Area 15 · Data, AI and consulting

Data and analytics

Here is how a sales summary gets made in plenty of companies: somebody exports figures from Comarch ERP Optima, pastes in an Allegro report, recalculates commission by hand and emails a spreadsheet to the board on Monday morning. That eats several days a month and produces several versions of the truth. Our job is to clean the underlying records, pull every source into one database and give you reports that update without anyone touching them.

3
services in this area
1 source
of figures for every department
Power BI
or Looker Studio
EU
the region we choose for your data

Signs it is time to act

Projects often start with “we need a dashboard”. A short conversation usually shows the real trouble lies deeper, in one of the places described here.

Sales and finance quote different revenue

We write shared definitions down. One team counts from the order date, another from the invoice date, a third deducts Allegro returns.

One customer, three spellings

We clean the records and add rules. “Kowalski sp. j.”, “KOWALSKI SPÓŁKA JAWNA” and an entry with no tax ID count as three separate firms in any report.

Month-end takes a week

We build a warehouse that loads itself. Excel exports and stitching sheets together swallow the financial controller’s time.

Stock levels are always a day old

We schedule automatic refreshes. When an Allegro listing sells faster than the report arrives, restocking decisions are made blind.

The warehouse team retypes supplier paperwork

We introduce automatic reading. Someone copies every line of supplier delivery notes and German or Chinese invoices into the system manually.

A dashboard does not fix data; it only makes it look better. A hand-built spreadsheet has one virtue: a person looks at every figure and occasionally spots nonsense. A colourful chart looks professional, so nobody questions it, even when duplicate accounts noticeably inflate the customer count. Data quality is therefore checked right at the start, before the first chart exists.

From Excel to a report that refreshes itself

Phase one covers a single area, usually sales or stock, and a handful of key indicators. Other departments join once the first report is part of daily routine. Everything happens remotely through secure access to your systems.

01

Agreeing the numbers

Together with finance and sales we settle how revenue, margin and marketplace returns are calculated. Skip this and every report gets challenged at its first meeting.

02

System inventory

We list the sources: ERP, shop, Allegro, spreadsheets. Each gets an owner, an access method and a note of fields holding personal data for your GDPR records.

03

Trial report

The first report is compared against the old Excel file. Every gap gets explained before the automated version reaches the board.

04

Expansion

New departments and metrics come on board. With the data model and loading already running, every later report is quicker and cheaper to produce.

Questions and answers

If you run on Microsoft 365 and your data comes from an ERP and SQL Server, Power BI usually wins: it models data well and makes permissions easy. Looker Studio plugs straight into Google Ads, GA4 and Sheets, so it suits marketing teams and online shops. You can also use both, with one shared warehouse underneath.

It should not. Extraction runs overnight or in small batches, ideally via an API or a database replica, never as heavy queries during office hours. That is exactly why reports sit on a separate warehouse rather than directly on the Comarch ERP or Symfonia database your whole sales team is using at the same moment.

Inside your Azure or Google Cloud subscription in an EU region or, if you prefer, on a server in your own data centre. A processing agreement is in place and permissions follow job roles: the sales director sees the whole country, a regional rep only their territory and clients. Every time a report is opened, the log records it.

Where records are in decent shape, the first report arrives much sooner, definitions included. With thousands of duplicates, clean-up becomes the bulk of the job. We give a date only after inspecting the database, never at the first meeting.

Let us get your numbers in order

Tell us which reports are still assembled by hand and where the figures disagree. We will look at the source systems and suggest a starting point.

Hours
Mon-Fri 8:00-18:00 CET, reply within one working day
Meetings
Online via Teams or Google Meet

We set strictly necessary cookies only: they keep the site running and remember the city you chose. Nothing here is used for advertising or tracking. More in our privacy policy.