Analysis of internal statistics

Internal statistics are the only place where you have all the basic data together, which is why you should consider the whole analysis of internal statistics. Here you will find not only the exact number of purchases, but also information about shipping, returns and complaints. It’s worth analysing your business’s most valuable data for several reasons. Shall we do it together?

What will you learn from the Analysis of internal statistics?

A lot! Some of the most common data we process for clients include:

• order data – comparing reality with Google Analytics data,
• number of returns – when paired well with Google Analytics, it can reveal which channel or campaign is generating the most returns,
• order delivery – how quickly I am able to deliver from order to customer (based on information from your system about order receipt and from the carrier about delivery),
• user segmentation (performing RFM analysis) – for example, by cost, order frequency, e-commerce categories, etc,
• identified users – customer categorization, long-term customer value, churn,
• comparison with competitors – comparison of product prices,
• analysis of product relationships in shopping carts.

What tools do we use?

We usually use the Keboola tool in combination with BigQuery. After a bit of magic, we get data that can then be visualised.

If we want to calculate something from this data (before visualising it), then we use Python, Keboola or compute on Google Cloud. The choice usually falls on the tool according to the complexity of the calculation, the volume of data and the performance needs.

Technical details 

Internal statistics can usually be exported from any e-commerce solution – whether it is a custom system, Shoptet or Opencart. In addition, you can also work with data from accounting systems (e.g. Pohoda, SAP). For eshop platforms, exporting is usually easier because the system is not as complex.

In addition to exports, APIs can also be used. However, this is individual according to the project – or the system or even the version of the system you pay for.

We are able to work with real time calculations, but the stumbling block comes in the data acquisition. The technology doesn’t allow us to retrieve data, for example, twice per second (it can take up to 30 seconds to evaluate a query through the API, as it downloads several MB or GB). Often, even an in-house system wouldn’t be able to handle it. That’s why we run these reports for our clients, for example 1x per day, sometimes 1x per hour.

Want to find out more analysis over internal statistics? Get in touch with us, let’s discuss everything important and get started.