How Data-Driven Marketing Campaigns Work

A solid yellow background with a tablet in the center. The tablet had various objects on top of it, such as a megaphone, magnifying glass, clipboard, and large bar chart. The objects could all be used by marketers analyzing big data to put together a data-driven marketing campaign.
A solid yellow background with a tablet in the center. The tablet had various objects on top of it, such as a megaphone, magnifying glass, clipboard, and large bar chart. The objects could all be used by marketers analyzing big data to put together a data-driven marketing campaign.

Marketing – or rather its predecessors advertising and sales promotion – moved from the gut-feel used in the 17th-19th centuries to something more structured and insightful when basic market research made its appearance in the 1920s.

But the data that flowed from that, while useful, wasn’t extensive because of the labour-intensive effort involved. An infant form of data-driven marketing had arrived. Then came more formalised market research and the emergence of George Gallup – he of the famous Gallup poll statistical method of measuring public opinion – and others of his ilk who made it into an early form of marketing data science.

Thanks to ongoing advances in market research and its analysis, Data was getting bigger, better, more accurate and being used more frequently by advertising agencies and marketers to make decisions and to justify what they were spending and why.

The advent of the public Internet in 1993, plus rapid advances in computerisation and digital technology, changed the game again. The age of big data and data science had arrived. Today, around 4.7 billion people use the Internet and 4.2 billion are on social media. An incredible 2.5 quintillion bytes of data is being created every day in 2021 (a quintillion is a million times a trillion, or a 1 with 18 zeros behind it) and that volume of data is expected to double every two years.

It equates to a lot of potential marketing data and consumer insights for marketers. Providing, of course, data science can sort the wheat from the chaff and an analytics expert can determine what’s useful for particular marketing strategies. There’s a difference between big data and useful data, and top-quality data science can make big data very ‘useful’ indeed for marketing purposes. In this blog, we'll explore how marketing teams can parse through the data they are collecting to inform how to increase the effectiveness of marketing campaigns.

Table of Contents

Data-driven marketing: The way of the future

Enter data-driven marketing as the way of the future. When marketing joins with data science to leverage big data effectively, marketers will likely achieve great campaign results. What does a data-driven marketing strategy entail? Let’s examine some definitions.

Defining a data-driven strategy

AT Internet, one of the French Internet industry’s pioneers, defines a data-driven strategy as thus: “A data-driven approach enables companies to examine and organise their data with the goal of better serving their customers and consumers. By using data to drive its actions, an organisation can contextualise and/or personalise its messaging to its prospects and customers for a more customer-centric approach.” The process, AT Internet adds, can also be termed ‘data analytics’ or even ‘data democratisation’ in that “the process of democratising data means making data accessible to as many people as possible within a company”.

DialogTech, a Chicago-based business specialising in analysing and optimising phone data for businesses, gives its definition as: “A marketing strategy that uses data – acquired through consumer interactions and third parties – to gain a better view into consumers’ motivations, preferences, and behaviours. Marketers then create personalized messaging and experiences that deliver the highest possible return on investment (ROI).”

People also ask:

What is data-driven marketing?

It is the use of data obtained from various sources, which is analysed and acted upon to give direction to an organisation’s marketing efforts. The data informs the plan.

Why is data-driven marketing important?

Because it uses the widest possible range of up-to-date information to drive the strategy. It doesn’t rely on assumptions, limited knowledge, or out-of-date research samples.

What is the meaning of data-driven?

These are decisions or actions that are guided by the gathering and analysis of data.

How do you do data-driven marketing?

The key actions are:

  1. Define your objectives
  2. Gather and analyse the data
  3. Devise a marketing strategy based on data insights
  4. Implement the plan
  5. Analyse the results

Why the uptake of data science has not been universal

That’s the theory and principles behind data-focused campaigns using data science. But do all businesses use it and what are the consequences if they don’t embrace data science as a way to drive successful campaigns?

