Lyft’s head of marketing science recently said that the only thing he believes in is experimentation. He measures current reality, perturbs it with creative and channels and spend, and measures the effect on metrics that matter. In short, he applies the scientific method to marketing.
But what is scientific marketing?
And how do you achieve it?
Most companies are familiar with BI, or business intelligence: using a company’s data to identify opportunities, manage risk, and analyze results. As marketing, the CMO, and the emerging Chief Growth Officer become more prominent, companies also need to engage in marketing intelligence … and their marketing intelligence stack.
Growth marketers, who are increasingly marketing scientists, need a marketing intelligence platform to measure the impact of experiments. The primary function of this stack is to provide insights for growth by connecting effort with outcome at granular and aggregate levels.
When marketers achieve this, they achieve a number of incredibly important things, all of which enable scientific marketing.
One: unified marketing data
First, when marketers unify their data, they get unprecedented data visibility at scale.
Doing so, however, requires a marketing intelligence platform that combines data from all marketing sources: email, web, mobile, app stores, and e-commerce systems, just to name a few. But it also includes campaign data and results data from paid advertising channels, including impressions, clicks/taps, and costs. Conversion data is critical too: app installs, purchases, custom events, revenue, and more.
Marketing science does not exist without unified marketing data, and data needs to be visible to be useful.
Considering doing this manually? Good luck.
The best BI analysts doing this manually spend 20-30 hours each week on their efforts, and still constantly run into normalization, standardization, and accuracy issues.
Two: intelligent insights
Second, marketers achieve the ability to answer questions they’ve never known how to ask before.
Processing all internal, external, and partner data in a central place helps marketers achieve full knowledge of the returns on their investments at a granular level. That means understanding both the full costs of each marketing activity, whether paid or organic, and the specific results those activities achieved. When marketers have this at a granular level, they have data on their results by overall campaign, by ad sets/groupings, and even by individual pieces of creative, or grouped sets of creative around similar themes.
Plus, with a marketing intelligence platform (MIP), marketers can slice and dice their data like never before — like calculating customer acquisition cost by creative asset type like video, banner, or playable — or even a specific piece of creative.
In addition, because you’re unifying all of your data — including campaign set-up data — brands can connect metadata that otherwise would never compute.
One example: dimensions that might exist just in your internal customer segmentation models. For instance, a prospect might convert into a customer via a campaign targeted at luxury buyers, but then actually buy a product focused more on utility. Traditional marketing reporting systems would report success in the luxury campaign and failure in the utility campaign, while the purchase records will show success in utility and failure in luxury.
Connecting disparate datasets is the key that untangles that Gordian knot.
And knowing — not just guessing — which customers respond to which messages helps brands communicate in smarter ways to customers in order to maximize profitability. And understanding true ROI for groups of campaigns helps brands know what really works.
In addition, marketers using a marketing intelligence platform know how their customers spider over devices and platforms towards the point of purchase.
With a MIP, marketers have a global perspective on their customers’ journey across devices and platforms. Marketers can see when customers who onboarded via a Google search ad re-engaged via a Facebook ad. And they can see when a customer begins with purchases online and moves to buying in-app.
This is critical, because if a customer sees an ad on mobile but converts on desktop, brands might think the ad campaign was a failure — pausing spend — and determine that organic is strong … only to see it mysteriously decrease after stopping the mobile ad campaign. All of this feeds into key business growth metrics such as ROI and true customer acquisition cost.
Absent that data, marketers don’t have accurate insights, and may make incorrect future budget allocation decisions.
Third, marketers gain the ability to automate tedious and time-consuming tasks.
Ingesting data, normalizing and standardizing it, and deriving actionable insights requires automation to be effective at scale. Sending data to partners for mediation, reconciliation, attribution, enrichment, and exporting it for additional internal analysis both need to be automatable to free up marketers for high-leverage strategic work.
Fraud requires intelligent automation to abate, marketers need configurable alerts when campaigns diverge from optimal parameters, and at the highest levels, campaign bids and buys need to be optimized by intelligent, learning systems.
Automation increases accuracy and enables marketers to focus on growth.
Real insights from scientific marketing drive growth
Once you have all the marketing intelligence platform components operating, you achieve real insights that change your strategy and decision making. A MIP can transform the way you think, analyze your past actions, recommend new courses of action, and even predict outcomes of potential new marketing campaigns.
This is marketing science in action, and it enables unprecedented growth in a hyper-competitive world.
And that drives value creation, which is ultimately what marketers are employed to do.