65: It takes a village to build a dashboard

What’s up everyone, today we’re taking a dive into the world of dashboard building.

Startups may not always have the luxury of having a dedicated data analyst on staff, which means marketers may need to get more hands-on with data. 

Yeah I haven’t had the data analyst luxury in my career very often! In episode 38, we discussed marketing reporting and how you can use key reports to help highlight impact and find new opportunities. But we’re not talking about reports here right?

That’s right, dashboards aren’t reports. They are living breathing snapshots of key areas you want to keep an eye on in your business.

Yeah I think a lot of people don’t make that distinction and just assume reports = dashboards = chart. Where should marketers be starting? With charts?

Scatter plots, bar charts, pie charts, maps, funnels, box plots… There’s a bunch of different chart types and visualizations at your disposal when you’re designing your dashboard, but this isn’t where you should start.

Here’s today’s main takeaway: When designing a dashboard, it’s important to focus on the decisions you want to make, rather than just the metrics you want to track. Before building your dashboard, consider your audience and bring together the right people to answer key questions. This will help you create a prototype of your first version.

Dashboard projects are close to both of our hearts. Both having worked for Klipfolio (a dashboard SaaS for startups and SMBs), we’ve spent a fair amount of time researching and writing about the internal dashboard building process.

There’s obviously a critical collaboration piece to this that would be an initial starting point for anyone taking on a dashboard project. 

Yeah one thing we always said about building good dashboards is that it takes a village.

So Phil, you’ve actually led the charge in this area at a few startups. What are some of the questions you should be asking as a marketer to get started?

Questions before building

The first questions to tackle as a team are: What metrics would you look at on a regular basis to measure performance and determine areas for growth? What metrics do you care about the most?

So ultimately, this depends entirely on your team goals and the top priority metrics we’ve selected as a group. These goals further inform how to prioritize views and metrics in our dashboard. What does this group of stakeholders look like when you’re starting to build things?

Stakeholder groups:

  • Main viewers: Who will be digesting or regularly looking at the dashboard
  • Marketing Ops/Data Ops: What resources to you have to help you build the dashboard
  • Designer and point person: Who’s scoping out the dashboard and driving project management as well as designing the end dashboard

Admittedly, in startup land, you’ll likely be wearing all three hats. I know I have. But in bigger teams, you’re working with a lot more moving pieces. 

Yeah I’ve gotten a taste of both of these. Small teams and bigger teams. There’s advantages to both. But I think regardless, it’s important to get a lay of the land first.

Yeah it might be helpful to walk through an example. You’ve been pretty deep in lifecycle marketing in your career. Maybe give us a real life example wearing a lifecycle hat. So Phil, you’re Director of lifecycle and you’re tasked with building out a lifecycle dashboard.

Here’s a list of example questions to ask yourself and stakeholders 

Yeah I like the lifecycle example actually. It’s broad enough to touch most parts of marketing so I can  use it as goal posts as we unpack some of this stuff.

Your goal with these questions is to figure out what metrics we care about the most, getting a benchmark and establishing a goal for each of these metrics and how they have been trending over time.

  • Current segment/vertical data we get on signups, are there specific segments we know we want to grow?
  • Current lead scoring on signup events, are we scoring leads based on email and domain and any other data we might be collecting?
  • What’s the current activation rates of signups after the first email, what’s our deliverability rate on the first email to signups?
  • Are there specific lifecycle status labels that we are currently using, ie Content lead/subscriber > Signup > Active/published site > Upgraded. Do we currently have micro stages/do we care about this detail, ie in between signup and active we might have, installed theme, created a page and created a menu.
  • Do we currently have the ability to attribute multi touch events for email engagements? Meaning, if a signup opens a pricing email on day 4 and they click the plans link and they buy 2 hours later, is that email getting $%?

With all of this information on hand, or at least identifying areas of focus and priority metrics, you can then start scoping out the first prototype of the dashboard, intentionally with too much information, with the hopes of cutting things out in following iterations. 

