GSP — Sales & Purchase Monitoring Dashboard
Designed a data visualisation dashboard that lets GSP's leadership track sales, purchases, margins and outstanding payments at a glance.
Client
GSP International
Industry
Trading & Distribution
Services
Data Visualisation, Dashboard UX, UI Design
The Challenge
The data existed. The answers did not.
GSP’s sales and purchase data lived in Excel exports shared over email. Leadership couldn’t see daily trends, compare buyers and suppliers, or spot margin leaks without waiting for someone to put a report together.
Why this is expensive in trading
In a trading and distribution business this is a direct commercial cost rather than an inconvenience. Margin is thin and moves constantly with buying price, selling price and payment timing.
A margin problem that surfaces in a month-end spreadsheet has already been running for weeks. Worse, every buying decision taken in the meantime was made blind — so the reporting delay does not just postpone the discovery, it extends the damage.
Outstanding payments compound the same problem. A trade that looks profitable on paper is not profitable if the money has not arrived, and a view of sales that excludes receivables flatters performance.
The subtler problem
Because every report had to be assembled on request, the questions leadership asked were shaped by what was easy to produce rather than by what mattered.
Reporting that depends on somebody’s time quietly narrows what a business looks at. Nobody asks for a breakdown they know will take two days, so the breakdown never informs a decision — and over time the organisation stops knowing what it does not know.
What needed to be solved
Live visibility instead of Excel exports circulated by email.
Sales and purchases viewable side by side, since margin only makes sense as the relationship between them.
Margin leaks and unusual movements visible early enough to act on.
Party-wise and product-wise comparison, to see which buyers and suppliers are actually moving.
Outstanding payments surfaced alongside trade performance rather than separately.
Drill-down from a headline number to the underlying transaction without exporting anything.
A view readable in under a minute by someone between meetings.
How we framed the brief
The goal was to turn raw data into one screen that answers the business questions management asks every week — without anyone having to build a report first.
What We Did
We designed the dashboard around decisions, not around data.
Our approach
Most dashboards fail because they are built from the database outward — every available field becomes a chart, and the result is comprehensive and unreadable.
We worked in the opposite direction: starting from the decisions leadership needed to make, and showing only what served them. Everything else was deliberately left out.
01 — Defining the questions
We sat with management to list the questions they ask most often — how are sales trending, which parties are buying more, where are margins slipping, what payments are pending.
Writing those down first did two things. It set the scope of the dashboard, and it ruled out everything else — which is the harder and more valuable half of the exercise, because every chart that earns its place makes the others easier to read.
02 — KPI hierarchy
Those questions became a clear hierarchy: headline KPIs at the top, trends in the middle, and detailed breakdowns below — so the most important numbers are always seen first.
The hierarchy is what makes a dashboard usable in thirty seconds. Someone opening it between meetings should be able to tell whether anything needs their attention without reading a single chart in detail, and only then go deeper if something looks wrong.
Designing for that glance is a different discipline from designing for analysis, and it is the one leadership tools actually need.
03 — Visualisation design
Sales versus purchase trend charts, party-wise and product-wise breakdowns, margin tracking and outstanding payment views. Every chart type was chosen for readability, with consistent colours for sales, purchases and alerts.
Pairing sales and purchase on the same axis was the central decision. Margin pressure in trading is almost always visible as a divergence between the two long before it shows up as a margin number — so putting them together converts a lagging metric into an early warning.
Colour was used as information rather than decoration. The same colour means the same thing on every chart, so the dashboard can be read at a glance without consulting a legend or relearning it each visit.
Outstanding payments were placed alongside performance rather than in a separate view, because a sales figure without its receivables tells leadership something misleadingly positive.
04 — Drill-downs & filters
Filters by date, region, party and product let users move from a high-level view to the exact transaction in a few clicks.
This is what closes the loop. Spotting an anomaly is only useful if the next question — which party, which product, which day — can be answered immediately, in the same screen, instead of becoming another request for a report and another two-day wait.
Services delivered
Data visualisation
Dashboard UX
UI design
The Outcome
Business health, readable in minutes.
Leadership now opens a single dashboard to understand how the business is performing today. Unusual movements stand out immediately, and buying decisions are made from live numbers rather than last week’s spreadsheet.
Reporting stopped being a task somebody had to complete before a question could be answered.
1 screen
Sales, purchase and margins together in one view instead of separate reports assembled on request.
Live data instead of Excel exports
Current numbers available on demand, replacing spreadsheets circulated by email after the fact.
Drill-down by party, product and date
From a headline KPI to an individual transaction in a few clicks, so the follow-up question never needs a new report.
Sales
Trends by party, product and region.
Purchase
Supplier-wise purchase tracking side by side with sales, so divergence is visible early.
Margins
Margin and outstanding payment views that flag issues before month-end.
What this unlocked
Management can ask a new question and answer it themselves — so what the business watches is no longer limited by who has time to build the report.
For sales, purchase & margins
Data instead of Excel exports
By party, product & date
