The EDRMS implementation project
The Department of Education and Communities (DEC) electronic document records management system (EDRMS) implementation project has been implemented to help DEC improve its information management to meet current and future business needs while containing storage costs. The project will also help DEC be more accountable and comply with its legislative obligations, including its obligations under the State Records Act 1998.
The project follows a detailed EDRMS scoping study conducted during 2007-08; and an EDRMS Treasury Business Case which was approved in 2009 for three-year funding from 2009/10 to 2011/12.
The existing records management system had already been used to manage corporate documents, but now DEC is implementing electronic document management within the same system (TRIM). This includes capturing ‘born digital’ records and, where appropriate, converting paper based documents into digital format through scanning facilities.
The EDRMS project will enable DEC to achieve best practice by efficiently managing all its corporate records in one authoritative system as a ‘single point of truth’. It will provide a way to transparently classify and sentence data from existing and future corporate systems and manage its legal, timely destruction.
The current records management system will be upgraded to a new version on a new technical architecture to support document management as well, and will be gradually extended from its current 6,000 users to up to 10,000 users over time. DEC will be gradual piloting and extending a web interface available in the new version that will be accessed through the DEC Portal.
The EDRMS implementation project is being staged over its three year lifecycle. The initiation and piloting stages are now complete, the development stage is nearly so, and the extension and review stages will follow during the last year of the project.
During the pilot stage, the EDRMS implementation project was trialled in some 18 pilot sites across a variety of State Office, Region and Institute business units between July 2009 and March 2011. These included the State Office human resource (HR) business unit at Blacktown.
The Human Resources EDRMS pilot
HR is of particularly key importance in the EDRMS project because some thirty percent of the records of DEC are HR records. These include records relating to the recruitment and management of teachers, including casual teachers, as well as DEC’s other employees. Many of these records need to be retained for long periods of time and contain sensitive, high risk information.
Until recently, the HR records have been kept in paper form. The sheer quantity of these files were a major contribution to the unsustainable storage costs for DEC as a whole. Access problems across locations as well as security risks meant that their management was an ongoing issue for the organisation. However, as much of the paper was received from external sources, simply 'going digital' was not an immediate option. Rather, it was clearly recognised that HR could benefit from the introduction of digitisation so that the records could be managed digitally thereafter. While much of DEC’s EDRMS project was never going to involve digitisation at all, for some key areas such as HR digitisation was the best immediate solution.
In 2007 there was a need identified to investigate enhanced records processing and management. As a result a business analysis was conducted to consolidate findings and develop an action plan. Subsequently, in 2009, the Department formed a project team, comprising staff both from the central Records Management Unit and from HR itself, to spearhead the pilot in Blacktown and, using EDRMS, to begin examining ways to manage the HR records better and ‘move into the 21st century.’
The project team determined that the aims were to:
- ‘go digital’ i.e. eventually remove paper-based processes
- reduce the storage space required
- create efficiencies, i.e. to reduce time lags in waiting for records to be found and delivered between offices and individuals, reduce money spent on both storage and transportation and ensure more efficient use of staff resources in handling records
- improve consistencies in recordkeeping practices across offices.
The end goal was a personalised and adaptable solution which was uniform across all areas of the organisation, and which satisfied business requirements.
The introduction of digitisation technologies was seen as an essential element to achieve the aims. However, the project team clearly recognised that digitisation was not going to be the full solution.
Digitisation was simply a tool in a broader process to improve and update business and records management practices concerning human resource management.
The HR pilot fell within the remit of the Department-wide EDRMS project, its original Treasury Business Case, and its approved budget. Central Records Management Unit staff played a key role in areas such as helping engage HR senior management, facilitating HR business needs analysis, customising HR training provision, and supporting HR project team members throughout the pilot experience. However a very considerable proportion of the resources required for the planning, delivery and management of the HR pilot came from within HR itself, from the HR members of the team inputting the expertise and enthusiasm to complement that of central Records Management Unit staff, and generally working within existing staffing and budget constraints.
A business case outline was put forward to the Human Resources Directorate Budget Committee to explore the possibility of digitisation and its introduction as a staged pilot in Staffing Services, Blacktown State Office. The scope included expanding the pilot across other offices of the Human Resources Directorate if the pilot proved to be successful. The pilot was endorsed by senior managers in Human Resources Directorate and supported by Administration Directorate and Information Technology Directorate.
