What it means
Almost every report a business already produces is descriptive analytics. Monthly revenue by region, headcount by department, average order value, churn last quarter: each takes a large pile of historical records and compresses it into something a human can hold in mind.
The work is summarising and comparing rather than predicting. It matters because decisions rest on a shared picture of the recent past.
If the sales director, the finance team and the board each carry a different idea of what last quarter looked like, the argument about next quarter goes nowhere. A clean descriptive layer settles the facts so the discussion can move on to causes and choices.
In practice, descriptive analytics is delivered through dashboards, standard management reports and simple ad hoc queries. The typical building blocks are counts, sums, averages, percentage changes, shares of total and comparisons against the same period last year.
Good practice is to pair every figure with a benchmark, because a number such as $460,000 in monthly revenue means nothing until you know last month was $400,000. The main limitation is that description alone can mislead.
A dashboard can show that conversion fell 12% without saying whether the cause was a pricing change, a broken checkout page or a seasonal pattern that repeats every year. That is the boundary where diagnostic analytics, which digs into causes, and predictive analytics, which forecasts forward, take over.
The common variant worth recognising is real-time or operational descriptive reporting. Instead of a monthly pack, the same summarising logic runs continuously so a warehouse manager can see today's pick rate or a support lead can see the current queue.
The maths is identical; only the refresh frequency and the audience change.
In practice
Real-world examples.
Example
A subscription fitness app publishes a weekly dashboard showing sign-ups, cancellations and net member growth by acquisition channel. The team sees that paid social delivered 1,400 sign-ups but 620 cancellations in the same month. No cause is identified yet, but the description is enough to trigger a deeper review.
Example
A manufacturing plant reports scrap rate by production line each shift. Line three shows a scrap rate of 4.2% against a plant average of 1.8%, which becomes the first agenda item at the morning stand-up. The report does not explain the gap; it simply makes it impossible to ignore.
Example
A professional services firm summarises billable utilisation by consultant and grade for the previous month. Partners can see immediately that senior consultants averaged 78% while juniors averaged 52%, prompting a change to how work is allocated.
Formula
Calculation
Percentage change = (Current period value - Prior period value) / Prior period value x 100. Share of total = (Segment value / Total value) x 100.
A coffee chain records revenue of $1,200,000 in the first quarter and $1,380,000 in the second quarter. The change is $1,380,000 - $1,200,000 = $180,000. The percentage change is $180,000 / $1,200,000 x 100 = 15%. Average monthly revenue in the second quarter is $1,380,000 / 3 = $460,000. If the in-store cafe channel contributed $552,000 of the quarter, its share of total is $552,000 / $1,380,000 x 100 = 40%, leaving 60% from wholesale and online.Case study
Seen in the real world.
Brightline Bakeries is an illustrative, fictional chain of twelve retail bakeries that ran for years on a single monthly profit and loss statement. Store managers had no view of their own performance until six weeks after month end, by which point any problem had already repeated itself several times.
The finance manager built a simple descriptive reporting pack: daily sales by store, average transaction value, waste as a share of production, and each store's rank against the group. Nothing in the pack attempted to explain anything. It only described what had happened, refreshed every morning.
Within a quarter the pack had changed behaviour without a single new policy. Two stores discovered their waste ratio was double the group average and cut afternoon baking volumes, while another realised its average transaction value was the highest in the group and shared its upselling script. This fictional example shows the quiet value of descriptive analytics: making the recent past visible often produces action on its own.
Watch out
Common mistakes.
- Treating a descriptive dashboard as an explanation, when it only shows what happened and never establishes why.
- Reporting a figure with no comparison, so nobody can tell whether $460,000 in monthly revenue is good, bad or ordinary.
- Building dozens of metrics because the data exists, which buries the handful of numbers that actually drive decisions.
Questions
People also ask.
Is descriptive analytics the same as business intelligence?
They overlap heavily; business intelligence is the broader set of tools and processes, while descriptive analytics is the specific job of summarising historical data.
Do you need a data scientist for descriptive analytics?
Usually not, because the underlying maths is counts, sums, averages and percentage changes that a capable analyst can produce with a spreadsheet or a reporting tool.
What comes after descriptive analytics?
Diagnostic analytics investigates causes, predictive analytics estimates what is likely to happen next, and prescriptive analytics recommends an action.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%