As marketers, we are constantly encouraged to be 'data-driven'. For starters, we are expected to keep a close eye on acquisition costs, track funnel metrics, and keep tabs on conversion rates.

On top of that, we use audience data for segments, behavioural data for optimization, and customer feedback data to find and solve customer experience issues.

With all of these various ways of using data, though, how can we be sure we're doing our jobs correctly? That is, are we truly marketing in a 'data-driven' fashion?

To find out, Econsultancy recently invited dozens of data-driven marketing experts to Digital Cream Sydney to discuss what they do, the problems they face, and how they overcome obstacles. Through roundtable discussions led by Danielle Grant, senior marketing manager at PayPal, best practices emerged which attendees agreed were key to the success of data-driven marketing.

Before we start, though, we'd like to let you know about an Advanced Mastering Analytics training course we are offering on November 8th in Singapore. You can find out more and reserve your spot here.

The keys to data-driven success

So what do successful data-driven marketers do that others may not?  

1) They obsess about data management

Participants said that the first step toward data-driven success is to devise a data management strategy, and a good data strategy starts with a data audit.

The audit does not need to be complicated; it may simply consist of writing down the answers to a few key questions: 

  • What data do you have? Make a list of everything you use and identify gaps.
  • Where is it housed? Ensure you understand where data originates and 'lives'.
  • Who currently has permissions to access it? Is it everyone who needs it? 

Then, once data assets are identified, marketers should also establish a data governance policy. The policy should state who is responsible for maintenance, who has editing rights, and what data is available to everyone. Ideally, the policy should state who the data 'owner' is and avoid having too many additional stakeholders involved.

Next, those with editing rights to the data should keep the shared data clean. As one participant said, "if something bad goes into your data, then you should expect bad stuff coming out". 

Finally, there should be a common data language which allows the organisation to unify the data. Bring in experts if need be, one attendee suggested, as the benefits of having well-integrated systems far outweighs the cost of a temporary consultant.

2) They test hypotheses to get insights

Another distinguishing feature of data-driven marketers is that they avoid hacking the data to prove a pre-determined point. With the amount of data available now, one attendee pointed out, you can prove almost anything.

Instead, experts approach their data with a hypothesis of what they believe to be true and use data and analytics techniques to either prove or disprove their hunch.  

Doing so also avoids 'analysis paralysis' and ensures that insights are gathered in a reasonable amount of time.

3) They stick to one attribution model

One of the main topics discussed by delegates on the day was marketing attribution models. Specifically, which attribution model is the 'right' model?

Instead of giving a quick answer, though, the subject matter experts asked another question, "what story are you trying to tell?" If marketers wanted to identify the superiority of a particular channel, then focusing on its place in the acquisition process should be a priority. If it was that acquisition happens across many channels, then a more complicated attribution model should be considered.

Also, data-driven marketers should consider a channel's impact across the customer funnel rather than what gets people to the site.

Whatever is chosen, though, a single attribution model should be used consistently across campaigns. Changing attribution models makes it difficult to compare data across various marketing activities and confuses stakeholders. 

A single model also makes the inevitable internal dialogues on the subject much more productive.

4) They insist on transparent agency relationships

Another habit of successful data-driven marketers identified by participants is that they maintain a transparent relationship with agencies and avoid being kept at a distance from campaign data.

While many acknowledged that there is often pressure to get campaigns out the door quickly, data-driven marketers must be strong and insist on knowing: 

  • how audience data is acquired and used,
  • how well each target segment is performing, and
  • what steps are being taken to optimize performance.

This is not only to 'keep an eye' on agencies, but, as one attendee described, "if there is shared ownership of information, there can be a shared ownership of the outcome, be it positive or negative."

The north star

A final point raised was that whatever the approach, data-driven marketers should have the best outcome for the customer as their 'north star'. They should avoid using tricks or anything remotely unethical in order to improve numbers.

Doing so, everyone agreed, will deliver the business results that people are looking for from data-driven marketing.

A word of thanks

Econsultancy would like to thank all of the marketers who participated on the day and especially our Data-Driven Marketing table moderator, Danielle Grant, senior marketing manager at PayPal.

We hope to see you all at future Sydney Econsultancy events!

Jeff Rajeck

Published 17 October, 2017 by Jeff Rajeck

Jeff Rajeck is the APAC Research Analyst for Econsultancy . You can follow him on Twitter or connect via LinkedIn.  

193 more posts from this author

You might be interested in

Comments (1)

Zsolt Keszler

Zsolt Keszler, Founder at Dee-Wine

Great article, Jeff, thank you. I would like to add a couple hints about Data Governance. We all know the "sh*t goes it, sh*t comes out" proverb - yet most companies do not understand the relevance of keeping data constantly 'clean'.

1) Clean starts with answers to the questions of what is core data and what is nice to have, as well as defining what quality levels for each of these the company would like to achieve. Getting an agreement on these rather simple questions is a difficult job: as by many companies there is no defined owner of the data, therefore fostering a single opinion about this question is a hurdle many cannot pass.

2) One (or first) time cleansing can bring a lot of results but at the end it is like a diet - you cannot do it on one day only, it has to be a lifestyle. And that is exactly where new problems arise:

a) you need to define who "owns" the data - it is the owner's responsibility to keep it clean.

b) defined ownership will not automatically bring results - telling sales that it is up to them to keep customer contact e-mail addresses up-to-date will not get the job done.

c) motivating responsible colleagues is key - data quality should be part of the company's performance metrics system.

Sadly, many companies forgo a comprehensive approach to data governance - accumulating excessive marketing costs and poorly executed campaigns.

about 1 month ago

Comment
No-profile-pic
Save or Cancel
Daily_pulse_signup_wide

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.