ADTYPE

Visualize campaign data to calculate future projections

Combine data from sales, accounting, inventory, marketing or other data sources to generate unique insights about your company.

COMPANY

AdType

YEAR

2019

TIME

4 months

ROLE

Founding Product Designer working with founders, responsible for analysis, product design, project management and QA testing

As the only designer in a small team, I got to play several roles to fill in the gaps.

Responsibilities and learning opportunities

As a Product designer, I was talking to data analytics and sales people to understand what reports we tend to create for our customers based on the data we get and with engineers to figure out how to connect different data sources into dashboards. Founders helped with defining what stories we want our reports to tell.

As a Project manager, I got to break down the design into deliverable chunks for the engineering team and make sure that the different use cases are documented as well as provide progress updates to the founders.

As a QA, I was the first who got to play around with new releases, gather bugs and improvements that were documented in Jira.

AdType noticed that their customers were not able to provide key data that they already had because of the complexity of the products used for data analytics.

Context

AdType was a digital marketing agency employing specialists to help clients plan marketing activities and improve campaign performance through data.

Many just stick to the tool they purchased years ago unaware of newer and simpler alternatives.

The goal is to provide a high-level overview of client business data, including offline metrics. Most importantly, it helps companies get more insight from what they already have.

Designing in an unfamiliar environment

I have not had the chance to build data-heavy dashboards previously and the world of understanding marketing jargon and the meaning of different data points was a learning journey.

Discovering what I don’t understand

Working on such a data and marketing-heavy product was a challenge for me but going through a design thinking process helped me to pragmatically learn what I needed to know:

  • Started by talking to senior stakeholders to understand what we wanted to accomplish

  • Documented findings to start building a source of truth

  • Facilitated ideation workshops to get more insights and align different points of view

  • Created a list of feature requirements to keep details top of mind

To combat my lack of knowledge of data-heavy systems, I made sure to iterate quickly and often and share my understanding with stakeholders to gather feedback. Each iteration taught me more about different aspects of marketing campaign performance measuring.

Research through in-depth conversations

Because I had not designed such a high-complexity data visualisation tool before, there were many things to figure out.

  • I did competitor research to understand how they do data visualisation to learn that we could lead the way with a clear and understandable story from the data. Explain it a bit.

  • Understanding metrics was crucial to tell the right story in each report.

  • I iterated one feature at a time to shorten the feedback loop and accelerate our understanding of the product.

MVP’s over polished designs

Iterating quickly was the point to keep things moving. The design goal was to make it easy to see how a large part of a business is doing at a glance. Colours were used sparingly to show crucially important information that stands out from normal operations.

A responsibility for what I design

I created detailed tickets describing all the features and how they worked and had recurring sync’s with engineers.

Design acceptance was sometimes tough to enforce because an not everyone saw a few pixel offset here and there as an issue to involve developers.

How to model future growth?

The most challenging part of this report was visualizing all the necessary data points and metrics in an easy-to-understand way while still being able to modify different attributes.

You can also switch to Sales Growth which would only include data from sales made online and offline. Users can create custom projections by adjusting different metrics like:

  • Amount of new customers entering the sales funnel

  • Average Order Value for new customers

  • Revenue gained by existing customers

  • Average Order Value of Existing Customers

Who am I designing for?

As the development of the tools had only started a few months before I joined, no data was available to understand who would be using the tool.

The assumption was that the main customer base would be Senior people in Marketing, Sales, Management and our team, of course.

Challenging old ways of doing

The primary challenge was to change user habits by introducing a flexible digital tool to users who were used to Excel spreadsheets. The slowest adopters were people in management roles.

Facilitation as a mechanism for learning

The first version of the product was built with little to no design oversight. It worked but it led to inconsistency in colours, layouts, button sizes and general confusion on how to get around.

People were used to handing down Google documents and hand-drawn paper sketches which often led to more confusion. I facilitated several discovery workshops that led to better communication and more shared product knowledge.

Understanding the underlying concepts

I am not a marketing professional but I joined a team of many highly knowledgeable ones. This meant that when I was introduced to new requirements, I asked people to explain what it meant and how it worked.

Over time marketing speak stuck and the exposure to concepts raised my baseline understanding.

Solving for information density

Combining data from tens of spreadsheets and other data sources in diagrams, data tables, fitting that into a single view can get complicated. But users made that a requirement to see a full picture.

Prospects Database

There is a view for each of the databases but the only difference between them are the title and numbers displayed.

