Article Summary
Every leadership team has had this meeting: the CRM says the quarter closed one way, the finance system says something else, and the operations team has a backlog number that doesn’t match either one. Gartner research puts the average cost of poor data quality at $12.9 million a year across the organizations it studies — a figure large enough that even a fraction of it, scaled to a $25M–$100M manufacturer, is real money spent reconciling spreadsheets instead of making decisions.
Gartner’s research identifies the specific mechanism behind that cost: inconsistency across data sources is named as the single most challenging data quality problem organizations face, and it’s the direct result of data being stored and maintained in silos — systems with overlapping, gapped, or contradictory records that were never built to talk to each other. For a manufacturer, that usually means a CRM that tracks opportunities one way, a finance or ERP system that recognizes revenue a different way, and an operations system tracking orders and backlog on its own separate timeline. None of the three systems is necessarily wrong. They’re just not describing the same reality.
The deeper issue Gartner points to isn’t technical, it’s organizational: business leaders broadly agree that data quality matters, but rarely see it as their own responsibility, and don’t always understand how the data in their domain connects to outcomes outside it. Data quality, in other words, doesn’t fail because nobody cares — it fails because everybody assumes someone else owns it.
For a manufacturing leadership team, the practical result shows up right before the board meeting: someone spends a day (or three) manually reconciling numbers that should already agree, and the figure that finally gets presented is the one that survived the argument, not necessarily the one that’s accurate. Fixing that isn’t a software purchase. It starts with agreeing on which system is the source of truth for which number, and naming who’s accountable for keeping them aligned.
Why Do Three Departments Report Three Different Numbers for the Same Quarter?
Because the systems that produce those numbers were built separately, adopted at different times, and are rarely required to reconcile with each other automatically. The CRM logs an opportunity as “closed-won” the moment a rep marks it that way. The finance system recognizes that same revenue on a different date, tied to invoicing or shipment. The operations system tracks the order against production and delivery timelines that may not align with either. Each system is internally consistent. None of them was designed with the other two in mind.
Gartner’s research on data quality names this directly: inconsistency across sources is the most commonly cited data quality problem organizations report, and it stems from data living in silos with overlapping, missing, or contradictory records. Left unconnected, standardizing that data into one trustworthy number becomes significantly harder — not because the underlying data is bad, but because nobody forced the three systems to agree on a shared definition in the first place.
What Does This Actually Cost a Manufacturer?
Gartner puts the average cost of poor data quality at $12.9 million a year across the organizations it has studied — a figure drawn from a broad cross-section of company sizes and industries, not a manufacturing-specific benchmark, so it shouldn’t be applied to any single company as a literal number. What it does establish is the order of magnitude: this isn’t a minor annoyance, it’s a cost large enough that major research firms track it as a distinct line item.
For a $25M–$100M manufacturer, the more useful question isn’t “does our number match $12.9 million,” it’s “how many hours does leadership spend every quarter reconciling numbers that should already agree, and what decisions get delayed while that happens.” Forecasts built on a CRM number that finance doesn’t trust get revised downward out of caution. Headcount and investment decisions get deferred because nobody in the room agrees on the baseline. The cost isn’t just the reconciliation time — it’s the decisions made late, or made on a number everyone privately doubts.
Where Does the Inconsistency Actually Start?
It starts earlier than most leadership teams assume — usually at the point where a lead or opportunity first enters the CRM with no agreed definition of what “qualified” means, and it compounds every time that record moves to another system without a shared standard. By the time an opportunity becomes an order, and the order becomes a shipped, invoiced, revenue-recognized transaction, it has usually passed through three or four systems, each with its own field names, its own timing rules, and its own definition of what counts as “done.”
Gartner’s research is explicit that this is a silo problem, not an individual-error problem: data quality suffers most when systems that should share a definition are never connected, and standardization becomes progressively harder the longer that gap persists. For a manufacturer, that gap is usually widest exactly where marketing- or sales-generated demand data meets finance’s revenue recognition rules and operations’ delivery data — three functions with three different jobs, using three different systems, none of which were built to reconcile with the other two automatically.
Why Doesn’t Someone Just Own This?
Because, according to Gartner’s research, most organizations have an ownership gap rather than a caring gap. Business leaders across functions generally agree that data quality matters. What’s missing is a clear sense that it’s any one person’s job — and an understanding of how the data a given department produces actually connects to outcomes in a different department. Finance may not realize how a CRM field they’ve never opened affects revenue recognition. Sales may not realize how their opportunity stage definitions ripple into operations’ production planning.
The result is a familiar pattern: everyone nods that the numbers should match, and nobody is accountable when they don’t. Gartner’s research frames data quality as a business discipline requiring cross-functional ownership and collaboration, not a technology project handed to IT — which means the fix has to start with naming an owner, not buying a new system.
What Does Actually Fixing This Look Like?
It starts with three decisions, none of which require new software. First, pick a single source of truth for each core number — which system’s figure for revenue, backlog, or pipeline is the one everyone agrees to trust, even if another system’s number differs. Second, write down the shared definitions that number depends on: what counts as a qualified opportunity, what counts as closed-won, what counts as recognized revenue, and when. Third, name a single owner accountable for keeping those definitions consistent across systems, rather than leaving it to whoever happens to reconcile the spreadsheet before the next leadership meeting.
None of that requires ripping out a CRM or an ERP system. It requires agreement on what each system is allowed to be the authority on, and a standing process for catching drift before it turns into a pre-board-meeting scramble. If your leadership team currently can’t say, without an argument, which system is the source of truth for revenue, that disagreement — not a new tool — is usually the first thing worth resolving. That’s typically where a diagnostic conversation starts: not with a system recommendation, but with a clear map of where the numbers actually diverge and why.
Frequently Asked Questions
Why do different departments report different numbers for the same period? Because the systems producing those numbers — typically a CRM, a finance or ERP system, and an operations or order-management system — were built and adopted separately, with their own field definitions and timing rules. None of the numbers is necessarily wrong; they’re measuring slightly different things.
How much does poor data quality actually cost a company? Gartner research puts the average cost at $12.9 million a year across the organizations it studies. That figure spans company sizes and industries rather than being manufacturing-specific, so it’s best used as an order-of-magnitude signal rather than a benchmark to hit exactly.
What’s the single biggest driver of data inconsistency? According to Gartner, inconsistency across data sources — caused by data sitting in disconnected silos with overlapping or contradictory records — is the most commonly cited data quality problem organizations report.
Whose job should it be to fix data inconsistency across systems? Gartner’s research frames this as a cross-functional ownership gap: most leaders agree data quality matters but don’t see it as their individual responsibility. Fixing it requires a named, accountable owner and collaboration across the departments whose systems feed into the same numbers — not just a technology purchase.
How does a manufacturer start fixing disconnected systems? Start by agreeing which system is the authoritative source for each core number (revenue, backlog, pipeline), documenting the shared definitions behind that number, and naming a single owner accountable for keeping those definitions consistent across systems going forward.
