Scaling Digitization for Today’s State Archives
Discover how state archives can build a scalable digitization operation that serves archival collections and agency records—without compromising quality, compliance, or efficiency.
State archives are increasingly asked to support digitization initiatives for other government agencies. These shared-service opportunities can help justify technology investments and expand the archive’s role in statewide records modernization. They also introduce new demands around capacity, throughput, workflow management, and the variety of documents being processed.
In this on-demand session, see how a common capture platform can help teams handle diverse record types and volumes while maintaining consistent quality and operational visibility.
Challenges Covered
- Growing digitization backlogs and limited staffing resources
- Diverse record types, sizes, and physical conditions
- Manual indexing and metadata capture
- Limited visibility into workflow performance and bottlenecks
What You’ll Learn
- Strategies to increase digitization throughput and productivity
- Best practices for improving capture consistency and supporting FADGI-aligned image quality
- Ways to automate document classification, indexing, and metadata extraction
- How one state archive funded infrastructure upgrades while extending digitization services to other agencies
Technology Featured
See how ibml FusionHD® production scanners and ibml Coretex® intelligent capture software can streamline document capture, automate classification and indexing, improve operational visibility, and manage multiple applications and document types from a common platform.
Here is a list of audience questions and answers from our webinar
What documents still need a flatbed, overhead scanner, or other specialized scanner?
While ibml Fusion can handle a wide range of document types, including damaged records, onion-skin paper, mixed batch sizes, checks, and fragile archival documents, there are still certain materials that are better suited for specialized capture devices.
Examples include:
- Bound books and ledgers that cannot be disbound
- Oversized maps and engineering drawings
- Rare artifacts requiring non-contact overhead capture
- Extremely fragile historical materials that cannot safely move through a transport path
- Photographs or museum-quality materials requiring specialized imaging standards
Many archives use a combination of technologies, leveraging overhead scanners for specialized collections and ibml Fusion for high-volume production digitization. The goal is to ensure every document is captured using the most efficient and appropriate technology.
How accurate is the AI classification?
Accuracy depends on document quality, document diversity, and training data, but organizations typically achieve very high classification rates once document classes have been configured and validated.
One advantage of Coretex is that it does not rely solely on traditional templates. Instead, it analyzes document content, context, and structure, making it more effective when dealing with diverse archival collections and unstructured documents.
More importantly, Coretex uses confidence scoring. If confidence falls below a defined threshold, documents are routed for human review rather than being automatically classified incorrectly. This allows organizations to balance automation with quality control.
How accurate is AI metadata extraction compared with manual indexing?
Manual indexing traditionally achieves accuracy through human review but can be costly, slow, and inconsistent. Human operators make mistakes, especially when processing large volumes.
Coretex applies consistent extraction rules and confidence thresholds across every document. In many environments, organizations find that AI-assisted extraction delivers equal or better consistency than manual indexing while dramatically reducing labor requirements.
The key benefit isn’t simply accuracy. It’s achieving high accuracy at scale while only requiring staff to review low-confidence exceptions rather than every document.
Can the AI read handwriting?
Yes.
As demonstrated during today’s webinar, Coretex can classify and extract information from cursive handwritten records, including historical documents, vital records, and property records. The solution combines OCR, intelligent recognition technologies, and AI-based data extraction techniques to interpret handwritten content.
As with any handwriting recognition technology, results depend on factors such as:
- Legibility
- Writing style
- Age and condition of records
- Image quality
Coretex applies confidence scoring so uncertain values are routed for human review rather than guessed.
How much training is required to add a new document type?
One of the major advantages of Coretex is the simplicity of onboarding new document types.
Traditional capture systems often require:
- Template creation
- Zone definition
- Script development
- Ongoing maintenance
Coretex uses natural language prompts and AI-assisted configuration, allowing business users to define extraction requirements without extensive coding expertise.
Many new document types can be configured in hours rather than days or weeks using a small sample of about 10 documents per document type.
Can we use Coretex without buying an ibml scanner?
Absolutely.
Coretex is designed to process information regardless of image source. Documents can be imported from:
- Existing scanners
- TWAIN devices
- MFPs
- Electronic forms
- Email attachments
- Digital repositories
- Fax systems
Many customers implement Coretex first and leverage existing capture infrastructure before deciding whether to modernize their scanning hardware.
How much volume is needed before an internal operation makes financial sense?
There is no single threshold because every archive has different staffing, outsourcing costs, and project volumes. Although typically most of ibml customers are digitizing more than 30,000 pages per day.
The better question is: “How much money is currently being spent externally on scanning and digitization?”
If agencies are already outsourcing projects valued at hundreds of thousands of dollars annually, bringing even a portion of that work in-house may help justify technology investments while building permanent digitization capabilities.
A business case analysis typically evaluates:
- Current outsourcing spend
- Internal staffing costs
- Projected volumes
- Equipment requirements
- Long-term operational benefits
Can ibml help us build the business case?
Yes.
This is one of the areas where ibml frequently works with public-sector organizations.
We can help evaluate:
- Current digitization workflows
- Existing bottlenecks
- Outsourcing expenditures
- Staffing impacts
- Productivity improvements
- Funding opportunities
Many archives find that the business case extends beyond cost savings and includes improved records access, preservation quality, operational visibility, and support for statewide modernization initiatives.
Can this integrate with our existing ECM or records-management system?
Yes.
Coretex is designed to complement existing information management environments.
Metadata and images can be exported to:
- Enterprise Content Management systems
- Records Management systems
- Document repositories
- Case management systems
- Archive platforms
- Government line-of-business applications
The goal is to enhance capture and indexing processes without requiring organizations to replace downstream repositories.
Can we continue using our current scanners?
Yes.
Many organizations begin by integrating ibml Coretex software with their existing scanning environment.
As future projects arise, organizations can evaluate whether higher-volume production devices such as Fusion would provide additional productivity gains.
The transition does not need to happen all at once.
Can we start with AI and add high-volume scanning later?
Absolutely.
Many archives begin by addressing their biggest bottlenecks first.
If manual indexing, classification, and metadata creation are consuming significant resources, Coretex can be deployed before scanner replacement.
As volumes grow or shared-service opportunities emerge, organizations can later add ibml production scanners and leverage the same capture platform.
Can different agencies have different indexing requirements?
Yes.
In fact, this is one of the primary reasons organizations adopt intelligent capture platforms.
A property records department, vital records office, court system, and transportation agency may all require different metadata fields.
Coretex can:
- Classify document types
- Apply different extraction rules
- Capture different metadata sets
- Support multiple workflows
All from a centralized platform.
Can we export different metadata for different repositories?
Yes.
Coretex allows organizations to tailor output based on destination requirements.
Coretex can transform and deliver metadata in the format required by each target system, helping archives support multiple agencies and repositories without creating separate capture workflows.
Ready to Scale Your Digitization Program?
Watch the recording to learn how state archives can reduce manual effort, improve consistency, and build the capacity to support broader records modernization initiatives. If you have questions or would like to speak with one of our experts further to see how ibml can help your state archives build a scalable digitization operation.