You can see how much movement there has been within the database with indicators of positive and negative statuses. The change is visualized as well as explained in detail in the table below.

See customer demographics and use precision tools to start a campaign.

Since GDPR it is important how it affects your ability to market.

Using the map below shows you the location and the density of people in the database which can be used for marketing efforts in the real world using Display advertising.

Users should be able to track the overall health of the business based on historical and projected future revenue and campaign data

Packed with data

The initial challenge of Growth Modeller was to understand what is necessary to be visualized. Many data points need to be included for the report to tell the story right. The chart includes 6 different customer types at different times of their life cycle.

Find the most valuable customers

Given that you have acted on the information available, how do you know that you will meet the companies revenue goals this year? This is where the Trend Spotter comes into play.

Project future revenue

Users can see the potential revenue the business can earn going on the same trajectory and, by adjusting different attributes, businesses can plan their marketing activities, focusing on the areas with the biggest possible ROI.

Data controls

This is the place where you can see how relevant is the data within your reports. This place would show when was the last time online and offline data sources were synced and if any issues need to be addressed.

Trend Spotter

The main takeaway from this report is that you will know if your business will reach the end-of-the-year revenue target or if you need to make some adjustments to your marketing plan.

I was able to include more explanations about the data and metrics by suggesting that other — less marketing savvy — users might benefit from this report.

This report gives you a detailed insight into business performance and informs the user if revenue targets can be reached based on the current trend.

Widgets at the top visualize key metrics as if the Average Order Value has increased or decreased compared to previous periods.

Variance Report offers a detailed explanation of the metrics mentioned above and offers actionable insights and an explanation of what it means.

Variance report analysis is what you might think. A simpler explanation of the metrics mentioned above. The most valuable explanation here is the indicator that tells you if the revenue goal that was set will be achieved by the end of the year.

Variance report breakdown is there to visualize Variance report analysis, filling in any confusion you might have until now.

P.E.C. Database

Since the EU’s GDPR law, the importance of understanding which customers you can market to and which ones you should leave alone has become very important, given the possible issues that can come from ignoring this law.

  • Prospects - are people who have not interacted with your company in any way ex. bought e-mail lists

  • Enquirers - are people who have interacted with your company by signing up for an e-mail list, filling out a brochure with their contact information or leaving an empty shopping cart (to name a few)

  • Customers - as you might have guessed, are people who have purchased something from either your online or offline store

This view provides an overview of all three of these databases (yes, you have to separate them for more precise reporting) and sees in what condition they are.

For precise marketing efforts, each customer database needs to be healthy

On top, we have the analysis of all entries in the database. This includes duplicates, and people you can and cannot market to. Below that you have people to whom you can market sorted by their source and the ability to create a custom campaign with the included Campaign Builder.

You can see the fluctuations within all databases by month. Each metric in the chart represents all 6 customer types that make up a customer’s life cycle - starting as a new customer and ending as not purchasing anything for 3 years.

You can also see the movement between databases which helps you understand where people are dropping off, how well are they converting between databases and how much money movement between databases (customer states) costs. The cost is an approximation based on many different variables and put together by a very smart algorithm.

Having an overview of all three databases is useful for marketing purposes. If you want to know if you can achieve the end-of-year revenue target, this helps understand if you have enough people in the database that you can market to.

You can also dig deeper into each database where you will find similar information about each of the databases.

Product Report

This helps with understanding which products sell best at what time of the year to better plan and experiment with discounts.

Depending on the products sold, we can determine customer buying patterns and product profitability.

The report also helps to manage inventory by telling when a surplus is needed by anticipating an increase in sales.

For future iterations, it will be possible to select products, select a marketing channel, select users to market to and start a campaign from this screen.

This report can be viewed in pre-set time dimensions or can be set to show custom time frames.

Customer View

Here you can find all of a customer’s purchasing history, location history and customer status to better understand their value for the company.

We were able to determine how likely each customer was to make another purchase or not make one at all

Each customer is given a score that determines their value to the company based on engagement, purchases and other metrics. This means that you could easily identify people who have bought this certain product a few times but not this month, for example.

Final thoughts

I am very interested in complex data products because you can learn so much

Not only the complex ways that you can use browser cookies and aggregate data from many different sources to get a very detailed and precise view of your customers and their interests but also the challenge of getting familiar with the different ways to visualize data and constraints you have to work around to make the data tell a story.