Analytics

In order to ensure we are solving the right problem, we must define it. This means determining what “done” looks like, what is an acceptable timeframe, what is in and out of scope, and so forth - and actually writing it down. This helps ensure everyone is on the same page and triggers us to come back together if the project starts to drift from the agreed upon parameters.

What would have to be true to achieve our goal? What are the component parts? For example, if our project goal is to maximize donor retention, we could break this down by fund, by years of consecutive giving, by giving method. This helps me understand how the organization thinks about their problem and what levers may be available to impact it.

A fleshed-out goal breakdown will result in more opportunities than we can realistically tackle at once. Furthermore, some may be outside of our control or unlikely to yield much fruit. So, we eliminate those and focus on the areas the initial data suggest will be the most impactful.

Once the priorities are established, we can develop a more detailed working timeline, including roles and responsibilities for all parties involved, and start working the plan.

Now for the good stuff. We start to gather data, do summary analysis, and digging deeper as the situation warrants. This may involve data science and machine learning methods such as regression analysis, time series, clustering, decision trees, text analytics, and more.

We review the results of our analysis and translate them into business insights and potential recommendations.

Throughout the project, we will have regular client communications updating on project status and findings. At project end, I will package the findings and recommendations into a document that the organization can reference and share, along with any technical collateral such as reports, spreadsheets, and decision models.