- Expenditure Estimation


When you wish to determine what future expenses will be for specific things, you conduct a expenditure estimation analysis. The analysis takes place by making an AI / ML model that processes and learns on the basis of your consumption data and other relevant information, and which can then later be used to give an accurate forecast of how many funds you need to spend on a specific thing.

You can thus get certainty about your budgets and how they will evolve over time, so you know exactly how your expenses will develop in the future.

Requirements for an expenditure estimation

We need to have free access to your data in order to make a expenditure estimation analysis model as this is the basis for the model. Depending on how the data is structured, we may have to perform substantial data augmentation, as the data must be structured in such a way that the model can understand what has occurred at certain times. When the data is not of a proper quality, we are able to assist you in collecting the data properly and ensure that a time model can be applied.

In addition to having a clear idea of what one would like to have predicted, each model is built to predict one thing only. It is possible to set up several models if you want to predict several things at once. Furthermore, this will reduce the cost per model, since the infrastructure only needs to be set up once for each model.

Technical possibilities regarding expenditure estimation

An AI/ML model can be built that can predict future consumption with a expenditure estimation analysis.

A model is usually set up in the cloud, and you talk to it via a VPN. Local data processing can be used to anonymize the data that is sent back and forth. The model can also be set up locally with you if the data is especially sensitive, so that the data does not leave your premises.

The process regarding expenditure estimation

Q-analytics process when we do a expenditure estimation analysis is

  • Initial meeting
  • Workshop
  • Data is accessed
  • Model generation
  • Result / handover

Initial meeting

As part of the introductory meeting, we will discuss the considerations you have and the values you would like to be predicted. We will be providing input and advice throughout the process, allowing us to understand your needs and narrow down the solution(‘s) that will be most suitable for you. In addition to discussing how the model can be integrated into your systems, we will also discuss how this must be done from the very beginning.


During the workshop, we focus on the technical aspects of model generation and its integration with your systems. We have already determined scope and are now focusing on how to implement it. Furthermore, we will be presented with your data so that we can determine what exactly needs to be addressed. So that any questions we may have about this can be clarified.

Data is accessed

We get access to the data here so that we can build the model for you. We normally handle data processing in our office, but if your data is so sensitive that it cannot leave your premises, we can easily agree to generate the models with you.

Model generation

Based on the workshop instructions, we generate the model. Additionally, we produce graphs and reports that explain how the model performs so that you know exactly what you're getting.

Result / handover

We hand over the model to you and help you integrate it into your systems as well as present you with the results.