Top 10 Considerations When Building a Production-Level Document Conversion Environment
Even with careful planning and reliance on industry-leading technologies, it’s not uncommon for records managers and their IT support teams to experience frustration and disappointment with their document capture and conversion workflow environment. Time and again, there are complaints that the system was not as easy to use as promised, or that it did not provide the high levels of automation and efficiency that were anticipated.
Often, organizations do not ask the right strategic questions during all of the lifecycle phases of their document capture and conversion production environment, including, planning, implementation, production and decommissioning (see “Production Environment Lifecycle” graphic). Unrealistic expectations about the capabilities of automated systems or the use of technology for technology’s sake often undermine an organization’s ability to achieve its goals. But that frustration and disappointment doesn’t have to be the norm. By considering the following 10 simple recommendations, organizations can ensure that their document conversion environment meets – or even exceeds – their expectations:
1. Keep the End Use Case in Mind – When considering the implementation of a document capture and conversion production environment, organizations need to clearly consider what features are absolute necessities for the types of documents they will be converting. Many times organizations incorporate features that they think they need before seriously understanding the additional costs and processes that may be required to utilize those features. For example, if an organization wishes to create a long-term archive, it may choose believe that compliancy with a higher level of PDF/A standards (PDF/A-1a) may be best. However, what they may not consider is that complying with PDF/A-1a may add significant implementation costs because it requires human intervention for tagging metadata. So, unless an organization must be able to have its documents reflow over a mobile device, compliance with a lower level of PDF/A (PDF/A-1b) is typically sufficient. In addition to evaluating what features are absolute requirements, it’s important to be as precise as possible during the specification phase. By defining what is needed for its own unique use cases and why it is needed, an organization can create clear requirements and expectations can be met.
2. Set Realistic and Measurable Goals – Organizations must consider both the types of documents to be processed and the timeframe in which the project must be finished. For example, if newspapers are being digitized and full text searchability is required, an organization must take into account the amount of time it will require to run fine print through an OCR engine. Also, manual processes, such as inputting metadata or indexing documents, can impact the overall throughput of the workflow.
3. Use Standards Whenever Possible – Proprietary file formats often leave organizations beholden to a single vendor. Additionally, use of home grown solutions may cause and organization to be reliant on select, highly trained individuals. Both of these scenarios may cause costly problems in the future. By using standards-based technologies and software solutions with open architecture, organizations can easily find experts to manage and build new, customizable conversion workflows.
4. Know and Manage Existing Resources– For every document conversion environment, an organization must balance three key resources: money, time and features. So, if time is of the essence, an organization may have to sacrifice certain features and incur greater costs for increased processing power to achieve its goals. Or, if features, such as 100 percent accuracy of extracted data or perfect image quality, are vital to the resulting electronic records, the conversion process could ultimately take more time or cost more than expected. For example, for optimal accuracy, an organization may need more quality assurance and data entry personnel to correct the extracted content. And, for perfect image quality, an organization must take into consideration that the file sizes will be larger, requiring more storage and bandwidth for accessibility.
5. Don’t Automate for Automation’s Sake – Even though a technology exists, it doesn’t necessarily make it a perfect fit for every organization’s needs. There is not always a cost benefit for complete automation. In fact, sometimes a human brain can be more effective than a computer brain. Ultimately an organization must assess what absolutely requires automation and where human resources can be best, and most efficiently, used.
6. Measure Throughput of Alternative Workflows – Organizations should analyze a variety of workflows to determine which combination is the best fit for a project. For instance, if an organization has thousands of documents to enter into the system each day, it may be best to implement data extraction software and focus human resources on quality assurance of the electronic content. Conversely, if there are only a handful of documents to enter, it may better to manually enter the data instead of investing thousands of dollars on software. By seriously considering alternative ways to reach the end goal, organizations can select the best combination of resources and features to achieve the most efficient and cost-effective workflow.
7. Communicate Effectively – Vital to the success of any project is strong, regular communication between an organization’s key stakeholders – such as the internal IT team and the records managers helping implement the system – and the vendors that are building and maintaining the document conversion environment. The internal stakeholders should clearly and effectively communicate so that all expected goals and targets are being met. While, vendors should be proactive, available and willing to tell the customer “no” when asked for costly features that are not well-suited to the implementation.
8. Ensure Quality Assurance Through Testing – To ensure that the environment can handle the volume and types of documents being converted, an organization should run a large enough sample batch of actual documents, so the test is representative of what they plan to convert. This step will enable organizations to better understand how the process will work and how different types of documents will behave during conversion. By doing so, they can mitigate the risk of processing mistakes, address concerns that may result in overall poor performance of the conversion workflow and ensure that the workflow is achieving the goals set in step two above.
9. Continuously Improve Processes by Tracking Live Data – Once production begins in the new document conversion environment, organizations should monitor the progress of the workflow in real time. This step enables them to identify where bottlenecks occur and adjust both automated and manual processes to further improve productivity.
10. Re-Evaluate the Production Environment After Each Job – By considering the conversion data generated and evaluating the weaknesses discovered as each job, organizations can continue to fine tune their environments to achieve even greater efficiency.
Building and deploying a document conversion environment can be a complicated process. But by asking the right questions, planning in advance and selecting the right partners, organizations can avoid the mistakes that create problems and lead to disappointment and frustration with their new workflow processes.
About the Author
Mark McKinney, President of LuraTech, Inc. (www.luratech.com), leads the company’s American subsidiary, where he specializes in new business development, corporate strategy and marketing. McKinney has more than 12 years of executive and product development/management experience, having held various marketing and management roles at Wyse Technology, Axolotl Corp., Siemens and Uni-Data.