It was decided that the pilot and associated testing would be carried out at the Blacktown State Office as the project was being driven from there. The Blacktown State Office project team would then become the ‘centre of excellence’ and an advisory service for the rest of the organisation on digitisation issues.
The project would then be rolled out across the multiple sites that manage HR information including:
- the Employee Service Centres in Wollongong, Newcastle and Bathurst
- the Corporate Employee Service Centre in Oxford Street, Sydney
- State Office at Bridge Street, Sydney.
This rollout is still underway.
No back-capture digitisation was considered to be ‘in scope’ for the project (i.e. no digitisation of existing files) due to the sheer quantities of the backlog. However, from the start of the pilot and roll-out to each area all incoming paper-based correspondence was to be digitised.
In planning the pilot, the project team identified the need to:
- ensure the pilot reflected the aims and fed back into the delivery of the wider DEC EDRMS project
- understand the technical requirements and choose the best technical solutions to meet them
- develop processes and procedures for digitisation
- conduct a needs analysis to determine business needs and recordkeeping requirements
- adapt technical solutions and processes to address business needs and recordkeeping requirements in the most efficient way possible
- build a rigorous quality assurance process at the outset
- manage the process of change for records staff, HR staff and other staff within the organisation.
At the outset of the pilot the project team at the Blacktown State Office:
- determined the technical requirements including computer and scanner specifications
- determined the basic metadata requirements
- selected the best hardware and software providers for the needs
- installed and tested the technology
- tested the integration of the software with the EDRMS.
The plan was to automate as much metadata collection as possible and this was assisted by the selection of Optical Character Recognition (OCR) technology.
An important factor in the choice of software was that it already interfaced with their EDRMS to enable the swift capture of images. The software also had to be flexible so that it could be easily modified to meet the identified needs in each area, as there was not one set of standard processes that could be rolled out in all offices.
Testing allowed the team to determine a number of issues including:
- common causes of scanning errors and how they could be resolved
- what standard computer specifications they should adopt (more powerful processing computers without Internet access and other applications worked better)
- the effects of Multiple Windows platforms (32 bit and 64 bit, XP, Vista and Windows 7) on the efficiency of digitisation
- the effects of scanner driver updates and software conflicts
- the effects of dirty scanners on image quality
They also had the opportunity to develop and test future enhancements or ideas and ensured they considered how updates to the records management software and the replacement of older scanners with new ones would affect the operations.
Processes and procedures
The project team determined initial processes and procedures to ensure digitisation could be undertaken as quickly and efficiently as possible, while retaining a high degree of quality control. The actual digitisation was centralised and performed only by selected staff.
The project team automated as much as possible to reduce the burden on scanner operators. Their set up included:
- automated setup of the scanning and RM system file types, file compression settings and RM system database selection
- automated setup of the document scanner driver setting, e.g. dpi and colour setting
- automated setup of the scanning definitions directory so each Unit would only see the ones relevant to them
- automated profile setup to lock the scanning and computer setting to ensure consistency and to stop the fiddlers
- automated document scanning file titling population to ensure consistency and to reduce the error rate
- the automatic capture of essential metadata when the document is sent to the EDRMS including: creator, time of creation, document title and any notes
- adoption of a simple one click scanning with data verifying only approach
- locked down file formats, resolution settings etc.
- automation of computer user profiles
- the development, approval and application of standard forms and templates for record types and mail merges using upper case fonts to facilitate metadata collection (these were created by staff performing the business and forwarded with relevant documents to scanner operators)
- integration of barcodes into forms and templates where possible
- the automatic population of templates, where possible, from other databases
- bulk scanning where required.
The team determined that, using these strategies, they could automate 99.9% of metadata collection, greatly reducing the potential for human error.
Processes and procedures were also put in place for managing original paper records after digitisation. This involved ‘day boxing’.
One small but essential rule that needed to be set was how to distinguish between physical and digital files in the EDRMS. The project team came up with the following distinctions:
DIGITAL = Digital file only, no physical file created (including scanned images)
PHYSICAL = Physical paper file only, no digital images
PHYSICAL/DIGITAL = Physical file with digital and scanned images attached
As a result of setting these distinctions, files are identifiable and searchable in each category.
