Supply chains have never had access to more information. Transportation management systems, warehouse management systems, enterprise resource planning platforms, visibility tools and customer portals all generate data that can help companies understand what is happening across their networks.
But in today’s data-driven supply chain, access to information alone isn’t enough – the quality and trustworthiness of that data is what separates confident decisions from costly mistakes. “More information doesn’t necessarily equate to better outcomes,” said Mike Medeiros, executive vice president of operations for Penske Logistics.
Organizations certainly are not short on information. I think they're short on information that they can truly trust.
– Mike Medeiros
That distinction is growing more critical as supply chains become more complex and organizations turn to data, analytics and artificial intelligence to improve performance. Medeiros framed the issue in familiar technology terms: garbage in, garbage out.
Data can only improve decision-making when it is accurate, complete, timely and connected to the operational realities it is meant to represent. Without that foundation, even the most advanced systems can produce flawed insights, slow response times and uncertainty at the point of decision.
Data Quality Has Become a Business Constraint
For years, supply chain visibility was viewed as the primary goal. Companies wanted to know where shipments were, when inventory would arrive and whether orders were moving as planned. Visibility remains important, but it is no longer sufficient on its own.
“Visibility has really become table stakes in our business,” Medeiros said. “Showing you the location of where freight or inventory is located doesn’t necessarily mean you understand the implication to the supply chain.”
Achieving a deeper level of understanding depends on context and on the quality of the data flowing through the organization. If shipment statuses are incomplete, inventory details are inconsistent or system timestamps are misaligned, decision-makers may hesitate or, worse, act on the wrong information.
Poor data quality also creates operational challenges. For example, consider a shipment that is running late. The transportation team sees an updated estimated arrival in their system, but the warehouse team is still planning labor and dock assignments based on the original schedule. The customer, meanwhile, is looking at a portal that has not yet reflected either update.
Three teams, three versions of the truth – and each one making decisions based on incomplete or conflicting information. That disconnect can slow decisions, generate duplicate work and increase the likelihood of reactive problem solving. By the time everyone is working from the same picture, the window may have already closed to be proactive.
“Disconnected systems oftentimes create a lot of friction,” Medeiros said.
They create that uncertainty that leaves people second-guessing their decision-making process.
– Mike Medieros
For supply chain leaders, the challenge is not simply collecting more data. It is deciding which data matters most, ensuring it is reliable and making it available in a way that supports action.
Clean Data Creates Confidence
Supply chain data integrity is a core focus of Penske’s Supply Chain Insight platform, which aggregates data across multiple applications to create a single view of supply chain activity. The platform brings loads, orders and item-level data together in one place – giving customers and Penske’s frontline teams a shared, current picture of what is moving and what needs attention.
Many supply chains operate across systems that were each designed to perform well within their own function. A transportation system may provide strong data for loads and carriers, and a warehouse system may provide strong data for inventory and order flow - but the handoff between those systems is often where visibility begins to break down.
Medeiros said Penske began by identifying the critical data elements needed to support accurate metrics and visibility. Those elements vary by service, product line and system, but the underlying objective is consistent: determine which inputs must be right so the outputs are useful.
Penske’s approach is built around three phases: manage, monitor and remediate.
- In the manage phase, the company identifies critical data elements for each service line and promotes data literacy and stewardship across the organization. Data literacy means ensuring that the people entering, handling and acting on data understand what each field represents and why it matters. Stewardship assigns clear ownership – so that when a data quality issue surfaces, there is always a designated person or team responsible for solving it.
- In the monitor phase, the company establishes rules, alerts and processes to detect anomalies as they arise – flagging issues like missing timestamps, inconsistent statuses or incomplete records before they affect downstream decisions. Monitoring is not a passive process. It requires defining what good data looks like for each field and service line, then building the automated checks that surface deviations in real time. The goal is to catch problems as close to the point of entry as possible, before bad data has the opportunity to travel through connected systems and compound into larger issues.
- In the remediate phase, teams act on what they find, track metrics and close. This is where data quality moves from measurement to improvement. When an anomaly is defined, the focus shifts to understanding the root cause – whether it’s a system integration issue, a process gap or a data entry error at the point of origin – and correcting it at the source rather than simply patching the output. Progress is tracked over time to ensure fixes hold and that data quality scores improve across the organization.
Consider a shipment with a missing or incorrect appointment time. In isolation, that may seem like a minor data entry issue. But if that field drives delivery confirmation, customer reporting and carrier performance scoring, a single bad input can ripple across multiple systems and metrics. Penske’s framework is designed to catch that kind of anomaly early – flagging it through automated alerts, routing it to the right team for correction and tracking whether the fix holds over time.
“We’re actually measuring our data,” Medeiros said. “We have dashboards that go out in the field every single day that identify what the quality of the data that they’re inputting into the support systems is, and it allows them to take swift action when those anomalies are identified and detected.”
That discipline matters because supply chain data drives key performance indicators (KPIs), exception management, customer conversations and operational planning. When the data is trustworthy, teams move faster and with greater confidence. When it is not, even strong analytics can introduce doubt.
Medeiros added that bad data can be worse than no data at all. Inaccurate information that appears trustworthy can lead teams to act confidently on a flawed picture. The wrong KPIs can reinforce the wrong priorities. A misleading dashboard can give leaders a false sense of control at exactly the moment when they should be asking harder questions.
Context Matters for Supply Chain Data Quality
Aggregating data into a single location is an important step, but it is not the whole answer. Context determines whether information can be acted on – not just observed.
