Who would have imagined that digital marketers will still face challenges using their data to improve market performance in 2020? Over the last couple of years, some large organizations invested heavily in data-centered marketing. The results of data-driven marketing strategies reward organizations that implement the concepts correctly. This highlights the marketing data challenges facing organizations with ineffective data management strategies or who fail to capture actionable insights from their data.
By gathering and leveraging consumer data, internal sales and marketing data, as well as data from Google Analytics, digital marketers gain powerful insights critical to creating targeted campaigns that maximize their conversion rate and return on advertising spend (ROAS).
As many firms discover early in their analytics journey, that big data is not an end, but rather the beginning of the marketing challenges they face in using data effectively. For more info on how to empower your data, go to the website.
Before a company translates data sets from raw statistical figures into valuable insights, they must summarize, categorize, and analyze the data. These task represent the biggest marketing data management challenges. Despite sitting on mountains of data, managing that data is a Herculean task.
Outcomes of good marketing data management
Firms with good marketing data management produce superior returns for their brands. Examples of benefits achieved with good data management include (this list isn’t comprehensive, but provides a valid rationale for spending resources to optimize your marketing data management):
- Personalization. We know most consumers respond to offers tailored to their needs and decision-making style. Companies like Amazon and Netflix perfected suggestion algorithms resulting in increased sales and higher customer satisfaction, for instance.
- Better management of the brand. Data provides insights to guide all 4 Ps in marketing. Understanding how price impacts profitability, for instance, guides pricing strategy. Understanding which ads, on which platforms, on what schedule produce the highest ROAS helps set advertising strategy, is another example where data guides insights.
- Forecasting, which begins the budgeting process relies on information about customers, prior sales, and results from marketing strategies.
Characteristics of marketing data
Of course, the following holds for most data acquired by the organization either through outside sources or internal metrics, not just marketing data.
We talk about the 3 Vs of big data. I’ve added my own V, resulting in 4 Vs of big data, as follows:
Big data includes a variety of data from various sources. Data may reflect structured data (numeric data) or unstructured data (words, symbols, images, etc). IBM estimates that 80% of the data available exists as unstructured data that’s hard to interpret and doesn’t readily convert to insights. One strategy for handling unstructured data is to structure it. For instance, a number of tools help convert social media data into sentiment data reflecting positive, negative, and neutral attitudes expressed with words or images. These natural language processing programs (NLP), using AI (artificial intelligence) and ML (machine learning), sentiment analysis represents a tool to monitor sentiment over time, yet, even this simply reduction of words to numbers represents serious errors. These errors result because natural language contains an ambiguity that, without non-verbal cues, is challenging for even humans to interpret let along training a computer to parse the language into buckets.
Data comes at you pretty fast. In a world where real-time data comes from many sources, it’s challenging to analyze data on the fly so as to make better decisions. Take a look at this image of the NASCAR command center that guides staff to focus on certain video feeds for their live broadcasts.
Notice the dashboard contains elements such as social media mentions of drivers, cars, and NASCAR, itself displayed as real-time graphs from each platform. Each camera feed also shows on the dashboard along with other data used to create the final TV feed.
The company uses various tools and a large staff to manage the data coming into the command center.
Just looking at the name, big data, should clue you in that we’re talking about a high volume of data. Again, looking at the NASCAR command center, you see the volume of data coming in every moment.
The volume and velocity of data mean you need a computer with appropriate tools to process the data; no human or team of humans can analyze data with these characteristics. Yet, computers are terrible at gleaning insights. For that, you need skilled people who take data from dashboards, manipulate the data displays to highlight relationships between data elements, and combine all of this with experience to develop actionable insights.
This characteristic of big data is of my own invention based on decades of data analysis, but every analyst recognizes that data is inherently “dirty” and requires cleaning before insights are reasonable. Data cleaning involves removing data based on the totality of data. For instance, data cleaning might remove outliers because they represent errors in data coding or categorization, such as transposing numbers in columns. And, don’t believe the myth that computers don’t make errors. Yes, they follow their coding, but the code might represent a mistake.
Hence, data requires a cleaning protocol to ensure the veracity of the data.
Marketing data management challenges
As noted by IBM, numerous data management challenges await businesses in implementing analytics to improve decision-making and market performance. The top 3 challenges are:
- Data security as hacking leaves a firm open to lawsuits and data breaches potentially uncover proprietary data damaging to the company or its brands.
- Data comes from multiple sources and may contain conflicting information. For instance, data may represent different views of the same situation (my Google Analytics data and my Google Ads data always seem a little off, with no explanation from my queries to Google). Data likely represent multiple formats to conform to the needs of the original data source, hence restructuring that data using tools like SQL and Python may result in errors if not coded properly. Further, data from various sources might reflect different formats, for instance, you might end up with a jumble of structured and unstructured data requiring additional processing.
- Translating data into insights. Of course, this is the elephant in the room that stymies even companies with great marketing data management as developing insights from data requires not only excellent analysis skills but intuition. The best data analysts combine the art and science of analytics in the proper proportions to develop actionable insights to guide decision-making.
Marketing data management is challenging for the following reasons:
The most obvious reason marketers struggle to connect the dots between their data sets is that the collated data cuts across a broad diversity of metrics. When you collect vast pools of unrelated data, correlating that data to form insights is easier said than done.
For instance, the attribution model that’s employed to keep track of offline campaigns usually harvests aggregate data. On the other hand, an offline campaign attribution model usually collects person-level data. At the end of the day, it becomes difficult for data managers to marry their diverse data sources.
When marketers sit down to correlate consumers’ engagement with their campaigns against the teams’ goals, it’s hard to normalize different campaigns to arrive at a single view of the consumers.
Today, marketing teams are overloaded with tons and tons of data, to an extent that they can’t dig out the valuable insights. This over-abundance of data takes a dangerous toll on marketers. Unfortunately, without the correct insights, metrics are meaningless. One Gartner study into the pressing challenges found more shocking news. In many big organizations, even the most experienced data experts spend a huge chunk of their time preparing data, rather than conducting the necessary analysis needed to derive insights.
In the long run, resource constraints make it hard for marketers to deliver tailored consumer content from several perspectives. Firstly, marketers are thrown off balance when they want to narrow down their campaigns to suit individual consumers. Secondly, when data analysts spend more time sorting data instead of analyzing it, the insights gained are anemic.
Overcoming marketing data challenges
Here are some ways to overcome marketing challenges.
- Marketers must change how they deal with big data and resource allocations
- Organizations should hire experienced data analysts and provide them with innovative analytic tools
- Marketing measurement models should be unified to resolve the challenge of disparate data
- Big data must be processed with big machines. Marketers must invest in modern analytic systems that can break down large data into manageable sizes.
Need marketing help to plan your business strategy?
We welcome the opportunity to show you how we can make your marketing SIZZLE with our data-driven, results-oriented marketing strategies. Sign up for our FREE newsletter, get the 1st chapter of our book on digital marketing analytics – FREE, or contact us for more information on hiring us.
Hausman and Associates, the publisher of MKT Maven, is a full-service marketing agency operating at the intersection of marketing and digital media. Check out our full range of services.