Marketing data: guide your marketing better
  • Juska Glán

Marketing data: how to guide your marketing with analytics

Updated: 3 days ago

Marketing is useless without business-related goals. To reach your goals, you need to know where to go and how you are going to achieve it. Marketing data is the compass for your journey.


Then comes the questions "what is marketing data?", "What should we do with it?", Where can we get it" and "why?".


In this blog, we will go through the basics of marketing data, what should be taken into consideration in terms of your business, and how marketing data can be utilized to guide you forwards.


All the power is in your hands now, my reader: you decide the path you take. The question is if you want to go blindfolded or use a map to get to the right place.



The basics of marketing data


Marketing data is information generated and gathered from organizations’ marketing operations. Data tells “what has happened” in organization-specific channels, tools, and/or activities.


The common types of marketing data are:

  • Marketing metrics. E.g. Customer Acquisition Cost, Monthly Recurring Revenue, Click Through Rate, Cost Per Click… The list is nearly endless. Source examples: Facebook Organic or Ads or Programmatic buying (e.g. Adform)

  • Marketing KPIs. Customer Lifetime Value, Sales Volume, Number of Deals, qualified leads Source example: CRM (e.g. HubSpot, Salesforce)

  • Customer Data & Preferences. Consists of information regarding your customers. You could have their contact details or have complete data about their behavior and preferences. Source examples: marketing automation platforms, CRM, ERP, email marketing (e.g. HubSpot)

  • Sales & financial data. Leads, opportunities, proposals, MQL-2-SQL rate, and so on. Source example: CRM or ERP


The sole purpose of marketing data is to improve your business by developing better marketing results. Remember: metrics are single data points and analytics are the solution to tie your marketing information together. KPI’s, on the other hand, does not equal metrics. KPIs incorporate the objective and goals of the business and they are used to guide businesses’ future.


Of course, data itself does not shoot you to greatness.


Everything you want to do with data is dependent on the quality and content of the data. It is good to bear in mind that having all the data and using the correct data are totally different things. A map is as useful as it's purpose and content is.




Is your company ready to use marketing data?


You must evaluate your organizations’ maturity to use marketing data. According to my experience, the majority of marketing decision-makers say that their company uses marketing data.


Monthly reports from metrics like impressions, followers, post clicks, and keyword performance is a form of using data. Metrics themselves are called vanity metrics: they might be impressive figures but rarely describes anything about your real marketing performance.


Here's a fact: all companies can (and should) use data from marketing for business value. There’s a lot of ‘whys’. For example, it gives

  • Insights for decisions to grow marketing ROI = better decisions

  • It helps communicate the ROI of marketing to other stakeholders = other business units understand the value and marketing can hold itself accountable and develop constantly

  • It gives centralized and automated data assets = No lost resources on data management and experts can focus on important tasks

Additionally, there are operative advantages such as

  • Operative efficiency = Better processes, better ROI on work time

  • Better information regarding target audiences and current customers

  • Optimized marketing budget

  • Tying offline and online marketing together

  • Organizational data reading skills

Like with anything else, by starting to use data in your organization, you will find other profitable use-cases as you go.


The early adopters have been creating better results with marketing data for years and slackers are hopefully starting to understand why they should do so too.


There are two questions to start producing better results in e.g. digital marketing:

  1. What is happening in our digital marketing channels and where should we invest in?

  2. Why?


Marketing data and marketing analytics


I am going to introduce the three analytics maturity levels. Why I use analytics you may ask. Because analytics tie everything together in terms of knowledge, measuring, investing, developing, and optimizing. It's the nervous center that builds the map for you. The analytics maturity framework is relevant to all companies in all industries. Data can be of course used in other ways too - the essence is to understand what value the information can deliver.


Maturity levels are good to understand as higher steps mean better data. If you want to have more information regarding this from a data scientist’s point of view, check it out from here.

Think these as “levels of complexity”. It is not guaranteed that you should or that you even can take the following steps in presented, "logical order". It is important to understand what stage your business is at and what your business needs.


Answer (at least) for the following questions when you are starting a marketing data related project


  • Why are we doing this?

  • Where are we are the moment?

  • What do we want to achieve?

  • What is the value of the project/solution for my business?

  • How is the human factor taken into consideration in the process?

  • It is not usually about the technology – make your stakeholders understand the benefits


Technology itself does not change your business. The people using it and embracing the idea of better results will. All the operations produce actionable data which can be then used in the analysis for better results.


The maturity for analytics is not first about technology. The most important factors are your business goals, people, processes, and information culture: the human factor.


