Take control of the data you possess and the data you want to collect

If you want to implement a marketing automation system, you need to know what sort of data to collect and what to do with the data. Paradoxically, too much data is no benefit at all, as it’s difficult to draw conclusions from large datasets. The more data you process, the more expensive the system license also is, making you bear unnecessary expenses.

Here, you’ll see how marketing automation systems make it easier to collect and process valuable data and what sort of data are worth collecting and processing in marketing automation campaigns.

Why do you need to collect data?

Let’s start with an easy one – what do you need the data for? Why do you collect data? Nul points for answering “Just because they’re there”. They’re like fresh flowers. They cost a lot, wither fast, and cause some trouble if kept for too long. That’s why it’s not worth “stockpiling” data. Ensuring they’re up-to-date costs a lot, and using outdated data is worse than using none at all. It gives you the false sense that everything’s alright. When your campaign fails, it’s hard to find the underlying problem. Whether or not the data are up-to-date is the last thing you’ll be checking in search of the cause of your failed marketing and sales operations.

That’s why you should make sure you know what to do with the data you’ll be collecting. If you’re not certain a given piece of information will be useful for you, you’ll most likely never make use of it at all.

Data review

Implementing a new marketing automation or CRM system gives you a good opportunity to take a closer look at the data you possess and decide whether it’s worth transferring the data to the new system or integrating the data with it. You should pay special attention to:

  • The ‘last updated’ date – if the contact details in the record haven’t been updated for over 2 years, it’s most likely that the data subject will have changed their telephone number, job and employer.
  • The data’s legal status – can you say for certain that the data have been collected in line with the currently applicable regulations? Do you have the right to process these data? Are you sure that the data subject hasn’t demanded that you delete them?
  • The consent to communication – data entered into a marketing automation system will most likely be used for communication purposes. Make sure that consent for such use was given.
    Hint – data found in the CRM system are usually entered by salespersons based on the data they obtain through meetings, conferences or telephone calls. While the salespersons themselves may contact the data subjects, you’ll need additional consent from data subjects to communicate with them through e-mail or marketing automation systems on a mass scale.
  • The completeness of the data – if, for example, data on diet preferences are included only in a few percent of your records, it’d be worth reconsidering whether you need to enter the data into your system at all. Using such data on a wide scale will be unprofitable; complementing them, in turn, will cost a lot.
  • The data format – it’s good to work with an IT department or a database specialist in this respect. It pays off to check whether all data are in the same format before they’re imported or before databases are joined. The traditional xx-xxx postcode format, for example, is often disregarded, with just an uninterrupted string of numbers (xxxxx) used. If you overlook this, you risk having an inconsistent database and serious mistakes in segmentation or reporting.

This stage isn’t met with a lot of enthusiasm among the marketers but ensuring high-quality data is a must if you want high-quality leads.

It’s worth keeping in mind, too, that it’s merely a preliminary stage. It only ensures the adequate quality of the data as at the day they’re loaded into the system. After all, marketing automation systems “feed” on the data; they make it easier to collect, process and use the data in communication, for purposes such as segmentation or content personalisation. This means you also have to make sure the quality of the data does not diminish along with their processing.

How to maintain the quality of data in marketing systems?

Here are some good practices for using data in marketing automation systems that prove successful in any system you use.

  • Reduce the number of forms. Web forms are the most popular way of collecting data, the early leads. Data obtained in exchange for an e-book, newsletter subscriptions, webinar registrations – all of these are covered by various types of forms. Practice has shown me it’s best to have as few forms as possible. The risk of errors and mistakes becomes lower. Once, when we were making a new form, we mapped some fields wrong. Details about cities started going to the “Surname” column. It’s easy to imagine the chaos it caused. From that moment on, I’ve been very strict about quality and consistency in forms.
  • Limit the amount of data you collect. Five fields in a form already tests the limits of our customers’ patience. It’s not worth risking losing a potential lead just because you ask them to provide too much data. Another argument in favour of less data is efficiency – the less data there are, the faster the database will work. Set a standard for collecting data, and it’ll be easier for you to limit the number of forms you need to have. Just to clarify, I’ll add that marketing automation systems allow for using one form on many platforms.
  • Don’t force your customers to provide data. If you make them give you their telephone number, and all they want is just to download your PDF, you’ll get an answer along the lines of +11 111 11 11 11, which will be pretty useless. Particularly at the early stages of contact with your potential customer, it’s worth reducing the scope of data collected to a bare minimum – the name (to greet the recipient) and the e-mail address. Too much, and your database will be flooded with creative, vulgar and random strings of characters.
  • Don’t collect data too early in the customer journey. The top of the sales funnel consists in leads which are least likely to engage in a purchase. By turning them into qualified leads, you lower their number and tighten the funnel. There’s no sense in collecting data from everyone who visits your website. Most visitors won’t become your customer. Even if they subscribe to your newsletter or provide their e-mail address in exchange for an e-book, you still won’t have sufficient knowledge of their needs for you to collect their data at this stage. Fortunately, marketing automation systems make it easier to pinpoint the moment it’s worth asking the visitor for their data. After 3 visits in a week? Or maybe after being redirected to your website from an accurately targeted marketing campaign? The best moment can be determined in many ways. However, as practice shows, the first visit is not the right time to collect data from visitors. Here, the question of the budget returns. In most marketing automation systems, the price you pay depends on the amount of your records. A large number of low-quality records (e.g. of those who downloaded your e-book but who aren’t in your target group) means large license costs and practically no benefits at all.
  • Plan your updates. People change their jobs and addresses. Companies go bankrupt, rename themselves and merge. That’s why it’s worth planning the manner and frequency of data updates. It often pays to use data from the CRM or customer management system. If somebody updates these data in those systems, it’d probably be good to update them in the marketing system too. You should also take social media platforms into account, like LinkedIn, where users announce the changes in their employment status. From practice, I’ve learnt it’s worth doing that at least once a year.
  • Centralise subscriptions and cancellations. The GDPR has done great a deal to raise awareness in terms of personal data protection. Customers cancel their mailing list and newsletter subscriptions more and more often and are irritated when a company keeps on sending messages. However, apart from the customers’ anger, you may also have to face legal consequences. Make sure your customers are managed appropriately in this matter. Good marketing automation systems allow for subscriptions and cancellations to be managed centrally. With adequate policies designed and implemented in such systems, the customers who want to cancel their subscription are automatically removed from the legally required communication systems. In this case, sending messages to such customers by mistake is practically impossible.
  • Automate the detection and removal of mistakes. Many mistakes in databases come from the data subjects themselves. With typing errors, incorrect abbreviations (Sp. z o.o. vs Sp z oo) and shortened names of cities and towns, your databases are constantly subject to mistakes. Good marketing automation systems include units that can detect some of such mistakes and rectify them. If you want to process tens or hundreds of thousands of records, make sure that your marketing automation system is equipped with such a functionality. Otherwise, you’re doomed either to engage in a costly and laborious process of cleaning the database or to have a completely cluttered database.