Possibly because of a lack of data scientists, plus the daunting nature and cost of analysing and effectively using massive volumes of marketing data, there is still a surprising lack of uptake. One recent survey on the data scientist shortage by SnapLogic found that 73% of UK firms said they lacked the talent to complete artificial intelligence, machine learning and data science initiatives. Quanthub, a data skills platform, tracks the data scientist shortage and found that in 2020 there was a shortage of 250 000 data scientists.

According to Forbes business magazine, nearly 36% of companies don’t use all the big data they possess and 47% are only planning to implement a data analysis tool in the future. The same article quotes the Meaningful Brands 2019 report by Havas Group, which projects that 81% of European brands could go extinct because they don’t create relevant content and can’t offer personalised discounts.

Only 19% of companies conduct customer behaviour analysis, segment their audiences correctly and personalise their campaign offers, Havas found. Taking note of this, Forbes points out that using data science to produce such insights would enable those organisations to “know what customers want and when they want it” and that the cost of marketing teams ignoring this information might be a “fast exit” from business.

Vlad Flaks, CEO of OWOX BI, an all-in-one marketing analytics platform, observes that companies that don’t care about analytics and data science also risk wasting their advertising budgets. This is because their marketing personnel know too little about click-through rates, conversion rates, and other key metrics driving the customer journey and customer buy-in to campaigns. “How can you find out what advertising is efficient, how much money to spend on each segment, and what goods sell better together?” he asks.

Amir Orad, CEO of Israel-based business intelligence company Sisense, told a technology conference in New York in 2019 that companies that fail to implement data science to properly utilise the big data that they already have will likely go out of business.

overhead photo of a laptop and coffee mug on a pink surface

Developing a data-driven campaign strategy

So the effective uptake of marketing data to drive campaigns still seems somewhat patchy – thanks to unconvinced CMOs, slashed marketing budgets, limited human resources, complex data science and high-salaried data scientists. Therefore, it is safe to assume that organisations that can implement a successful data-focused marketing strategy should have a notable competitive advantage.

If you're wondering 'how do you do data-driven marketing?', here are some key data-driven trends for marketers to be aware of when developing a data-focused campaign strategy.

Shift to first-party data

Third-party data restrictions are growing because of the EU General Data Protection Regulation (GDPR) and other ever-increasing consumer data-privacy regulations worldwide. Consequently first-party data – accurate big data that organisations have captured from their own customers and analysed using data science – will become more important for campaigns in the future.  

Increased uptake of artificial intelligence and machine learning

Another trend we're seeing is an increased focus on artificial intelligence and machine learning in data science use cases such as predictive analytics and smart chatbots. According to the website Towards Data Science, bot analytics and chatbot development strategies will, in the future, “have a role in enhancing or even restructuring business processes”. Chatbots magazine sees similar trends.

Increased personalisation of marketing efforts

This is already one of the cornerstones of successful data-driven campaigns, but there is ever-more hyper-personalisation as consumers indicate that this is what they want. A 2017 study found 80% of respondents are more likely to do business with a company if it offers personalized experiences and 90% indicated that they find personalization appealing. Their figures are likely to have increased in the intervening years.

The continued rise of cross-channel marketing

The number of touchpoints that people use to interact with brands is continually increasing – from in-store visits to websites, social media channels, mobile phones, tablets and who-knows-what MarTech in the future. This means more opportunity for companies to gather big data and key insights, but also more ways in which the campaign strategy must be fine-tuned to be channel-appropriate.

More emphasis on data security and having the right data policies in place

Given the seemingly inevitable demise of third-party data due to legislative and other pressures, getting first-party big data from willing consumers is vital. But they need to be convinced your business is ethical in its use of personal data and has the appropriate data-security protocols in place.

Break down of silos

We're also seeing heightened collaboration between those who create the campaigns and the data analysts and data scientists behind the scenes. A siloed approach that separates the data scientists and marketing teams is counter-productive. A fully informed relationship that maximises understanding of the customer journey is far more beneficial. So too are KPIs that make the marketing team more data-centric. To break down these silos, larger brands are gradually introducing more marketing data analyst roles.

