Top Marketing Solutions For Retail CIOs In 2020

DELVE regularly attends events throughout the year, where we have the opportunity to network and learn from top brands in the eCommerce industry. For this year’s CIO Retail Connect Conference in Fort Lauderdale, Florida, Data Science Team Lead Ben Bodek spent two days interviewing and chatting with attendees and CIO’s to learn about their top challenges, and how they plan to solve them in 2020 and beyond. Read on to hear about the event and learn what retail brands are focusing on this year in digital marketing. 

Challenge #1: Breaking Down Organizational Silos

A rapidly shifting marketing landscape means that CMOs increasingly invest in technology. In fact, CMOs spent 26% of their 2019 budget on Martech, tied with media purchases as the #1 overall marketing expense

However, despite the increasing convergence of marketing and tech, many CIOs cite a disconnect between the marketing and technology departments as a top challenge they continue to face. 

This challenge is in large part due to differences in philosophy. IT teams traditionally take more conservative approaches to new technologies, especially since a security breach or system failure may cost them their jobs. Marketing teams, tasked above all with increasing revenue and staying ahead of competitors, rely on the ability to pivot, experiment, and push boundaries. The difference in perspective often leads to rising tensions. As one CIO put it, “Marketing views IT as the ‘NO’ in innovation.” 

When marketing looks at IT as a blocker rather than an enabler, they cut IT out of the process and take Martech into their own hands. The problem that arises includes organizational inefficiencies and unnecessary risks on the path to project success. Therefore, collaboration between IT and Marketing departments is vital for deduplicating efforts and achieving the best possible business outcomes with minimized risk. 

Solution: Break down silos with clearly defined roles and responsibilities, or consider using a cross-departmental mediator. 

CIOs rely on two main silo-breaking techniques to help foster marketing and IT collaboration:

Use a cross-departmental team or trusted partner as a mediator between the IT and Marketing teams. During the event, one CIO mentioned that their internal data science department acts as coordinator for marketing technology projects, where members of the data science team gather requirements and address the needs and challenges of both IT and marketing.

Clearly define roles and responsibilities. Discuss and assign ownership of projects at the organizational level to eliminate surprises and duplicated efforts. Encourage a regular cadence of communication between departments.

Challenge #2: Breaking Down Data Silos

Today’s CMOs and CIOs have more access to data from more sources than ever before. We’ve generated 90% of digital data in the world just in the last few years, and people in the ad tech space have analyzed only 1% of that. This is in large part due to data silos. 

According to a study by Adledger, brands, on average, currently use 28 technology vendors, and brands might use up to 8.5 additional vendors in the next year. The vendors use tools to collect and analyze valuable customer data, but creating a comprehensive 360-degree view of your customer becomes increasingly challenging with a disparate tech stack that doesn’t necessarily integrate well. 

Solution: Unify customer data

Many CIOs take one of two approaches to unify customer data:

Consolidate data into a data lake for analytics consumption. This approach pulls raw and granular data into a centralized location for analysis and machine learning. For example, data scientists use tech like Hadoop to consolidate and manipulate large amounts of data, but the scalability and elasticity of cloud solutions like Google Cloud Platform or Amazon Web Services are increasingly popular solutions for Big Data. 

Utilize Customer Data Platforms. CDPs manage the integration and activation of all customer data from a variety of sources. Vendors like Optimove, BlueShift, and Tealium offer compelling solutions. However, the jury is still out on how useful these tools really are for managing customer acquisition and retention, especially since it is a new, emerging technology. 

Challenge #3: Moving Beyond the “Buzz” of Machine Learning

Machine Learning is a hot topic in the ad tech industry.  With journal articles touting titles like “The AI Revolution is here…” and “Why marketers need to be obsessed with AI and machine learning…” it’s no wonder companies rush to add data science to their business and marketing strategy. 

But the reality is that many, if not most, data science projects fail. Keeping that in mind, CIOs work to cut through the buzz of machine learning and figure out how to actualize the strong ROI potential of ML, while also minimizing risk and avoiding common pitfalls. 

Solution: Maximize success in a data science project by starting small, applying ML to the right types of problems, and feeding it clean data.

Start Small & Fail Fast. Trying to do everything all at once is the most common mistake companies make when starting down the path of machine learning. As with any new technology, there’s a strong temptation to dive immediately into the deep end. But the key to success is patience and taking a crawl, walk, run approach to data science. We suggest starting with a single POC use-case, which allows you to evaluate the ROI of machine learning with minimal risk. Meet with all relevant business stakeholders and data scientists for at least 1 hour and brainstorm 5 – 10 potential use cases for machine learning in your company (this can be anything from customer segmentation to fraud detection) Then, plot each possible use case onto an Effort vs. Impact grid (see below)

Since this is your first machine learning use case, choose a use case that is in (or closest to) the top left quadrant. Do not begin work on another use case until you’ve completed a proof of concept of your first use case evaluated it for ROI.