Technology itself does not change your business. The people using it and embracing the idea of better results will.

When choosing a technology provider, in this case, analytics tech, or when you start to build a system in-house: remember that you must not forget to build an analytics strategy and get your organization (the people) behind it.


I recommend having a meaningful, business-related goal (a reason) to start using analytics. The same applies to marketing data. After you have a goal panned out: take it step-by-step. Building any solution to solve a problem is a journey. To make it successful, you need the right people and solutions.


Descriptive analytics: Reporting – What happened?

Let's start with the analytics maturity with reporting.


To start the change, you must understand what has happened? For example, let us imagine that your company has set up a campaign across different channels and you are looking to understand how the Omni- or multichannel campaign performed.


The purpose of the reporting is to help you learn from the past.

This is incremental: if you are constantly collecting information from your digital marketing (channels, campaigns, the performance of copies, and visuals of advertisement, conversions) is your next step just to report them to a stakeholder, or are you using the knowledge to develop your performance and hold yourself accountable? Your decisions to act on given data makes the difference. But how should you act?


All the companies with marketing operations (internal, external, extra-terrestrial) should aim to be at least one level of descriptive analytics. Sadly, this is not the case. The companies which understand their digital marketing mix’s performance and results are much more prone to tackle hard financial times and generate demand compared to their counterparts.


Keep asking the following questions: are we doing the correct actions in the right volume to reach our goals? [Rant] Remember that the baseline for any advanced solutions is to have a stable foundation to build on. [Rant/]


Basics of descriptive analytics:

  • Marketing data is collected and updated manually for example on spreadsheets or light connector technologies

  • Descriptive analytics, also known as reporting, can be harnessed for operational efficiency if automation is brought in

  • Visualized data usually consist of metrics and it can be mirrored to targets

  • Information is usually provided for a stakeholder to tell what has happened - I call this as a back-mirror effect

  • Common use-case: customer or stakeholder reporting from a finished campaign or previous month is a common use-case



Predictive Analytics: Analysis & forecasting – what will happen?


The next phase from a descriptive analytics/reporting and understanding of the past is to look for the future.


It’s complex as here we’re thinking and looking answers for “what will happen if we change X, Y or Z” and “how will it affect us if these trends continue like in scenario N” and “what could happen regarding our digital marketing mix?”.


There's usually a chasm between descriptive analytics and predictive analytics when companies are trying to make the shift from the former to the latter. Here's a good article about the chasm. There's also our picture for the matter as a picture at the bottom of the blog.


Having gathered and stored information helps you to build scenarios and "forecast the future".

Your data’s quality is equal to your analytics quality. According to Eric Bradlow, you have to have “better data, not big data”.


In many cases where companies are looking to achieve predictive analytics to develop their marketing, they realize that they lack the important step: having a solution to gather, store, clean, process, and unify their data while ensuring the quality of the data. This means that the data is not actionable.


There are two common ways to achieve actionable data-analytics.

  • Hire an expert or a team to manually handle the data from start to finish and build a customized process and system

  • Set up a solution to automatically produce you the goods

Bot have their best use-cases.


Everything boils down to data: if you do not have it, you will not get any results. Analytics is a presentation of the information that its fed with. Predictive analytics is an effective way to help decision making but it’s really important to understand that you need a solution that is able to fetch information from sources systems, store it to warehouse, transfer, and unify it for further use and then provide actionable insight into it.


Jouni Leskinen (LinkedIn), my colleague said it well: “The difference between descriptive analytics (or reporting) and predictive analytics is that in latter the outcome is a forecast, a sophisticated guess. A forecast always includes a probability and error margin. The data is indeed the most important factor. An excellent algorithm is useless without quality data, which in this case equals 1. enough disparate sources, 2. enough historical data, and 2. the required amount of variance to describe events with required probability."


Does this sound easy? Usually, it is not.


Example of predictive analytics:

  • Marketing data is automatically gathered and processed by a marketing data platform

  • Data is collected from multiple sources and data is processed and managed inside the solution for more advanced needs

  • Data visualization utilizes marketing data to be used by ad-hoc queries and advanced analytics

  • Information from marketing channels can be used on-demand

  • Marketing experts and teams are looking to make more sophisticated decisions with data. A specific use case could be marketing campaign optimization where gathered marketing data is used for better targeting and messaging.


