The IBM System/360 is greeted by fear and confusion when it arrives at the SC&P offices on Madison Avenue in the final season of AMC’s Mad Men.

Some employees of the agency infer their own impending obsolescence, or in the case of the copywriter Michael Ginsburg – having his mind infiltrated by the machine. While on the nose, it’s not an absurd depiction of the zeitgeist. The real net stock of computers and peripheral equipment increased by more than 20% in each year of the 1970s. By 1980, that figure was hovering around 40%.

Mainframe builders’ pitch wasn’t actually to have this technology make decisions independently. Rather, it could provide graphical representations of the numbers, supplementing the intuition of business leaders on an ad hoc basis so they wouldn’t need to commission slow and expensive analysts’ reports.

Today it’s unthinkable that an executive would walk into a boardroom and announce their plans without the endorsement of rich, recent data.

Marketing has perhaps been the slowest department to change, not least because it was traditionally the domain of humanities graduates and the kind of creatives that call the shots in Mad Men’s early seasons. But in the last decade, data has gone from merely validating and optimizing marketing tactics, to being utterly prerequisite to any significant decision. 

Logisticians and comptrollers announced their thirst for data, and the technology market responded by opening the faucet. CMOs wouldn’t have predicted they would be the ones drowning in it unless they’d taken Mad Men’s hapless Ginsberg seriously. “I can’t turn off the transmissions” he says, just as we now can’t look away from social media. “They're beaming 'em right into my head.”

So is it time to put the Paris Review aside and approach marketing as an engineering problem?

Of course not.

The delivery side of marketing is still fundamentally composed of human connection, even now ‘which people to connect with,’ ‘where,’ and ‘when’ are scientific considerations. The new formula for marketing excellence is a collaboration, the chief technology challenge being the extraction of relevant and actionable insight from the ocean of data. For their part, creatives must respond with greater discipline, iterating on hypotheses and eschewing derivative content. In large enterprises, that looks like cross-functional teams, but it’s not always feasible to allocate engineering resources to marketing problems.

That’s where software comes in. 

Marketing teams must leverage big data 

Here are five ways a creative marketing team can up their game by using smart tools to leverage the power of big data.

1. Data-driven PR

The scientific skill: Conducting original, high-quality research and/or consolidating second-hand research.

The tools: Smart surveys, government and other open source databases, Google Sheets, data visualization platforms, customer data platforms

The opportunity: A single story in tier 1 press outlet (or a slew of mentions in relevant local or industry press) can be explosive for a brand. But most companies throw resources at PR with nothing to show for it. If you’re not important, the media doesn’t want to make you important.

The secret to escaping this sisyphean predicament is that journalists love original research. SaaS companies can simply anonymize their own user data and give the press a compelling story to tell. Spotify does this on a huge scale. Even companies with a drought of internal data can simply go online and discover interesting patterns by pulling together a few third party open databases.

2. Deep personalization

The scientific skill: Keeping data from multiple sources as clean as possible and building API bridges between channels to get a holistic picture of customer behavior.

The tools: Data management platforms, web content management systems (alternatively, ‘sales intelligence’ software, ‘marketing automation’ software, and ‘personalization engines’ - this is such a huge market segment that vendors bundle different functionality in a bunch of different ways.)

The opportunity: Deep personalization is central to Facebook’s pitch to advertisers, and Netflix reportedly uses machine learning to personalize the hell out of their viewers' experience.

Obviously the more granular information you have about your customers, the more deeply you can target your marketing efforts. The possibilities are almost endless, but the central idea creative marketers should consider is that more audience segments means fewer people in each segment. In other words, creatives get to ‘perform at a series of intimate gatherings’ rather than trying to win over a stadium. Naturally, that means more conversions.

More than your average newsletter.

Every Thursday, we spill hot takes, insider knowledge, and news recaps straight to your inbox. Subscribe here

3. Unlocking platform cheat codes

The scientific skill: Interrogating web platforms, then conducting statistical analysis on the output.

The tools: API marketplaces, web scrapers, and statistical analysis software – especially the more user-friendly iterations of these tools.

The opportunity: There’s already plenty of generic research out there about ‘the best time to publish’ on various platforms, and some really awesome tools like ‘Later for Reddit.’ But for a real competitive advantage, you need to do some independent analysis and figure out the optimal tactics for your own goals. The inclusion of hashtags, word count, and ideal ad spend per platform are all questions that differ between organizations.

There are also plenty of unanswered questions about the algorithms behind each of the biggest social media platforms. Carrying out automated research at scale can unlock opportunities you might never have imagined.

4. Predict upcoming trends

The scientific skill: Statistical modelling, machine learning, gathering and analysing search intent and user behavior data.

The tools: Google Trends, marketing analytics

The opportunities: Predictive analytics lets marketers see the future. That’s important for two reasons.

First off, being able to align marketing activities to top level business objectives is what earns boardroom seats. Getting there means translating fuzzy digital metrics like click through rate and social engagement into accurate profit predictions. It’s a road paved with tricky problems like detailed lead scoring, lifetime value and churn rate calculations, and getting those numbers right requires a high volume of quality inputs.

On the opposite side of the coin is the ability to get out of the building entirely and make macro predictions about product-market fit. Creative leaders have long been expected to have their ear to the ground and understand cultural and social trends. But the stakes are higher now. Discipline has replaced intuition. 

5. Outgun huge companies on Google Search results

The scientific skill: Understand the signals Google’s algorithms look for to determine quality, relevancy, and trustworthiness on the web.

The tools: SEO software 

The opportunity: Earning organic traffic from Google search results is probably the fastest route to marketing ROI, but it hasn’t traditionally been the most predictable.

That’s because there’s an enormous number of signals that combine to decide what ranks for a given query. Attempts to game the system with simple hacks have consistently been met with penalties, and today Google updates its algorithms more than ever. 

A quick recap

There’s a good reason the demand for data scientists is booming. Strategists as far back as Sun Tzu understood that better information means better decisions, and today’s world is flooded with unstructured and inconsistently represented information.

Almost every business function has established its methodology for diving into this mess and retrieving useful insight, but marketing remains largely haphazard in utilizing data for most use cases - personalization being the notable exception.

When data scientists apply their skills to marketing challenges, they’re able to decipher what’s going on behind events like Google’s BERT update, givingcontent writers and strategists concrete actions to earn reliable search rankings. By aligning granular digital marketing metrics to accurately forecast revenue, marketing leaders make themselves indispensable to their businesses.

Competitive advantages are ripe for the picking by anyone with the know-how to interrogate social platforms and open-source datasets. Finally, data is sometimes even more valuable to audiences and publishers than it is to the organizations that collect it. Turning data into a story is one of the most powerful moves marketers can make.

All of these are ways marketers can be more data-driven, beyond the more obvious personalization use cases. Creatives with no STEM background don’t need to fear the rise of the machine; they can absolutely execute these ideas with the help of smart tools and a little discipline. 

Ready to become more data-driven? Make sure your marketing strategy includes finding the right big data analytics software to help get you on track.