Apply Machine Learning, where it makes sense- it’s not always the best solution to every analytics problem. As Abraham Maslov famously said, “I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.” It’s easy to view machine learning as a silver bullet for your analytics needs. But in reality, ML is just one tool in the larger analytics toolbox. Just like you wouldn’t (or at least shouldn’t) use a hammer to pound in a screw, ML is not the right approach for every analytics problem. Many companies fail by throwing machine learning at problems that they can more easily solve by taking a much simpler approach.

Your model is only as good as your data (and your business implementation) At the event, one presenter suggested imagining data science as a race car. The car itself is the machine learning model and technology. It’s sleek, shiny, and powerful, so it’s only natural that you’d want to hop right in and take it for a spin.

Data is the gas for the race car. Without lots of it (most people don’t associate race-cars with fuel efficiency), you’re not going anywhere. In fact, if your gas (or data) isn’t clean, pure, and reliable, your race car will run inefficiently or not at all.

Next, think about you and your business stakeholders as the drivers. While the car itself is fast and powerful- it needs someone to steer it in the right direction and put all of that horsepower to good use.

What’s the point of this extended analogy? Unless you want to admire a stationary vehicle, you need to invest in fueling it with all of the clean, relevant data it needs. And to win the race, you need to make sure the car stays on track and moving in the right direction. You can have the most cutting edge model in the world, but without a well thought out, and executed business implementation plan, your data science project will likely fail.  Even with the shiniest racecar in the world, without the right fuel and a good driver who knows where they’re going, we promise you will still lose to your competitor in their 1980’s Subaru wagon.

It’s easy to get caught up in an exciting new technology, but with the right approach to data science, top brands in eCommerce drive ROI and gain valuable insights about their customers. 

Learn more about how we can help your brand build a unified measurement foundation with Google Cloud here. And if you’d like to follow along with us at events, make sure to connect with us on Facebook, Twitter, and LinkedIn. 


Ready to take your ads to the next level?  

DELVE is your strategic partner for site-side analytics, campaign management, and advanced marketing science. As experts in the Google Marketing Platform and Google Cloud Platform, DELVE drives client growth through a data-driven mindset that converts digital inefficiency into hard ROI.

SEE EXAMPLES of our experience and reviews from our clients.

Contact us to learn more about how we help our clients get advertising right.

Sarah Ash
sarah.ash@delvedeeper.com


Top Marketing Solutions For Retail CIOs In 2020

DELVE regularly attends events throughout the year, where we have the opportunity to network and…

Top Marketing Solutions For Retail CIOs In 2020

DELVE regularly attends events throughout the year, where we have the opportunity to network and learn from top brands in the eCommerce industry. For this year’s CIO Retail Connect Conference in Fort Lauderdale, Florida, Data Science Team Lead Ben Bodek spent two days interviewing and chatting with attendees and CIO’s to learn about their top challenges, and how they plan to solve them in 2020 and beyond. Read on to hear about the event and learn what retail brands are focusing on this year in digital marketing. 

Challenge #1: Breaking Down Organizational Silos

A rapidly shifting marketing landscape means that CMOs increasingly invest in technology. In fact, CMOs spent 26% of their 2019 budget on Martech, tied with media purchases as the #1 overall marketing expense

However, despite the increasing convergence of marketing and tech, many CIOs cite a disconnect between the marketing and technology departments as a top challenge they continue to face. 

This challenge is in large part due to differences in philosophy. IT teams traditionally take more conservative approaches to new technologies, especially since a security breach or system failure may cost them their jobs. Marketing teams, tasked above all with increasing revenue and staying ahead of competitors, rely on the ability to pivot, experiment, and push boundaries. The difference in perspective often leads to rising tensions. As one CIO put it, “Marketing views IT as the ‘NO’ in innovation.” 

When marketing looks at IT as a blocker rather than an enabler, they cut IT out of the process and take Martech into their own hands. The problem that arises includes organizational inefficiencies and unnecessary risks on the path to project success. Therefore, collaboration between IT and Marketing departments is vital for deduplicating efforts and achieving the best possible business outcomes with minimized risk. 

Solution: Break down silos with clearly defined roles and responsibilities, or consider using a cross-departmental mediator. 

CIOs rely on two main silo-breaking techniques to help foster marketing and IT collaboration:

Use a cross-departmental team or trusted partner as a mediator between the IT and Marketing teams. During the event, one CIO mentioned that their internal data science department acts as coordinator for marketing technology projects, where members of the data science team gather requirements and address the needs and challenges of both IT and marketing.

Clearly define roles and responsibilities. Discuss and assign ownership of projects at the organizational level to eliminate surprises and duplicated efforts. Encourage a regular cadence of communication between departments.

Challenge #2: Breaking Down Data Silos

Today’s CMOs and CIOs have more access to data from more sources than ever before. We’ve generated 90% of digital data in the world just in the last few years, and people in the ad tech space have analyzed only 1% of that. This is in large part due to data silos. 

According to a study by Adledger, brands, on average, currently use 28 technology vendors, and brands might use up to 8.5 additional vendors in the next year. The vendors use tools to collect and analyze valuable customer data, but creating a comprehensive 360-degree view of your customer becomes increasingly challenging with a disparate tech stack that doesn’t necessarily integrate well. 

