The data opportunity for improving product Quality
With more data becoming easily available, Quality departments have turned to that data to solve their problems early (ideally even before they occur), deal with incidents faster, and put in place robust measures to avoid future issues.
However more data just adds to workload unless one has the tools to handle it. If your teams have to manually sift through many documents looking for that needle in haystack info, or is doing a lot of manual data entry, you may have an opportunity to make them happy and improve the effectiveness of your department!
The Quality data overload
Let us look at the vast variety and amount of data that a Quality department needs to work with to do its job:
Fundamentally, you have two different kind of data sets that a Quality department deals with – structured data such as in Tables and unstructured data as in text or images
| Quality Business Process | Structured Data | Unstructured Data |
|---|---|---|
| 1. Quality Control & Testing | LIMS results, MES process parameters (temp, pressure), ERP inspection lots. | Lab Notebooks, Photos/Videos of line checks, Operator free-text notes. |
| 2. Supplier Quality Management | Supplier Rating Scorecards, Audit Scores, Non-Conformance Reports (NCRs). | Supplier Contracts, Audit Reports (text-heavy assessments), Quality-related email correspondence. |
| 3. Complaints & Feedback | CRM records (codes, resolution status), Call Center Logs, Warranty/Return Data. | Customer Complaint Narratives (open-text description), Social Media text/sentiment, Call Transcripts. |
| 4. Corrective/Preventive Action (CAPA) | QMS records (status, due dates), Root Cause Analysis codes, Implementation metrics. | Root Cause Investigation Reports (detailed analysis text), Verification of Effectiveness (VoE) Reports, CAPA review Meeting Minutes. |
| 5. Traceability & Recall | WMS/ERP lot numbers, batch dates, ingredient consumption, ship-to location records. | Paper Batch Records (historic logs), Receiving/Shipping Documentation, Regulatory Filings/recall notices. |
| 6. Quality Audits & Compliance | QMS Audit schedules, Checklist scoring, Finding/Observation codes, Follow-up tracking. | Audit Reports (textual findings/recommendations), External Regulatory Standards (PDFs), Auditor Notes/transcripts. |
| 7. Document Control (The Foundation) | DMS metadata (ID, version, approval dates, owner fields). | Standard Operating Procedures (SOPs), Work Instructions, Quality Manuals, Specifications, Training Materials. |
The Unstructured data pain point
While it is getting easier to deal with structured data using Excel or BI tools such as PowerBI, unstructured data still remains a pain point. Some companies have the additional pain of dealing with structured data in an unstructured format, for example, lab measurements coming in as pdf documents. And then of course we have the really unstructured data such as customer complaint narratives, and detailed investigation reports. Quality departments either have to work manually with these documents if that is absolutely necessary or they tend to ignore it if there is no capacity.
Better data handling, of unstructured data will free up a significant time – both time in day-to-day operations, as well as fire-fighting. Your Quality function can then focus its attention on value-add tasks such as implementing robust quality measures for guaranteeing future quality.
Case in Point: Fixing the CAPA Bottleneck
As an example, let us look at the Corrective and Preventive Action (CAPA) process, the backbone of quality improvement. While your QMS tracks the structured data (CAPA number, status), the critical root cause investigation findings are recorded in a lengthy, unstructured text document.
When a Quality Director needs answers, the team is forced to manually read, search, and subjectively categorize these detailed investigation reports. This slow, subjective process wastes valuable time and leads to dangerous missed connections.
For example, a human reviewer might easily miss that 15 recent deviations, across three different sites, were all linked to the same specific Vendor X component, because that crucial link was buried in the free-text of individual reports. Furthermore, you never have the possibility to go back and look at your documents in a different light to pull new insights.
Intelligent document review tools automatically scan these investigation reports and immediately tag the critical fields (e.g., vendor, equipment ID, SOP reference). This allows the system to instantly identify that "Vendor X" is a recurring root cause, allowing you to launch one focused supplier audit instead of fighting 15 separate fires. And it saves time too, as investigators spend less time sifting through information and more time implementing preventive measures.
You may want to check out how Augmend can help, if you would like to tackle important unstructured data challenges Because better product quality rests on better quality data.
Do you want to see Augmend in action?
If you're looking to improve your Quality department's productivity and make better use of unstructured data, Augmend might be the solution you've been waiting for. Click the Tackle your challenges button in the navigation bar to discuss your data challenges and check whether we could solve them using Augmend