulfilment and support evidence are needed for root-cause investigation.
Risk thresholds are example rules. Suggested actions require human judgement and additional evidence.

The methods behind the views.

Explore the same Power BI, metric-grain and validation topics in the existing insight library.

Why AI dashboard mockups need human validation Power BI metric grain Define refund rate without misleading the business

Discuss your reporting questions

Need a clearer connection between traffic, sales and operations?

Start with your business questions, metric definitions and review workflow.

Contact HYDRADATA