What kind of data is it worth collecting?

With the formal aspect of processing data covered, let’s turn to the kind of data worth collecting in marketing automation systems. Here, the concept of Digital Body Language comes into play.

Traditional communication involves significantly more content being communicated non-verbally. By analysing a person’s body language, we can gain much more information than only by listening to them. The same goes for digital communication. Analysing your customers’ “non-verbal” behaviours can significantly broaden your knowledge of their needs, preferences and purchase decisions.

What is Digital Body Language?

Digital Body Language (DBL) is a set of behaviours in which those who use the Internet and other digital resources engage. Thanks to analytical and marketing tools, you can observe those behaviours and adapt your communication strategies accordingly.

The most comprehensive idea of DBL was developed precisely for the purposes of business marketers. Steven Woods, founder of Eloqua (now part of Oracle Marketing Cloud), developer of one of the first and most advanced marketing automation tools, described it in his book, Digital Body Language.

The idea is that the complete profile of a potential customer depends both on directly (verbally) provided data (e.g. those provided in a form), and indirectly (non-verbally) provided data that stem from observing the customers’ behaviours.

Data provided in a form will tell us who receives your contents; observations of these persons’ behaviours will tell us about their interests and about their interest in your offer.

What can you read from one’s behaviours?

Quite a lot, as it turns out. The easiest thing to do is to count the number of sites your visitor has browsed through. If you add factors like the frequency of site visits, the Call-To-Action click-through rate, the number of forms filled etc., then you gain an interesting image of the interest taken in your offer. It’s also worth considering whether the lead “consumes” diverse contents. A person who visits your website three times will be less of an interesting lead than a person who downloads an e-book and participates in your webinar in that time. This means you can consider the interactions both in terms of their quantity and quality. The more quality contact the lead engages in (through webinars, e-books and interactive courses), the more important the lead.

Verbal and non-verbal data are assessed through lead scoring matrices which are available in the more professional marketing automation systems.

The graph below shows the lead’s significance depending on the position they hold on the horizontal axis, and the degree in which the lead is engaged, based on their behaviours and content consumption on the vertical axis. In line with this example, the most valuable leads are directors and members of management boards who test your software and participate in your webinars.

Lead scoring graph


Clearly, it’s worth considering what you need the data for before you start collecting and processing them on a mass scale. Just like fresh flowers, data get old quickly and become redundant.

It’s worth treating the implementation of a marketing automation or CRM system as an opportunity to organise the data and specify the procedures governing their collection and processing.

You must also ensure that the quality of the data doesn’t suffer once all data are in good order. I hope you’ll find some of the good practices which I’ve shared here useful. However, it’s always best to ask a database specialist for more advice. Larger databases can’t be managed without expert knowledge.

Finally, you have to keep in mind that good marketing automation systems also allow for collecting data which are not provided directly. You’ll achieve the best results by complementing the data with details on the lead’s behaviours, such as the amount and the nature of the contents they consume on your website. It’s the ability to analyse a person’s DBL precisely that makes marketing automation systems so efficient in generating high-quality leads.


Digital and B2B Marketing practitioner. Helps B2B Companies to fully utilize the potential of digital sales and marketing tools. Previously Head of B2B Marketing team at Samsung Electronics Poland before holding several marketing and channel development roles in Microsoft Poland.

CEO and Head of Consulting at Grow Consulting.

Over 15 years of experience in building marketing and leads generation strategies in B2B market. Pioneer of Marketing Automation deployment in B2B space.Effie Award winner in B2B category. Blogger, contributor to professional media : Marketerplus, Brief, Marketing w Praktyce. Author of the book “ABC of B2B Marketing” first handbook for B2B Marketers in Poland.

Contact us if you plan to develop or rebuild your digital marketing strategy in your organization.

Contact us