Data-driven marketing: How to start
The basis for performing marketing is quality and good data. Read how to thrive in a data jungle and become a data-driven marketing organization to ensure fast insights with the best value.
What is data-driven marketing?
Data vs. feelings. Discussion regarding these two entities is raving – which one should you rely on your marketing on and which one is the “truth”. This is unnecessary: you need both.
Data cumulates from every marketing action, be it offline or online. Only the format and source system of data changes. Data is the part that provides understanding and drivers for growth through information created by marketing actions.
Marketing has multiple, high-important functions to deliver: sales, communications, brand & its values, internal functions to name a few. Behind all this, an expert is making decisions, creating a strategy and operational actions to reach set goals (which should be always set before anything is done)
Data-driven marketing is a world-calls combination of data and feelings/instincts. Data is gathered from all the valid data sources and processes for strategical and operational use. The most important job for it is to support decision making and show how marketing is performing.
Additionally, the best customer experience help companies win competitors. But it’s not only marketing and sales that need to work together: the whole organization must get behind this initiative. Even external stakeholders such as partners, agencies and other service providers must be held accountable to provide the best possible customer experience. Data is the best buddy to understand your customer experience: Without data, it’s hard to understand how’s your customer experience doing
1. Centralize your data
Get your data together.
Companies want to increase their performance using data, but the data is spread around in disparate sources. Ultimately, it’s very hard to integrate and turn into actions to drive growth. The marketing technology stack is full of applications for many specific tasks. Let’s not even talk about external data. Oh yes, later we will.
A new generation of companies is emerging which has a lot of easily to deployed SaaS solutions on the grasp and in use. The data is “stuck” in separate solutions and the analysis is done using the analytical features of each solution or by manually building combined excel-sheets. Sounds handy, huh?
Marketers often must rely on analyzing each channel separately from the ad platform analytics like Google Analytics or Facebook Business Manager. Most modern SaaS solutions and almost all marketing platforms have APIs which can be used to extract the data out of the solution. It’s still hard to pull out the data into one place in a format where data from various channels could be integrated. Add some company-specific data or unstructured data to the mix and the process gets even more complicated.
2. Get quality data
Quality data is crucial for increasing performance and this is a huge opportunity in marketing.
Not even the best algorithms or models could make up the lack of a high-quality dataset in data-driven marketing. How do you build those datasets you may ask? (Wikipedia answers what is a data set)
Every organization has a dashboard excel sheet or visualization tool where the most important metrics are gathered and reported to the management. The problem is that often every metric available belongs to somebody’s “most important list” and this leads to huge amounts of dashboards filled with tactical or so-called vanity metrics which makes it hard to make actual strategic decisions. Or any decisions.
It’s relatively easy to bring together metrics from multiple data sources into a dashboard: there are even free tools for this. But often in the dashboards, the metrics are displayed in channel-specific reports. This leads to a siloed view of marketing. Even though a dashboard is a tool to distribute the data for the organization and important stakeholders, high-quality datasets are required to get the right stuff to the right people.
Consider these when evaluating data for marketing:
- Instead of just tons of metrics, define a few important KPI’s, which the organization wants to systematically improve and gather all the data that is needed to produce those KPI’s.
- Examples of important KPIs: MRR, Customer Lifetime Value, Cost of Acquisitions, Marketing ROI
- Focus on the outcomes, not outputs (for example, impressions and clicks are output-values). Opportunity, sale or recommendation are outcome-values.
- Concentrate on really understanding your funnel dynamics or flywheel or whatever model you use to describe your interactions with current, churned and future customers.
- Ask questions like “how can I make it happen” instead of “what happened”. Evaluate can your data give answers to these questions.
3. Metrics are a good start, but…
Measure performance (from actions to overcome), not only metrics (action).
Marketers have growing opportunities to deliver real business results and growth. More and more CMO’s have the same performance goals & targets as CEO’s (there are even indicators that CMOs are increasingly likely to shift to CEOs instead of their other C-suite counterparts).
As the questions asked when figuring out how to drive growth are different than when just reporting what has happened, the techniques for data harnessing are naturally different too.
Metrics dashboards are not enough to satisfy the demands of data-driven marketing. Metrics are a good start, but they simply are not enough. You can get important information from metrics, such as clicks, but you should know the reason and correlation behind that action.
Getting answers to relevant business questions to improve performance requires the integration of all relevant data and the ability to dig deeper into the actual outcomes instead of just clicks or leads.
4. Data warehouse – “the second coming”
Data warehouses have been for decades the “go-to” solution for other business units. Fortunately, marketing is starting to tag along. Traditionally the use of data warehouses has been reporting oriented and the data has mostly been originated from internal sources. Marketing is opportunity oriented, which sets special requirements regarding agility and speed in the data-driven decision making.
The traditional data warehouse is optimized for slicing the data, drilling down on details and getting answers fast to questions like what happened in the past, just like most of the marketing dashboards used today.
When data warehouses were built, the use of the data was decided before it was loaded into the database. Often only the specific domain-relevant data was extracted from the sources to avoid the data warehouse getting too big and expensive. You guessed it: this approach leaves a lot of business value on the table.
In “ancient” times the data was forced into an optimal structure for the low latency queries. All this was done before the data was loaded into a data warehouse, which is why it usually took a lot of time before the business users got the first results to support their decision making.
Now is the second coming of data warehouses. Cloud and big data technologies have matured into building blocks of modern data platforms and services can be built for agility and performance. Modern data platforms can handle an almost infinite amount of data and semi-structured data is the norm, especially in marketing analytics.
Data-driven marketing: glance to your future
Performance Analytics requires a different approach to use the data. In performance analytics, the questions are “what will happen” and “what should I do to make it happen”.
As we are looking for ways to predict the future, the data must be the best possible regarding marketing actions and environment: the data needs to be organized in a format where it can be analyzed.
The demand for an analytics data model in marketing is high, but it has to be up and running fast and it has to be agile enough to react to change because changes occur all the time. This applies to the trade of marketing more than for most other business units and functions. Marketing is all about taking on opportunities, experimenting, learning what works and optimizing actions based on that.
The need for more advanced methods of dealing with data in marketing is caused by the following trends:
- The best customer experience wins – everybody in the organization is in customer service
- The amount of data and data sources used by marketing is increasing
- The number of external data sources is bigger than the number of internal data sources
- The number of data types and formats is increasing
- The speed of business and decision making is increasing
- The digitalization of products and customers has forced the companies to recognize that metrics are not enough, the value is in the performance analytics
- Everything can be global in the modern business landscape
To get and stay ahead in the competition, the path from data to business value must be fast and the whole organization must be empowered to make insights from the data to drive their performance. The ones who can make good decisions fast will win the game.
Taking your company’s marketing to the next level with data can be complicated: take the right steps and make it easier.
A bit about the author. I’m the director of Research and Development at Madtrix. Concentrating on growing our business as the solution for complicated, manual marketing reports and dashboards that make all our lives much more difficult than it needs to be.
To speak directly to me email email@example.com