Moving the needle — How I set up a Dashboard for C-level Platform Strategic decisions (part 1 — how it all started)

A brief introduction

I joined Basecone coming from a PMO position at a previous company and a much bigger company. My primary role as a Senior Program Manager is to provide alignment, visibility, and efficiency using whatever tools and methods I may find fit to a group that has grown across 2020 and expects to grow much more in 2021.

In summary, I will divide my experience in building a C-level Dashboard into four parts:

1. The ask and the strategy to set up things in motion (how I joined Basecone and, in month #1, I got multiple requests from C-level to build up a strategic dashboard and how I got processes undergoing to make it happen);
2. What work was done and how (the ins-and-outs of using JIRA + Excel + PowerBI integrations to produce meaningful graphics);
3. The Review (how the process of reviewing the dashboard works, and with who);
4. What I would do differently (lessons learned from almost a year of working through the project);

How it all started

After my first month at Basecone, I got into my second one-hour conversation with Basecone’s CEO, Nico.

During our conversation, Nico asked me if I could produce a Dashboard with data that would provide the Senior Management team with clear insights on how our Product teams were spending their time (Product Vision, Customer Requests, and Platform improvements) so he could if we were spending time in what really mattered, in each month/quarter, depending on the company strategy. At the same time, the CTO, Lorenzo, told me he wanted to have a place where he could see the Platform overall status in terms of the number of Releases, their type (Standard, hot-fixes), and the Quality of our Software in Production so he could also steer our Agile teams, into the right direction, together with the Head of Product (Wouter).

I knew, beforehand, that would be a massive endeavor, across many months, given that Basecone:

1. Was still at an early scale-up stage, which means streamlined processes were far on the horizon;
2. Had no agreed communication flow making it hard to process multiple needs from multiple stakeholders (everyone would ask everything to everyone);
3. These type of initiatives take months to get done correctly;

I was also finding it challenging since the experience I gathered over the last three jobs was now being challenged by this single and simple URL where all the data would be put together to produce strategic decisions that would affect the way we approach the future of Basecone.

How I approached the problem

I used a simple continuous evaluation model framework for all the below points:

1. GATHER THE DATA: Interview your stakeholders and try to find what they are looking for

If you’ve worked in a Strategic PMO role before, you probably have a good idea of what your stakeholders are looking for and what metrics make sense to surface. Each company is very different, and the same goes for the Senior Leadership group, the Departments, your Agile teams, etc. It would help if you ALWAYS tried to understand where others are struggling and what they want to see so you can adapt your project to suit their needs (and challenge, when needed).

2. ANALYSE THE DATA: Understand the big topics you want to surface

After my initial conversations, it was clear to me what questions we were trying to answer:

1. How often are we releasing (Release Cadence), how faster are we in Go-To-Market, and what are the details of each one of those Releases (dailies, bi-weeklies, standard, hot-fixes)?
2. Are we building good and stable software for our customers(Software Quality)?
3. How are we spending our Agile team’s time (Execution Effort breakdown by type)?
4. What type of features are we developing, and who’s requesting those (Department Development);
5. Are we distributing our resources into the correct places(Headcount Distribution across departments, teams, expertise, and guilds);

3. APPLY CHANGES: Make sure you have correct data and the right processes in place for future analysis

For you to have data that you can trust, you need to have four things:

1. Understand what data you will need to make the Dashboard live (make sure you break down the group into some release cadence for better stakeholder management and to evaluate progress as you go — you don’t want to spend your time in things that don’t matter);
2. Work together with the teams to find a way to produce the data you need(that will highly depend on what tools your company uses);
3. Create weekly, and monthly quality checks to make sure everyone is filling in the data in the same way (that will grant you data quality);
4. Make sure you define an SSOT (Single Source of Truth), so everyone knows where they need to go if they need to search for data and that only that data is valid(that will deprecate those 10123 Excel spreadsheets everyone maintains that only generates waste, confusion, and miss-alignment across the company);

For you to have the correct processes in place, you need to make three things:

1. Do not act as you own everything and do not come with an “I know what needs to be done” type of approach (this will get you nowhere, and you will face much greater resistance if you don’t listen to people and match their needs);
2. Gather the people who want to participate in the process definition and brainstorm for a solution (this will grant you more buy-in from individuals and teams, and, as we all know, it will probably get you a better result and something more suitable for your company — no company is the same);
3. Make sure you have only had enough processes and enough changes in place to get what you need (do not create processes for the sake of processes. People will not buy it, and you will eventually create more friction for adoption in the long run);

4. MONITOR: Make sure you continuously look at the data both from a quality and discussion point of view

As companies grow, you will find yourself in a position where the data is all messed up. That’s normal, even if you have enough control over the data itself. To have some control over your data quality, make sure you run weekly, bi-weekly, or monthly analyses to make sure you check for data inconsistencies and outliers. Any research you want to take and decide the next steps is only as good as your data quality is correct; otherwise, you could be looking at something that is not real.

ALWAYS run analysis periodically on the processes in place. Companies are always evolving structures, so the process you created 12 months ago may not be adjusted to the current reality, and you may be sticking to a process that does not generate any real value.

I hope you had fun reading the above. In my next medium post, I will show you the tools and methods used to get to the dashboard.