The project team used the internal records management expertise within DEC to conduct a business needs analysis to:
- understand the context in which the records were being created
- identify and document business processes – the ‘bigger picture’
- determine the recordkeeping requirements within those processes
- determine how and what should be scanned and where digitisation could be introduced into the processes
- determine at what point the documents should be scanned
- determine appropriate naming conventions for documents and files, other metadata and disposal requirements for the records
- determine appropriate access and security that need to be applied to the records.
Their aim, as expressed by one HR manager, was to 'make records the oxygen that flows through the body of HR.'
The needs analysis was conducted in consultation with the staff in the relevant business units. This enabled staff to express their views and have a sense of ownership of the process. However, the translation of the needs into processes was done by the project team. The aim was to make processes as simple as possible for staff.
Standard document titles were determined for the records produced by each business unit. Principles for document titles included that:
- they were to be concise, meaningful, consistent and accurate
- acronyms and abbreviations were not to be used.
Where these conventions are adopted, staff can read document titles in the EDRMS and know what they are without having to open the images.
All documents had to have a ‘parent’ file folder in the EDRMS, also with standard titles.
Other metadata collection
The records team determined what other metadata needed to be collected about the digital images. This included the minimum required metadata which was captured when the records were saved into the EDRMS.
The team determined that one of the important metadata elements, that had to be kept for personnel files of employees and teachers, was date of birth. This information had not always been kept in the past in relation to these files. Without it, some of the records could not be managed appropriately.
For example, in General retention and disposal authority - Administrative records, records documenting the appointment and service of general employees where an adequate summary record is not held, must be retained for a minimum of 75 years after date of birth, or 7 years after last action, whichever is later, before being destroyed.
Forms and templates
The needs analysis revealed what forms and templates were required for particular business units. Where possible these were to be standardised across different business units.
A lot of work went into creating these forms and templates to ensure they were simple to understand and complete and could facilitate swift digitisation and metadata collection.
By trial and error the team worked out that for scanning and metadata extraction:
- some fonts worked better than others (mono-spacing fonts like Courier are better)
- uppercase worked better than lowercase or mixed case
- use of barcodes increased accuracy
- spacing needed to be provided around the terms being extracted
- it is best to automate and forced form data validation were valuable to use where possible.
They also set standards including that:
- dashes (-) were to be used as separators, e.g. between parts of titles
- ID should be captured in 3 fields (this way, if one field is corrupted and is missed in quality assurance, the information remains accessible).
Drop down menus were also provided wherever possible where choices had to be made for metadata population of forms to reduce the potential for error.
Work was also done on where the metadata on the forms would map to in the EDRMS.
Once checked and authorised forms were made read-only. The team added a small picture of a scanner to each approved template and form to indicate it was approved and scan friendly.
If particular business units were attached to the existing forms, the project team did not force them to re-engineer the forms. Rather, they worked with the existing forms and manipulated them. This enabled staff to find comfort in familiarity and be less resistant to change. Gradually the staff members themselves began to make suggestions about how to make the forms more efficient.
The records team ensured that the access and security needs for sensitive/privileged information were identified (in liaison with staff and the sponsor in the business unit). These were protected using the security and access controls available in the EDRMS.
In addition, the digitisation project team:
- ensured that forms and templates, when approved, could not be changed
- established access rights to use scanners
- locked or restricted the ability to change scanner software settings to reduce the potential for security breaches.
Quality control was an important element of the project to maintain consistency and reduce the error rates. The records are high risk and may sometimes be required in court. If they could not be found and retrieved quickly the digitisation effort would be wasted.
Determining exactly what was needed beforehand, including scanner settings, and automating as much metadata collection as possible by using forms and templates greatly reduced the potential for quality problems.
In addition, three levels of quality assurance were devised:
- Initial quality checks were performed by the operators of the scanners as soon as the image was scanned and before they would be captured in the EDRMS. Rescanning would occur immediately if errors were detected.
- An integrity report was run each day of the digital images produced. These were then checked by the team leader of the digitisation project. If errors were detected, they would be sent back to the operator responsible to be rescanned.
- A third quality check was then performed by another supervisor against set criteria. About 5 to 6 items in a box were checked. If errors were detected, they would be sent back to the operator responsible to be rescanned.
Sending any digital images with errors back to the operator who digitised them had the advantage of reinforcing areas of responsibility. It also enabled the team to detect when a staff member was unsuited to digitisation work.