“The difference between descriptive analytics (or reporting) and predictive analytics is that in latter the outcome is a forecast, a sophisticated guess. A forecast always includes a probability and error margin. The data is indeed the most important factor. An excellent algorithm is useless without quality data, which in this case equals 1. enough disparate sources, 2. enough historical data, and 2. the required amount of variance to describe events with required probability." Jouni Leskinen, CEO of Madtrix


Prescriptive Analytics: Optimization – what is the best achievable outcome?


Would you like to know what would be the next best thing to do regarding your marketing mix and to get the best possible outcome?


There are possibilities and solutions for that. Do you know what is the most important thing? You got it: *data* (and well-established foundations).


You need quality data and from diverse sources to be able to make the most out of your optimization.

Do not discount descriptive and predictive analytics off. They are always needed as they provide invaluable information form customer behavior. This information gives you the possibility to build and grow your business as it uncovers buying journeys, preferences, interests through clicks, and actions.


An example from prescriptive analytics would be a PPC (Pay Per Click) cost-to-result analysis. It would analyze your PPC channels (for example Facebook, LinkedIn, Google Ads, and Adform) in regards to cost-per-click and then connect the gained value (conversions with it) and then give optimization possibilities from the accumulated data. This is an ROI-centric case in its heart but result-oriented marketing is one that keeps the boat afloat and running.


Prescriptive analytics is the pinnacle where we can start to produce optimized actions that will make an impact on your marketing. Opinion alert: it is always the responsibility of a marketing expert, a marketing team of a marketing decision-maker to prove the value of their operations. These actions are done better when you have the best possible marketing data at your disposal.


Recommendations:

  • Understand your businesses situation, problems, and needs

  • Build the foundations first: have data and analytics strategy, set up the foundational processes, have a meaningful business goal

  • Generate data from multiple touchpoints from many channels

  • Have a solution to get data stored, unified, and granulated to be used for analytics

  • Ensure constant development and backing of your organization



Note to the reader


Do you remember the phrase ‘technology is not the most important’? As a business developer working in SaaS tech-company you might think that as an odd statement to give. Hear me out.


Let us have a totally off-topic example.


We would like to start a new podcast (Let us name it as “The Best”). The first thing would be to establish the technological and operational groundworks to be ready to distribute our content. We will define the equipment and tech with online research, contact podcast professionals in our network, and then produce the framework for content for first shows. Sounds easy enough.


Now we have everything, right?


The Best is a simple process. If you produce a podcast as an individual project: you might be good to go. But think if you want to produce it for your company. You need: highly targeted and interesting content -- great guests will not do harm. The whole process should be tied to your communication and marketing process and goals. It should be held accountable as a channel and activity. Everything is planned, operated, and monitored by a human. Further decisions are made by a human.


How has this anything to do with data? This example is not tied to the subject in any way. Think about making a podcast without

  • Humans

  • Targeting it for humans

  • Strategical plan on how to make it happen and what to achieve with it.

  • How to orchestrate the people related to the podcast.

Marketing channels, technologies, and solutions require humans to function. Even though you can automatize marketing and sales processes, you cannot automatize your customers, users, team, and partners just yet.


If you automatize everything (which a good direction to aim) and you won’t have knowledge of performance and how to measure it, how will you guarantee that the baseline results from automatized marketing are good?

Over to you: marketing data is the map

We went through the journey of marketing data knowledge and analytics together. Now you can yell like Jack Dawson in Titanic. We have had an analogy about the map in this blog. It is your decision if you go for the iceberg or sail past it and reach greatness.



The optimum direction is not easy to find in marketing (or any other field for that matter). Usually, the easiest way is to go with the “I feel”-reasoning. We all feel something. The guesswork should be minimized in decision making.


Be it an operative marketing expert or marketing decision-maker, an aligned direction with a marketing organization directs the ship to the correct compass point. If there is a common direction to work for, the people will always know where to aim for with their actions.


My recommendations for utilizing marketing data:

  • Talk to your peers and with other experts on marketing data

  • Understand what are your organization's specific marketing (data) problems and more importantly “why?”

  • Form a data strategy if you have not done it already

  • Pick solution partners that provide you more value than just a software

  • Measure only the things you want to improve and that is related to your business goals

  • Connect and communicate your organization with data. Share your “map” and results with stakeholders

  • Inspire and motivate people with data: create transparency with results and give credit with it


If you are not happy with the amount or quality of the information in this blog, send me an email to juska.glan@madtrix.io or message on LinkedIn and tell me what I missed.


Here's a picture describing the stages of marketing analytics maturity on a more granular level.


The basis of everything is data.

Rautatieläisenkatu 6, 00520 Helsinki, Finland