Exactly. Next we can talk about metrics that flow in from those questions.

What metrics you should consider for the first prototype

The critical piece of this phase is to spend time understanding the most important things to monitor and give ourselves time to explore different ideas before rolling out a finished dashboard.

Here are the core areas of a lifecycle dashboard, with a focus on conversion rates, starting at signups (explicitly did not scope content lead > signup):

  • Signups, signups by segment, signups by lead score
  • Confirmations, signups > confirmation %, deliverability
  • Active (published a site)
  • Behaviors (installed a theme, >2 pages, menu)
  • Email metrics, engagement score, top emails, ab tests
  • Conversions to plans, signups > conversions %, % in first 30 days, % after 30 days
  • Upgrades, plan breakdown
  • Revenue impact

Yeah that’s a lot obviously, depends how long you want your dashboard to be but we’re still in the prototype phase here so more is better and you can always remove stuff later or create a second dashboard.

The main takeaway of this episode though as we said is that When designing a dashboard, it’s important to focus on the decisions you want to make, rather than just the metrics you want to track.

So how do we do that?

Focus on the decisions you want to make

Something we want to keep in mind as we narrow the list of important metrics are the decisions we want to be able to make. The goal of our example dashboard is to monitor the lifecycle marketing performance and identify growth opportunities. That means answering questions like:

Are we improving sign up engagement and conversions over time? 

Are specific segments or campaigns driving better conversion rates than others?

Should we double down or kill this experiment/email

So ultimately, the focus of the dashboard should be on Signups > activated(published site) rates and Signups > upgrade conversion rates in the first x days and the viewers should be able to see the impact across the funnel over time. 

So now that you have a better idea of all the metrics you want to start with, one of the next steps you can start thinking about is chart types, how you’d like to ideally display your data.

Choosing chart types

Scatter plots, bar charts, pie charts, maps, funnels, box plots… There’s a bunch of different chart types and visualizations at your disposal when you’re designing your dashboard, but this isn’t where you should start.

Dataschool has an awesome guide on picking charts:

Choosing the right chart types is an important step in the process of designing a dashboard. The right chart types can help to convey information effectively, and can make the dashboard more engaging and intuitive. However, choosing the wrong chart types can lead to confusion, misinterpretation, or even misinformation.

When choosing chart types for a dashboard, it is important to consider the goals and objectives of the dashboard, and the audience for whom it is intended. Different chart types are suitable for different types of data, and for different purposes. 

For example, scatter plots are useful for showing the relationship between two variables, while bar charts are good for comparing values across categories.

It is also important to consider the formatting and layout of the chart types. Different chart types can be formatted in different ways, and the right formatting can help to highlight the most important information, or to draw attention to trends or patterns. For example, using colors, labels, or axes can help to make the chart more readable and interpretable.

I’m going to throw fire here that you don’t need to design a piece of art. I can also plug powermetrics, since we have the ability to dynamically switch chart types 🙂

How to ensure people actually use your dashboard: ID stakeholder, involve in design process, involve in testing and QA.

So getting key stakeholders in the room, asking important questions not just about what metrics are most important, but also what decisions you’re trying to make. Then you can start thinking about chart types and drafting your dashboard. Doing these pieces first will help ensure your dashboard is used often instead of being another tab left unopened. 

Conclusion

Building a great team dashboard is like constructing a towering skyscraper. It takes a village of skilled and dedicated workers to lay the foundation, raise the walls, and finish the details. Just as a skyscraper can’t stand on its own, a team dashboard needs a diverse and talented team to make it functional, useful, and beautiful.

✌️


Intro music by Wowa via Unminus
Cover art created with Midjourney

See all episodes

📫 Get new episode notifications

By subscribing, you agree to receive emails from us, but we’ll only send you an email when we drop a new episode, usually on Tuesday mornings ☕️

Processing…
Success! You're on the list.

Future-proofing the humans behind the tech

Leave a Reply