Clarofy Knowledge Base: Evaluation

    Make decisions based on your analysis.


    Clarofy contains tools that allow you to make a statistical evaluation of your data, which will help make decision about the ‘population’. This is different from the exploration tools where you might make a visual judgement but may not be able to exactly quantify the measurement/slope/improvement. An evaluation would be required for decision-making, especially when resources need to be allocated.


    Here are a few things to think about when you are at the stage where you have the results of your statistical tests and are ready to recommend or valuate a potential change:

    • Back to the start: What was the hypothesis and why did we do the analysis? Generally speaking, we want to make changes within our organisation which will improve performance.
    • What is the best way to make changes? This will be different for every team and every organisation, but remember that often, you can be technically ‘correct’ but unhelpful to the business. There will be factors like time and budget that will come into your evaluation.
    • In many mining or processing contexts, fast decision-making and quick implementation are often more valuable that the ‘perfect’ solution
    • Take the time to understand the risk vs reward situation relating to the problem and solution – this should be part of your initial hypothesis-building and business understanding stage.
    • Change is constant– the ‘right’ answer for now might be different in 3, 6, or 12 months. Plant dynamics should be constantly evolving – don’t spend too long trying to find the perfect solution.

    Once you start communicating your Evaluation, you will find that you need to be able to influence with data.

    This is sometimes as important as your results themselves! Here's a few tips:

    FIRST, get to the point

    • Restate the aim/hypothesis
    • Summarise your findings, opportunity and recommendations upfront
    • Be clear about next steps

    THEN, explain your methodology

    • Go into details about the steps you took

    Make sure you document your assumptions

    • Date Periods, Filters etc


    Clarofy t-test

    Hope these tips help, and let us know what else you'd like to see in these articles!

    Analyse your data

    Before evaluation, you need to test and measure the significance of your data.

    Read Previous Article

    Deploy your data

    You've almost made it all the way! In a typical Data Science workflow, the Deploy phase is where the true value is realised.

    Read Next Article