Marketers are under greater pressure than ever to optimize, measure, and ultimately predict outcomes of their efforts. Marketers working in an agency environment, serving clients, who are measured on the success of campaigns and dependent on that success to drive client renewals feel that pressure even more intensely.

Many marketers have figured out the measurement of marketing programs—most agencies can tell their clients how well a campaign performed after-the-fact. But, can you provide business and investment insight or predict how a future campaign will perform? Maybe not.

While the phrases “Artificial Intelligence” and “Machine Learning” are in every trend-report, its adoption is slow in non-IT or non-enterprise environments. For some, the idea of machine learning is still obscure and it’s not clear how to leverage machine learning tactics like predictive analytics or optimization to support clients and their marketing efforts. Or, in some cases, offering services like predictive analytics is simply too expensive and not profitable for an agency.

However, data and the tools to leverage that data for marketing has become more accessible. There’s been a “democratization of data,” if you will. And, as data scientists and marketers, we’ve seen and believe in the power of machine learning and its ability to improve agency services and outcomes for clients.

We know that there are barriers to providing these kinds of services. The most common barrier we hear from our clients and prospects is that they don’t have enough access to data or clean data.

Data doesn’t have to be a barrier. There are solutions for data prep that don’t require a whole team of data scientists. Here are some suggestions to overcome the data prep barrier and an example of an agency who overcame the data access hurdle.

Data Preparation Doesn’t Have to Hamstring Analytics

According to the Harvard Business Review (HBR), analytics outcomes are not matching analytics spending. HBR cites data integration efforts and lack of talent as primary reasons. Data preparation often hinders the growth of analytics programs.

To overcome this cycle, data preparation can occur in tandem with analytics efforts. In fact, analytics should happen without high fixed costs and organizations need to be wary of the ‘perfect’ data trap.

When it comes to data management, there are a number of tools out there, like LityxIQ, to help marketing analytics departments generate insights from data, before it’s “perfect” using approaches like “windowing” or “partitioning”.

  • How long between marketing activity and purchase?
  • What’s the total dollar value of purchases by our client’s customers?
  • Are customers spending at an increasing or decreasing rate?
  • Which channels are driving the most customer spending?
  • Which messages and targeting have the highest return on investment for our client?

Using a data platform with an easy-to-use interface enables marketers and marketing analysts to uncover new patterns in data and answer the questions above, more easily. Instead of writing specifications and tickets, analysts can run their own scenarios. This approach minimizes the need for support from an IT department and makes it more affordable to perform data services for clients.

For example, our agency client, Concord Direct, uses LityxIQ to provide deeper insights on donors to their nonprofit clients.

Concord is a mid-sized agency focused on non-profits looking to bring a unique and modern approach to their clients.

The challenge Concord was facing was that the size of the data they wanted to report on to their clients belied the complexity of their requirements. A simple three field file of 10-years giving history needed to be rolled up by dynamic periods of time and several member segmentation schemes needed to be represented.

Using a combination of LityxIQ and Tableau, Concord was able to create a valuable report for their client. LityxIQ was used to automate and perform complicated data preparation for Tableau which was set up with a variety of calculated conditional fields, parameters, dashboards and ultimately a storyboard for final presentation to the agency’s client and they were able to accelerate and smarten their decisions about marketing investments.

They didn’t have to hire data scientists and they were able to provide their client with an added-value service that improved the client experience.

Agencies and their clients have a lot to gain by adding advanced analytics to their service offerings. For more examples on how greater use of data can impact marketing efforts, check out our collection of case studies.

And, stay tuned for our next post for guidance on how to use predictive analytics to improve programs across digital and offline channels.


Sign up here to subscribe to the blog

Subscribe Now