Strategies to follow and pitfalls to avoid from my own experience.

LinkedIn recently invited me to provide insights on several AI-powered “conversation starters”. Their goal is apparently to provide a base onto which experts can contribute their own ideas, examples, and experiences.

One article I contributed to was about maintaining and updating dashboards to reflect changing business needs. (You can view the original articles here with my comments added.)

While the original did a passable job of listing things one should do, it lacked what I consider to be even more important -- what not to do. With that in mind, I decided to share some dos and don’ts based on my experiences; these observations are mostly based on a Talent Acquisition (TA) operations dashboard creation project I led; the general principles, however, may apply to other business areas as well.

First things first: Why is it important to use dashboards in TA?

Modern talent acquisition has moved beyond the passive “post-and-pray” method. Technology has dramatically changed the way we attract and hire talent. One of the changes it brings us is that it produces reams of data which can potentially improve decision making. I say “potentially” because the raw data must first be processed and interpreted in a way that’s easy to read and relevant to decision makers.

This is where dashboards come in. My company produces large amounts of data in operational, experiential, financial and diversity spaces. Pulling data from these various sources, analyzing them, and presenting the results interactively to decision makers has become critical to our TA team’s success.

In addition, dashboards improve TA’s operational efficiency by making each person’s performance visible to the rest of the team. While some may be uncomfortable at first with this visibility, it helps identify individual areas for improvement; ultimately this produces better results for the whole team.

Finally, an automated and standardized dashboard provides a single source of truth -- automatically updated with the latest numbers -- for TA, HR and hiring managers. Hello, improved partnership!

Below I have laid out key dos and don’ts for three phases of the process: design, development, and maintenance.

One piece of advice for all three phases is to apply agile methodology whenever possible. TA dashboard creation and maintenance is an ongoing process, and the final product may differ significantly from the initial vision. In my own experience, designing and launching the dashboard created several unforeseen challenges. Using an agile approach enabled us to test as often as needed, iterate quickly, and communicate changes with stakeholders.

Design phase

DO have a vision about a robust and comprehensive final product, but be realistic about what you can actually accomplish. My original vision was to have a dashboard showing all types of TA data, but I later had to give up on certain types due to potential legal/compliance risks or the complexity of the data sources.

DO collaborate. The essential idea of the project is to provide data from multiple platforms in one place, but this process is much more than simply creating charts or lines. It’s also essential to involve the right people to ensure you are doing the right thing. For example, you need to first understand all TA platforms and how data is stored, structured, and able to be transferred. This requires collaboration with the technology owners and sources -- ATS, CRM, survey platforms, interview selection platforms, sourcing platforms, online shared folders, etc. You will also need to consult with legal and compliance to understand certain limits, especially if you are using an external analytics platform (e.g. Tableau or Power BI) rather than the tools provided within your company’s existing HR platform (e.g. Oracle or Workday).

DON’T rely only on your users’ input. Of course, user feedback is important, and your design should be user centric. User input is sometimes incomplete because it comes from individuals only focused on one part of a larger picture. It’s your responsibility as a project leader to consolidate and prioritize all these inputs. In my own experience, I found useful guidance and inspiration from companies in the tech industry, which have many early adopters of technology utilization and dashboard development. I am glad I was able to incorporate some best practices I’ve learned into my own company’s dashboard. Many elements taken from their examples would never have been included in our platform if I had relied solely on what my users said that they needed.

DON’T aim merely to create a pleasant customer experience; go above and beyond to ensure your platform can make an impact on the business. In a corporate setting like my company, a TA dashboard might be used by hundreds of users in different functions (e.g. HR, TA, and hiring managers) at different levels. It’s impossible to satisfy everyone’s needs and wants. Rather than trying to please everyone, focus instead on whether your dashboard is telling a strategic and useful story.

Development phase

DO follow an iterative process – in other words, a series of steps that you repeat to test, tweak, and improve the end product. When developing TA dashboard, there’s no better way to ensure the accuracy of the dashboard and build trust with its users.

DO choose the simple visuals whenever possible. Telling a complex story with simple visualizations is difficult but doable. My favorite authors in the data storytelling area are Nancy Duarte and Brent Dykes. I also highly recommend a company called storytelling with data, which offers many free resources on its website as well as paid classes and seminars.

DON’T simply pile in the metrics; instead organize them to tell a story and help users better understand the big picture. In my project, we intentionally structured the story telling and keep the structure and flow consistent across all tabs. My company’s dashboard is organized into three distinct sections. At the top are all of the filters. This is followed by a group of charts that explain what’s happening: hiring volume country or job type, for example. Under that is a series of figures providing more in-depth analysis that helps explain why it’s happening, such as the funnel pass through analysis. This peeling-the-onion approach is logical and has proved most popular with our users.

DON’T overcrowd the dashboard. Include only the most important and relevant metrics. If using data to monitor and improve operations is part of your job as well, then you know we track a LOT of metrics. Many of these are too specific or uncommon to put in a dashboard intended for use by a broad range of employees.

Maintenance phase

Once the dashboard is up and running, it will need continuous improvement.

DO coach your users to get the most out of the dashboard. Two things I like to do to help users maximize the dashboard:

1. Introduce benchmarks so we know where we are;

2. Explain the interconnections between metrics. One example could be that the candidate distribution analysis for a past period shows the number of applicants for one successful hire. If that number is higher than the number of current applicants in the funnel for a similar role, then it’s a good strategic indicator that more efforts need to be put on getting applicants.

DON’T underestimate the problems that may arise in your dashboard. The three most common problems I have encountered are missing data, broken data connections, and incorrect calculations. Some of them we simply didn’t expect to see, and we had to spend a lot of time fixing them. Because it’s impossible to predict all future problems, it’s critical to ensure that you have the time and resources available to fix them when they appear.

------

Disclaimer: The views expressed and opinions expressed in this article are those of the author and they do not necessarily reflect the official policy or position of any other agency, organization, employer, or company. Assumptions made in the analysis are not reflective of the position of any entity other than the author. Since we are critically-thinking human beings, these views are always subject to change, revision, and rethinking at any time. Please do not hold them in perpetuity.