Project Cost Estimate. The Most Reliable Way to Evaluate a Project

# Strategy
Alex Borodin
September 25, 2019

Before making a deal, everyone prefers to know all the details of future cooperation. One of the first questions a client asks the developing company is a project cost estimate and the time necessary for completion one or the other job. The issue can be tricky since it’s not easy to evaluate work instantly.

Now we’ll try to describe the time estimate process in our company.

What Can Slow Down a Project 

To create a more exact deadline, we use Jira software, our internal statistics, and probability theory taking into account the most common risks that may arise. The principal factors that usually influence a deadline include:

  • An error in the original estimate.

  • Growth of the requirements.

  • An effect of productivity variance and communication expenses.

The list is not exhaustive.

Our Approach to the Estimation

We start with the Jira tasks creation for all the issues. After that, we download them in the form of a table and put the anticipated time frame in front of each part. That’s the initial illustration of the project we are going to work with. Then the data goes into a project simulator that runs the project 500 times. It allows us to calculate how much time we should add to the primary estimate assessing all the risks by the law of probability.

After the calculation of various possible scenarios, we get a table of data presented below. 

It shows the probability of completing the project in a set amount of time.

Let’s see the diagram based on the simulation results.

The vertical axis demonstrates the number of instances run by the simulator. The horizontal axis shows the number of hours needed for completing the project, given that some or all of the adverse events occurred during 500 runs of the simulation. The blue area represents the finished project.

To better understand the subject illustrated in the diagram, let’s draw an imaginary vertical line that divides the blue area into two parts with an area equal to each other. On the x-axis, we’ll get the number of hours required to complete the project with a 50% probability. However, 50% is not enough. To be confident that we’ll get the job done by a specific time, we need to have at least 95%. From the figure, we can see that this probability corresponds with 197.44 marks on the scale. So, this is the number of hours we give to the client.

Conclusion

As you can see, our approach to establishing the timeframe for completing each project is scientifically well-founded and reliable. We don’t want to give our customers false expectations saying that we can do the job faster. In the long run, it would only disappoint them and destroy trust. There are different situations, but it’s always better not to sacrifice quality for speed. Everything is up to you. Whatever the decision is, we will always be happy to provide good services in a reasonable amount of time.

Author
Alex BorodinChief Operating Officer

See More Articles

At VT Labs we are a full-cycle software development company specialising in eCommerce. Whether you’re looking to develop a product from scratch, platform migration, revamp your existing product, design an app, or do UX testing, we’re your one-stop shop.