Cartoon sketch by marketoonist showing the silo mentality of office. Cartoon characters look beyond a castle built in an office while others look on stating "out silo mentality may be getting out of hand".

Are there any of these trends that stand out above all others as being vital to big data marketing in 2021?

Personalised marketing

All are extremely important, but arguably personalisation is the most crucial to successful campaigns. As even the most basic online course in marketing will teach you, a one-to-one relationship with the potential customer is the Holy Grail for all marketers.

The only reason that one-size-fits-all ‘spray-and-pray’ marketing activities existed in the past is because there wasn’t a viable way to narrowcast the message to a target audience of one, except in highly specific circumstances. B2B marketers have generally been able to do this more easily, simply because of their highly defined audience. But now modern technology and big data are, indeed, making hyper-personalised messaging viable – whether it be via targeted website interactions, email campaigns, social media touchpoints, serving great content to highly specific audiences, or through other targeted mechanisms.

“Essentially, data-driven marketing expands upon the 25-year-old marketing philosophy first voiced by Don Peppers and Martha Rogers; their then-revolutionary personalized, ‘one-to-one’ marketing philosophy. Focusing on each prospect and customer using data-driven marketing has elevated the one-to-one concept to new heights that had to wait for technology to advance,” observes DemandJump, a US-based marketing software company, in a blog post.

Cross-channel marketing

Cross-channel marketing is another that’s vital, simply because the number of channels – and therefore touchpoints – where consumers expect to be engaged during campaigns continues to proliferate. TikTok, for example, was a largely unknown social media platform only a few years ago. So, a successful modern business needs to be everywhere and communicating in different ways.

A diversified content strategy

“You should also be diversifying your content strategy with a wide array of types in order to reach the highest potential audience on more channels,” advises Social Media Week. “This will also help your project build authority, which is critical in retention, churn reduction and overall brand loyalty.

“Yes, have a blog. It is a great SEO tool and you should already know how to be targeting keywords by now. But in addition to this, make videos, create infographics, launch a podcast, guest on other podcasts and YouTube channels, embrace networks like TikTok and Instagram, and contribute to other blogs. You should be producing diverse content constantly to be able to market to different channels.”

Examples of successful data-driven campaigns

The following are a selection of effective campaigns that used data science and a data-driven strategy to maximum effect.

IBM Watson marketing campaign

The winner of the UK’s Campaign Tech Award 2020 for the Best Use of Data/Insight was IBM Watson, the computer system capable of answering questions posed in natural language.

As the publication Campaign Live explained, to celebrate its 30-year association with Wimbledon tennis, IBM’s marketers wanted to showcase the capabilities of the Watson’s artificial intelligence during the 2019 tournament. But the question for the marketing team was how to reach people when most matches are played on weekdays and accessing live updates is difficult for Londoners travelling on the Tube?

The answer was a complex and surprising campaign that used AI to create highlight packages to broadcast on digital out-of-home. Watson edited action from hours of footage using unstructured data (such as crowd noise and player emotions like fist-pumping) and created a highlight clip within two minutes. Every point on all six show courts had to be analysed simultaneously to show the clips on screens across London.

IBM’s analytics showed a steep increase in Top-of-Mind Preference scores: from 8% before Wimbledon to 21% afterward. They also scored a significant increase in Social AI share of voice during the campaign, from 14% to 22%.

According to Campaign Live, the award was given in recognition of innovative marketing activity rooted in data-driven customer insight.

‘Politics of your diet’ campaign

Grubhub, the American online food ordering and delivery platform, has a strategy of creating interesting content from the data that it harvests from its customer interactions. It then partners with publishers, which use these data-generated stories to create native advertising that feels natural.

Triggerbee, the Swedish-based analytics and automation company in a blog, explains: To expand its partnerships to publishers focused on politics, GrubHub analysed how the food choices of their users correlated with their political leanings and then connected those to congressional districts around the US held either by the Republicans or Democrats. 

