top of page

Garbage in.... is Garbage Out. Why Data Quality must be a #1 C-suite focus in the Age of AI

Writer's picture: Cate GutowskiCate Gutowski

Updated: Nov 6, 2024

No matter how much you invest ... you won't get an ROI on AI-driven investments without clean data. To win in the age of Artificial Intelligence, Data Quality must become the #1 Imperative for every C-suite leader & Board Director.


Want to learn where your organization stands? Take our free Machine Data Quality Quiz.


Industry Insights by Cate Gutowski, CEO of Quantis.ai



solutions to clean received data and validate
head of maintenance and head of automation viewing beverage machine


‘Bad Data Quality’ is costing beverage manufacturers billions


According to Gartner, ‘bad data’ costs organizations $12.9 million per year on average. The McKinsey Global Institute confirmed poor-quality data can lead to a 20% decrease in productivity and a 30% increase in costs



What is your current state of your Data Quality... and what is it costing you?


Do you know what bad data is costing your firm? Most organizations don't, and that is a problem.


At Quantis.ai, I’ve seen how important data quality is to our customers. As a specialized AI technology firm focused on serving the needs of the beverage industry, we are proud to partner with the world’s top beverage firms. Each brand we serve is often worth billions of dollars and is cherished by loyal customers who count on their favorite refreshment tasting the same, every time.


These brands understand deeply the connection between quality and customer loyalty, and I've been impressed with their commitment to making quality job one. As a former Six Sigma Black Belt, I can appreciate their focus on quality.


For example, when a brand achieves the coveted level of 'six sigma quality', it indicates that for every 1 million beverages produced, there were a mere 3.4 defective beverages produced.  For many, this is still too much... but, statistically, this is an achievement of world-class quality.


Given the extreme importance of quality to the brand reputation of the top brands in the Beverage Industry, they often rely on neutral third parties like Quantis.ai to independently measure and test data quality levels in their plant operations to verify quality.


They appreciate our ability to test quality at a machine level, because they understand that this is ground zero for all plant operations.


When we are able to confirm high levels of data quality at a machine level, it creates the foundation for a new, emerging asset for beverage manufacturers - clean, high-quality data at a plant infrastructure level.


KPI of important machine data from filler machine

Beverage manufacturers are prioritizing the investment required for firms like Quantis.ai to independently test and verify data quality levels of the most commonly utilized beverage machines such as fillers, mixers, and more because they understand that when data quality at a machine level is not meeting their standards, there is a downstream impact.


One of the most alarming outcomes is we've witnessed came from one customer who shared, "When machine data is not meeting our specification, then this means there is a negative downstream impact to Plant Operations.


The impact is that the data feeding the KPI reports that my leaders depend on is not accurate, and this is a problem, because we depend on those reports daily."


...when data quality at a machine level is not accurate, there is a negative downstream impact to Plant Operations.

Yikes. Can you imagine telling your leaders that the reports they are using to run their operation can no longer be trusted? But, this is an example of the downstream impact to an operation with poor levels of data quality.



Should your AI-driven Investment Strategy be internally… or externally focused?


Poor data quality can not only impact internal operations, but also customer relationships. It is easy to understand how this can happen.


Today, there are well-intentioned firms who are pushing their teams to innovate and adopt AI technology. As we rush to innovate, it is normal that mistakes will be made.  I like to tell my team that every day, we are either winning… or learning. All of us love to win (especially me!) But, if we are not winning, then we better be learning.


Some days, we learn what works. Other days, we learn what does not work – and both should prepare us to win more frequently if we are rapidly implementing what we learn.



What are the risks of externally focused AI-driven Investments?


Chevrolet learned what did not work in the age of AI – they had a service chat bot that agreed to sell a customer a new Chevrolet Tahoe for one dollar, and Air Canada was forced to compensate a passenger who received an incorrect refund information from its service chatbot.


Both firms recently learned that sometimes, choosing to learn about AI alongside your customers can be a risky strategy, as customer trust is fragile… it takes years to earn trust, and only seconds to lose customer trust.


head of automation on their cell phone remote working

Enterprises are prioritizing internal vs. external AI-Driven Investments


We see the majority of enterprise customers choosing to learn about AI by focusing internally first – as they feel there is less risk. Enterprises investing in AI prefer to experiment in areas of AI that can help improve daily decision-making or optimize internal processes. Regardless of where you focus, leaders must set their teams up for success. 


One of the best ways for leaders to do that is to invest in building a strong foundation in data quality.

For customers in the Beverage Industry, we believe that that building a solid foundation of clean data at a plant level will be the #1 most important investment they make in the next 3 years to position their firms to win in the age of AI.


CAN line ina beverage factory

Garbage In … Means Garbage Out


I think about data quality for Quantis the same way that I think about my health.


If I want to maintain my good health, I must prioritize healthy eating. Growing up, my Grandfather, a former Army General, was a big champion of healthy habits. For example, when he wanted to motivate us to eat more vegetables, he would boldly declare, “garbage in… is garbage out.” 


