New automatic calculation of project phases and algorithm improvements

Features & Updates

For the last month we have been hard at work with a significant update to our Automatic Learning Algorithm (ALA) as well as automatic projet phase calculations. Both of these updates are relevant for all users no matter whether you are using agile or a more traditional project model. The updates have gone live in the solution and can be used immediately.

Improvements to the Automatic Learning Algorithm (ALA)

The update ensures that the algorithm is now tightly integrated to the underlying structure of This means that the algorithm is now much more clever when it comes to forecasting new projects. Furthermore, we have now given the option for users to choose whether to take advantage of the algorithm or not. This can be disabled individually on project level.

Of course we always recommend that the algorithm is turned on, as that gives the best basis for measuring and increasing performance on future projects. The new updates to the algorithm also relies on the power of our complete tagging system, meaning that the more tags are used throughout, the faster the algorithm will learn and adapt to how any given company operates.

User Interface (UI) updates

Additionally, UI updates makes it much more clear to the end user what is predicting and how to override values of the predictions if that is wanted. Instead of starting from our industry standard initial parameters, it is now also possible to enter company specific parameters for the algorithm to initiate from. Using this will enable a faster learning curve for the algorithm since it is more appropriately suited to the specific company.

As an example:

Task 1 is estimated to take 10 hours but depending on the historic performance of the specific company, project type, client, tasks, actual performance previously and other historic metrics, knows that it is much more likely to take 15 hours to complete the task and will thus suggest this to the user. The same goes for estimates done in points e.g. story or function points.

Thus aids the user intuitively and in a much more accurate way than what could have been done manually before. This is all thanks to the built in intelligence and automatic data mining of the solution. This is not achievable using standard tools such as MS Excel.

New automatic calculation of project phases

Easily create automatic SoW's

Our other new mainline feature is automatic calculation of project phases using 3-point estimation. This enables rapid simulation and generation of estimates based on historical performance. Using estimates from only 1 or more project phases enables to predict the size and effort of completing the remaining phases.

As an example:

The project model of a company goes through 4 phases called requirements, development, test and deployment. By only estimating e.g. development and test, can simulate and calculate the effort for the two remainders (requirements and deployment) with a high degree of accuracy.

It is an extremely flexible and fast way of creating estimates/SoW's that are reliable, as the algorithm constantly monitors actual effort from historic data for each phase that has been defined in the system and applies them. Again, this functionality can apply to both agile and waterfall styled projects.

We hope you enjoy the new features. Our development roadmap is also constantly updated so keep an eye out for new upcoming functionality.

Niels Frederiksen

My name is Niels Frederiksen and I’m co-founder and lead developer in Forecast. 

Related Posts