The creative concept was to “test the politics of your diet”, which generated some interesting stories about how, for example, Democrats are more likely to order veggie burgers and Republicans hamburgers.

“[The] results were pretty interesting and potentially made it easier for the company to [provide] their data to secure new partnerships with political publishers,” Triggerbee commented. “It’s easy to see how this same technique could be applied to other industries, allowing them to expand their native advertising base and grow their revenue.”

A hand holding a phone with the GrubHub app up, in front of a Just Eat sign

How big data informed the Lion King campaign

An example of how marketing data can inform your campaigns is The Lion King movie case study. Meltwater, as a data-oriented company, provided a solution to movie execs and their marketers looking to ensure that the 2019 remake of the 1994 classic would not follow the downward spiral of public ambivalence and apathy that bedevils so many remakes. The insights gained by Meltwater from social media enabled the film’s marketers to fine-tune their strategy.

For example, by analysing over 5,000 mentions in the news around the world and more than 2-million social mentions just in November 2018, the importance of promoting the return of James Earl Jones to the remake (he also starred as a key character in the original) was clearly understood.

Fans were ecstatic when they heard the booming ubiquitous baritone voice of Jones (who played the character Mufasa) narrate the trailer for the remake and they soaked in the nostalgia of the reanimated classic. In comparison, global megastar Beyonce’s involvement in the project had only a relatively small number of social mentions in the same timeframe. This interesting insight revealed that even the presence of an undoubtedly talented superstar such as Beyonce could not match the level of nostalgia that needed to be emphasised and harnessed when promoting the new version of the movie.

If you're interested in how Meltwater media intelligence can be used to inform your own data-driven marketing strategy,

Other examples of effectively using data analysis

Other examples of effective use of marketing data analysis include:

  1. The Weather Channel in the US analyses the geographic location of the millions of visitors to its website to sell highly targeted advertising opportunities to advertisers. For example, anti-fizz hair products to site visitors from humid climates, or skin moisturisers to people living in arid parts of the country.
  2. Automakers analyse the data that high-tech modern cars send back constantly via the cloud, to tailor personalised new product offers to those customers. For example: accessories, specialist driver training, or even new vehicles better suited to their lifestyle.
  3. YouTube, Netflix and other video-streaming platforms use constant analysis of past customer preference data to suggest other videos, movies and music that are likely to appeal to that customer.
  4. GreenPal, a lawn care/lawn mowing company, analysed census data in its home city and created an online ad campaign targeted at an up-and-coming suburb where people wanted lawn care but were price sensitive. By emphasising that GreenPal could offer a cost-effective service at a specified price, the click-through rate increased by 200% and the on-page conversion rose by 30%.
An image with bright coloured boxes and lines, representing data storytelling

Data-driven marketing campaigns: In summary

To quote Forbes columnist Vlad Flaks again, effective use of data analysis gives you a competitive edge when you plan campaigns.

“Experience and even tested hypotheses aren’t a sufficient basis for decision-making,” he says. “Your opinion or what you’ve done for years can completely differ from what your customers want today and are ready to pay for. Personalization and smart interaction design are the two main features of market leaders. Your task is to know everything about your customers and forecast their next wishes and purchases.”

As marketing software company DemandJump observes: “As tools that leverage AI to combine and analyse various sources of data become more widely adopted, you can expect data-driven marketing to become the norm across all industries. There's no better time than today to start using data-driven marketing principles for your business.”

In short, if you’re looking to achieve extraordinary strategic business growth, data-driven marketing may be the answer. Gather big data; use machine learning; employ data scientists; embrace data science; utilise analytics. Success could be yours!

Do you have any tips around how data-driven campaigns work, optimising marketing data analytics, or perhaps you have a great data marketing example you’d like to share with us? We’d love to hear from you, so tweet us @Meltwater. Alternatively, if you want to learn more about using media data to guide your strategy, fill out the form below!

Loading...