This phrase did work to inspire me to eat more vegetables, and, it captures the essence of the challenge facing the c-suite and the boards of publicly traded enterprises: without accurate, high-quality data … then garbage in … is garbage out. As every enterprise considers investments in AI in 2025 and beyond, they must also consider whether or not their firm is prepared to get a return on those investments.  As Boards review 2025 plans, they should be asking: do we have the right data foundation before we invest in this AI technology? And, more importantly, how do you know?


...without accurate, high quality data then garbage in ... is garbage out.

Boards should also be asking: How do we know if our Data Quality is accurate?

This simple question is critical. In the age of AI, your organization cannot afford to get this wrong. Each time you invest in capital equipment or technology, you have a right as an investor to understand how data quality is measured and reported. We would encourage all firms to ask technology providers, OEMs, and manufacturers of all types: how can we confirm that we have high quality data? How do we know? Most importantly, how can we validate it?


There are multiple resources in the industry who can independently measure and verify data quality levels for your organization. Given the importance of clean data to the future of AI, we recommend you get multiple quotes and independently verify and measure your data quality at a machine level.



CFO and CIOs are positioned to play a critical role in enabling Data Quality


Typically, we tend to think of CIOs as the ones responsible for data quality levels in an organization. However, I believe this is old and outdated thinking. We've known for a long time that data is the new oil, so we need our thinking to catch up to the current business reality.


Modern CFOs understand the strategic importance of clean data to their future revenues and profitability. They also understand that AI will force every industry and every business to evolve and change, and that this is a space they should own and drive. I recently spoke at an industry event for CFOs, and it was clear that they understood deeply the importance of validating data quality levels prior to investing in AI technologies.


After all, without a strong foundation of clean data, even the most sophisticated and well-resourced firms will not be able to achieve an ROI on those strategic AI-driven investments because poor quality inputs will always lead to poor quality outputs, which can paralyze firms for years to come by preventing high quality decisions from being made every day on the shop floor.


two shop floor employees reviewing the line

Delivering Results requires a focus on Data Quality as a Key Competency


At Quantis.ai, we’ve defined the term ‘data quality’ as the accuracy, completeness, consistency, and reliability of the data used within an organization for daily decision making. If your teams cannot rely on the data for daily decision making, then you do not have high quality data in place.


For a firm to determine if the data is high quality or not requires the organization to dive deep into the detail, and to create definitions and standards around both types of data. If your firm does not have these foundational definitions in place, then you do not have a focus on data quality.


This is a good litmus test for c-suite leaders who have team members presenting business cases to their leadership teams asking for investment in AI-driven solutions in 2025.


chief financial officer at a beverage factory viewing pocket factory dashboard

Strategic preparation delivers stronger returns on AI-Driven Investments


All of us have been tempted to invest in a new, shiny AI-driven object. But, before you do.... C-suite leaders should consider asking their teams: What good is an investment in a new AI-based shiny object … if you don’t have the assurance that the data that will feed the AI models is of the highest quality? 


This may not be a popular question, but it is an essential one. Without high quality data, your team will be unable to make high quality decisions.


The impact? You will not deliver a return on your investment.



Is there a connection between Data Quality levels ... and Beverage Quality?


AI algorithms and large language models learn and make predictions based on the data they are fed. 


This comment surprised me and stopped me in my tracks.


I asked him to educate me further. His point was that if data quality is poor in a manufacturing environment, then we need to take it as seriously as he sees his firm taking cybersecurity. He explained that he was concerned that poor levels of data quality should be treated with a sense of urgency - because hundreds of millions of consumers are trusting beverage firms to get quality right due to the extreme importance of consumer health and safety.


His point was that poor data quality levels could cause teams to make incorrect insights, and ultimately... poor decisions. He was concerned that these decisions ... or lack of decisions... could result in significant financial and reputational damage.


Poor data quality should be considered as dangerous as some cybersecurity threats?


head of automation and head of maintenance


Data Quality is not a 'check-the-box' exercise, there is real business impact


The message is clear: data quality is more than a check-the-box exercise— it’s the foundation for ensuring your future of operational success. As AI technologies continue to grow in importance, we believe it's clear that manufacturing is at a pivotal moment. C-suite leaders and Boards are realizing that high-quality data isn’t just an asset; it’s a requirement to enable accurate insights to the shop floor. Those accurate insights are critical - without them, you are unable to empower your teams to make the high quality decisions required to maintain consumer trust. If you are curious to learn where your organization stands, take a few minutes to do our free Machine Data Quality Quiz.


In my next article, I will dive into three core strategies you can implement beginning tomorrow -- to build a solid data quality foundation for your beverage operation. From practical steps to strategic insights, we will help you to unlock the power of your data at a machine level and provide you with simple tips that can help you to lead your team to succeed in the age of AI.



33 views0 comments

Recent Posts

See All

Comments


bottom of page