A shipment status may show that freight is delayed, but teams also need to know the customer priority, delivery window, available alternatives and downstream impact. An inventory count may show what is on hand, but teams also need to know where that inventory is located, whether it is allocated, whether it is damaged and whether it can meet the timing of the order.
That’s why trusted data must be both accurate and operationally meaningful. It should help teams understand what is happening, why it matters and what decision needs to be made next. The progression moves from visibility – knowing where things are – to context – understanding what that means – to intelligence – know what to do about it. Each step depends on the quality and completeness of the data that came before it.
AI in the Supply Chain Raises the Stakes for Data Quality
Artificial intelligence (AI) in the supply chain is intensifying the need for clean, structured and trustworthy data. As Penske deploys AI across its operations, one principle has become clear: those capabilities are only as reliable as the data behind them. “In order for AI to truly have an impact and be a force in your business, you really need good quality data inputs,” Medeiros said.
Penske’s AI strategy is built around an associate-in-the-loop model, where AI surfaces insights and recommendations while experienced associates apply judgement and context to the final decision. That approach is intentional. AI can process data at a scale and speed no team can match, but it cannot determine whether the data it is working from reflects operational reality – that responsibility stays with the people closest to the work.
“This approach ensures that customers benefit from advanced analytics and automation without losing accountability, because there’s always an associate overseeing what’s happening,” said Vish LK, vice president of financial administration for Penske Logistics, who also oversees AI strategy and industry outreach for the company.
There’s flexibility and service excellence that come from associate expertise, but at the same time, there’s speed and accuracy that comes from AI.
– Vish LK
Accuracy is especially important in supply chain operations, where decisions can affect production lines, critical medical supplies, groceries, automotive parts and customer commitments. Organizations that want to use AI effectively must start by understanding their data environment – where information comes from, how it is defined, who owns it and which fields are critical to decision-making.
How Supply Chain Visibility Reduces Friction
Supply chain complexity often stems from the number of systems, partners and handoffs involved in moving goods. As organizations adopt more technology, they can unintentionally create more fragmentation and silos of data.
Supply Chain Insight addresses that challenge by connecting data across transportation, warehousing and supply chain management into a single, shared view. Users can track the full journey of every load – including all pickup and delivery stops and the orders tied to each – making it easier to spot delays and understand their downstream impact. They can also search for any item or order, review performance trends and troubleshoot issues without switching between systems. That shared visibility means a transportation manager, a warehouse supervisor and a customer are no longer working from different versions of the truth – they are looking at the same picture, in real time, and can act on it together.
“When you can manage and monitor the performance and health of your supply chain in one place, we’re seeing more expedited decision making, more confidence, and it really allows us to get to that desired state when we’re anticipating where the next disruption is coming from, rather than reacting to a problem that’s already occurred,” Medeiros said.
That shift from reaction to anticipation is where clean data creates measurable operational value. When teams can see inventory movement, transportation status and potential exceptions in one place, they are better positioned to act before a disruption creates bigger issues or becomes more costly.
Better Decisions Start With Better Inputs
The goal of any analytics platform, AI tool or visibility solution is to deliver trusted insights - and that requires clean data, connected systems, common definitions and processes to identify and correct anomalies. Medeiros was direct on the point: data quality is not optional. “Bad inputs create bad outcomes,” he said. “If you don’t trust the data, you can’t trust the decisions you’re making on a day-to-day basis.”
Frequently Asked Questions About Supply Chain Data Quality
What is supply chain data quality?
Supply chain data quality refers to the accuracy, completeness, timeliness and consistency of the information flowing through a supply chain. High-quality data means that every system, team and partner is working from reliable, up-to-date information that reflects operational reality – enabling faster, more confident decisions across the network.
Why is data quality important in supply chain management?
Data quality is the foundation of effective supply chain management. When data is inaccurate or inconsistent, teams may hesitate, act on flawed information or work from different versions of the truth. Poor data quality can slow decision-making, create operational friction and undermine the performance of even the most advanced analytics and AI tools.
What is the difference between supply chain visibility and supply chain intelligence?
Supply chain visibility means knowing where things are – the location of a shipment, the status of an order, the level of available inventory. Supply chain intelligence goes further, providing the context needed to understand what information means and what action should be taken next. Progression moves from visibility to context to intelligence, and each step depends on the quality and completeness of the data that came before it.
How does AI affect supply chain data quality requirements?
AI in the supply chain intensifies the need for clean, structured and trustworthy data. AI tools can improve forecasting, identify exceptions and accelerate decision-making – but they cannot overcome poor inputs. Inaccurate or incomplete data fed into an AI system will produce flawed insights, regardless of how sophisticated technology is. Organizations that want to use AI effectively must first establish a strong data quality foundation.
What is an associate-in-the-loop model for AI in logistics?
An associate-in-the-loop model means that AI provides insights and recommendations while experienced associates apply judgment and context to the final decision. Rather than replacing human expertise, AI augments it – combining the speed and analytical power of machine learning with the accountability and situational awareness of experienced logistics professionals.
Better Data, Better Outcomes
Supply chain performance has always depended on making the right call at the right time. What has changed is the volume of data available to support those calls – and the risk that comes with acting on data that cannot be trusted.
Organizations that treat data quality as a strategic priority, not as an IT concern, are the ones best positioned to make faster decisions, respond to disruptions with confidence and get more value from every technology investment they make.
Learn how Penske Logistics helps customers build supply chains they can count on.
