Data processing and data transformation

Data processing and data transformation are some of the activities where we extract data from different systems, merge it, clean it and use it in subsequent data analyses. From different piles, we usually make one big but usable pile of data. After the data analyses, the visualizations take a turn.

Thanks to data processing and data transformation, we can combine data from, for example, Google Analytics, Google Ads, Sklik, Facebook, internal systems or Google Spreadsheets, and calculate or predict the cost of customer acquisition, cost of one single shipment, or performance of a specific brand in an e-shop across different marketing channels or devices.

In addition, we are able to import data back into tools – for example, purchase price information into Google Ads or an emailing tool (reverse ETL).

Practical examples

There are no limits when it comes to the processing and transformation of the collected data. Let’s take a look at a few practical examples that we are dealing with for our clients.

Common realizations

  • finding out what the real cost of 1 purchase is (credits, packaging, shipping) and how much of the order remains
  • what the ROI is – return on investment; how much it costs to acquire a person and what the retention needed to generate a profit is
  • Evaluating what brands work across the e-shop (while taking into account different devices, marketing channels and laboriousness of shipping)
  • predicting sales trends based on available data
  • merging data from different systems, countries and currencies (Google Ads in CZK, Facebook in Euros, external costs in South African Rand) across multiple projects = cost per item calculations

More advanced implementation

  • data processing and real-time margin calculation – after conversion is made, a real-time margin is calculated. It is imported into advertising systems, so the marketing agency is able to better optimize PPC campaigns according to the product performance
  • analysis of the performance of individual brands sold – costs per brand are budgeted when we do not advertise directly on the brand, but e.g. on the general brand of the client 
  • optimization of Google Analytics loading speed – with large projects it takes several tens of seconds or minutes to load a specific report – we download the data, transform it into relevant reports, and then we perform the visualization (it is similar to the Google Analytics report, but the loading is much faster)
  • evaluation of offline activities (billboards, flyers) according to changes in classic behaviour (very closely linked to data analysis)
  • geolocation – data collection for more precise determination of customer location, specific distances are then calculated within the analyses (e.g. for transport)
  • prediction of revenue development based on available data

Safety limits and requirements

Working with customer data requires maximum security. That’s why we work in Google Cloud, which offers an immense number of security certificates.

We also urge our clients to use anonymous IDs instead of email addresses and real user data. For example, Shoptet does not support this anonymous ID solution – we usually convert emails to anonymous user IDs ourselves.

Our experience

Our data processing and data transformation clients include Outdoor Concept (Rock Point, Hannah) and Parcelsport.com etc.

More detailed information

Is Excel no longer enough for you but you still need to work with data? Try data merging and its transformation.

Scroll to Top