Three guiding principles for building better marketing analytics dashboards

Three guiding principles for building better marketing analytics dashboards
Emilio Basaldua is vice president of analytics enablement within RAPP’s Marketing Sciences Centre of Excellence, where he brings analytics technology solutions to clients across the RAPP U.S. network of offices. Emilio creates and enables the activation of applied data systems related to measurement, analysis, and optimization within a digital ecosystem.

Modern marketing campaigns are constantly churning out data that can allow companies to continually improve customer engagement. But for that data to be useful, it has to be organised and presented in a way that key stakeholders can understand and act upon.

Enter the marketing analytics dashboard. These visual data representations provide a single, unambiguous source of truth about campaign performance. They allow stakeholders without a background in data science to easily see patterns, trends, and outliers, and to garner insights that enable rapid optimisation and better decision-making on future campaigns directed at individuals.

What you see is what you get

The best dashboards use a variety of dynamic visual aids like charts, graphs, and dot plots, pulling data from multiple disparate analytics tools in real time. They incorporate sensible design and an intuitive UX to illustrate performance with respect to key performance indicators like revenue, engagement and conversion rates, source tracking, attribution modeling, and other significant metrics.

Whether you’re leading an agency’s creative team or in-house brand marketing department, you can significantly enhance your team’s effectiveness and value to your organisation by harnessing the power of visual analytics dashboards. If you want to start developing better dashboards, for your own use or for client reporting, remember the following principles:

#1: Start with the end in mind

Your dashboard will do more than just visually display performance metrics; it should tell a story and answer important business questions. Decide on the individual story you want to tell and the questions that need answering prior to beginning design work. Choose KPIs that are directly linked to overall campaign or business objectives and that can be easily monitored based on the analytics tools at your disposal. The data you display should be actionable and demonstrate how specific variable changes can affect overall performance.

It’s also important to be realistic about your data collection and analytics capabilities. You don’t want to be making important decisions using data that isn’t trustworthy. Dashboards can be tailored to reveal micro-level insights by audience segment or cohort, channel, placement, line of business, tactic, sub-tactic, and many other data “slices,” but those insights will only be as good as the integrity of the data that goes into them.

#2: Measure what matters

Once you realise how much data a single digital campaign can generate, you might want to include all of it in your reporting. But a dashboard isn’t a visual data dump. Prioritise macro- and micro-level perspectives that lead to discovery rather than information overload. In other words, once you’ve set your KPIs, use your dashboard to provide context. Don’t waste time consolidating data that isn’t directly linked to those KPIs or building visualisations that you can’t act on.

For example, scatterplots can compare click-through rates and click-to-open rates of an email campaign, revealing the effectiveness of certain subject lines and content that individuals consume. This information is useful only if you have the resources to make in-flight optimisations. When you’re thinking about the metrics you’ll include in your dashboard, make sure you account for the time and personnel resources needed to act on the insights you’ll discover. Before building your dashboard, you’ll need to determine roles related to data collection, transformation, dashboard development, and user administration, and you’ll want to select the dashboard platform that best suits the support resources you have in place.

#3: Focus on individuals — the end user

When it’s time to start development, keep the perspective of the individual front and centre, because user experience is everything. Your dashboard should be seamless and intuitive and should incorporate a navigation framework similar to what users might encounter when surfing the web or using common software tools. It should also account for design best practices. For instance, colours should be used consistently across the interface. If the conversion rate metric is a blue line on one visualisation, then it shouldn’t be a red bar on another.

Don’t include fancy charts just for the sake of it. For example, while a sleek animated chart might look impressive, it’s often hard for viewers to glean the significance of the illustrated changes while chart objects are constantly moving around (a bar chart showing period-over-period changes might get to the point faster and save dashboard real estate and development hours). By contrast, interactivity in the form of filters, parameter controls, highlighters, actions, sheet swapping, and other functions can allow viewers to consume and manipulate information at their own pace.

Remember, a dashboard is designed to make developing actionable insights easy — but that doesn’t mean developing a useful one is easy. It takes careful planning and creative execution to build a sophisticated marketing analytics dashboard capable of truly moving the needle. Keep the above principles in mind when you set out to build yours, and you can turn reporting into an organisational competitive advantage.

Read more: Marketing dashboards matter: Why marketers must go beyond traditional BI tools

Interested in hearing leading global brands discuss subjects like this in person?

Find out more about Digital Marketing World Forum (#DMWF) Europe, London, North America, and Singapore.  

View Comments
Leave a comment

Leave a Reply

Your email address will not be published. Required fields are marked *