Solution: Unify customer data

Many CIOs take one of two approaches to unify customer data:

Consolidate data into a data lake for analytics consumption. This approach pulls raw and granular data into a centralized location for analysis and machine learning. For example, data scientists use tech like Hadoop to consolidate and manipulate large amounts of data, but the scalability and elasticity of cloud solutions like Google Cloud Platform or Amazon Web Services are increasingly popular solutions for Big Data. 

Utilize Customer Data Platforms. CDPs manage the integration and activation of all customer data from a variety of sources. Vendors like Optimove, BlueShift, and Tealium offer compelling solutions. However, the jury is still out on how useful these tools really are for managing customer acquisition and retention, especially since it is a new, emerging technology. 

Challenge #3: Moving Beyond the “Buzz” of Machine Learning

Machine Learning is a hot topic in the ad tech industry.  With journal articles touting titles like “The AI Revolution is here…” and “Why marketers need to be obsessed with AI and machine learning…” it’s no wonder companies rush to add data science to their business and marketing strategy. 

But the reality is that many, if not most, data science projects fail. Keeping that in mind, CIOs work to cut through the buzz of machine learning and figure out how to actualize the strong ROI potential of ML, while also minimizing risk and avoiding common pitfalls. 

Solution: Maximize success in a data science project by starting small, applying ML to the right types of problems, and feeding it clean data.

Start Small & Fail Fast. Trying to do everything all at once is the most common mistake companies make when starting down the path of machine learning. As with any new technology, there’s a strong temptation to dive immediately into the deep end. But the key to success is patience and taking a crawl, walk, run approach to data science. We suggest starting with a single POC use-case, which allows you to evaluate the ROI of machine learning with minimal risk. Meet with all relevant business stakeholders and data scientists for at least 1 hour and brainstorm 5 – 10 potential use cases for machine learning in your company (this can be anything from customer segmentation to fraud detection) Then, plot each possible use case onto an Effort vs. Impact grid (see below)

Since this is your first machine learning use case, choose a use case that is in (or closest to) the top left quadrant. Do not begin work on another use case until you’ve completed a proof of concept of your first use case evaluated it for ROI.

Apply Machine Learning, where it makes sense- it’s not always the best solution to every analytics problem. As Abraham Maslov famously said, “I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.” It’s easy to view machine learning as a silver bullet for your analytics needs. But in reality, ML is just one tool in the larger analytics toolbox. Just like you wouldn’t (or at least shouldn’t) use a hammer to pound in a screw, ML is not the right approach for every analytics problem. Many companies fail by throwing machine learning at problems that they can more easily solve by taking a much simpler approach.

Your model is only as good as your data (and your business implementation) At the event, one presenter suggested imagining data science as a race car. The car itself is the machine learning model and technology. It’s sleek, shiny, and powerful, so it’s only natural that you’d want to hop right in and take it for a spin.

Data is the gas for the race car. Without lots of it (most people don’t associate race-cars with fuel efficiency), you’re not going anywhere. In fact, if your gas (or data) isn’t clean, pure, and reliable, your race car will run inefficiently or not at all.

Next, think about you and your business stakeholders as the drivers. While the car itself is fast and powerful- it needs someone to steer it in the right direction and put all of that horsepower to good use.

What’s the point of this extended analogy? Unless you want to admire a stationary vehicle, you need to invest in fueling it with all of the clean, relevant data it needs. And to win the race, you need to make sure the car stays on track and moving in the right direction. You can have the most cutting edge model in the world, but without a well thought out, and executed business implementation plan, your data science project will likely fail.  Even with the shiniest racecar in the world, without the right fuel and a good driver who knows where they’re going, we promise you will still lose to your competitor in their 1980’s Subaru wagon.

It’s easy to get caught up in an exciting new technology, but with the right approach to data science, top brands in eCommerce drive ROI and gain valuable insights about their customers. 

Learn more about how we can help your brand build a unified measurement foundation with Google Cloud here. And if you’d like to follow along with us at events, make sure to connect with us on Facebook, Twitter, and LinkedIn. 


Ready to take your ads to the next level?  

DELVE is your strategic partner for site-side analytics, campaign management, and advanced marketing science. As experts in the Google Marketing Platform and Google Cloud Platform, DELVE drives client growth through a data-driven mindset that converts digital inefficiency into hard ROI.

SEE EXAMPLES of our experience and reviews from our clients.

Contact us to learn more about how we help our clients get advertising right.

Related Posts

Uncover Lower CPAs for Insurance Marketing with Multi-Touch Attribution

Marketers know how important it is that their paid advertising reaches their target audience with…

Improve Data Analysis with Google Data Studio 360

A massive amount of data accompanies the everyday work of a marketer. They need to…

Top 5 Programmatic Advertising Challenges and Solutions

Marketing leaders aim to control spendings more precisely now than ever before. Regardless of whether…

Top 5 Programmatic Advertising Best Practices for Better Marketing Efficiency

What is the role of display, video, and audio advertising in your brand’s digital marketing…