Original paper records were initially retained for one year before being destroyed to allow for quality control checks. This will be reduced as the error margins have been significantly reduced.
As the pilot progressed it was determined that it was better to conduct quality checks before the boxes of original paper records were sent to storage. That way time could be saved on the retrieval.
Managing the human element
The project team also put a great deal of thought into managing the human side of the pilot. All agree that this was an essential step in achieving success.
It was the project team’s view from the start that, with a digitisation project, 'all hinges on people management.'
Strategies the project team determined to manage the ‘human element’ included:
- obtaining senior management sponsors for the project as a whole – the technology was demonstrated and the focus was on the efficiencies that could be gained
- only dealing with one area at a time (not the 'big bang') approach – they would learn, debrief, tweak and adjust the approach before moving to the next area
- obtaining champions at the manager level in each area when the area was being targeted – they would demonstrate the software and its capabilities to the manager and ‘sell’ them on the concept and the managers would then tend to become sponsors in the areas
- identifying key people in each area that could be used to support the project and assist other staff members further in digitisation – generally the people who could have the confidence and ability to sell the project in a positive light (these staff received additional technical training)
- negotiating time frames that were suitable for the areas included in the pilot – as it had to happen in tandem with the usual work
- building strong relationships with the staff - allowing them to feel a sense of ownership and acceptance of the change
- conducting the needs analysis upfront – to determine the context and requirements of each area and to ensure that identified requirements were met
- providing onsite training by professional trainers, customised from the overall EDRMS project training model to meet the needs of HR specifically – tailored for different levels of knowledge and computer ability, kept as simple as possible
- supporting staff by having a training manual to refer to afterwards (this included screen shots to assist understanding)
- supporting staff by having a ‘help desk' at the Blacktown State office
- assisting staff with the change of culture - so they could deal with not having physical documents anymore and adapt to the use of digital images.
The conclusions drawn and lessons learned from the HR EDRMS pilot are applicable both to the extension of EDRMS to other parts of HR, and to the wider DEC EDRMS project.
While the HR pilot – and the other pilots across DEC – are now largely concluded, the rollout to other HR areas within the DEC, and across DEC as a whole, is still underway.
Listed below are some keys to success, problems encountered, unexpected benefits and changes ahead.
Keys to success
Generally the pilot has been highly successful in establishing model practices: so much so that areas where it is being rolled out have hardly needed additional support from the help desk!
Project team members see that the keys to success are:
- a cohesive project team with a passion for excellence and continuous improvement
- a slow rollout – elements of the process could be refined and improved over time, digitisation processes were not rolled out until areas were deemed ready
- careful testing of the software and development of customised technical solutions
- as much automation of metadata as possible to reduce human error – the forms and templates were invaluable for this
- the needs analysis – although it took time, it ensured that the technical and recordkeeping solutions were correct and closely aligned to business practices
- the change management initiatives to bring records management staff, senior management, managers and staff on board
- quality assurance planned from the start which enabled processes to be refined and accuracy to be improved immediately
- the maintenance of clear documentation about the EDRMS project as a whole including planning, processes, specifications, the quality assurance undertaken, outcomes etc. This will enable the organisation to prove they meet requirements of the General retention and disposal authority: original or source records that have been copied
There were small problems experienced which needed refinement as the pilot progressed.
For example, initially, little thought was given to the management of the original paper records after digitisation. Prompt decisions has to be made regarding:
- how long to store the originals for
- how to reference the physical location
- what criteria to check the digital images against for quality assurance
- how to ensure that paper records were not recirculated or removed from the containers.
Other problems encountered included that:
- the lead time for rollout was underestimated. There is a lot more ‘hand holding” involved than initially expected
- the space requirements were underestimated. More space was required around workstations to conduct the digitisation than was initially expected. In addition, more office space had to be allocated for the storage of original paper records awaiting digitisation
- more onsite training was needed than was initially factored into the project. Training needs to be accessible to staff on an ongoing basis due to staff turnover and changes over time
- initially all scanning has been in colour. This has lead to a large file size than is always necessary. The team are now investigating add-ons to the software that allow it to pick up and scan in colour only when necessary.
The pilot has already resulted in some unexpected benefits:
- Senior HR management as well as staff at all levels are engaged in the process and so share greater understanding of the importance of good recordkeeping.
- Human resource management staff are recognising that keeping full and accurate records is a key part of the jobs, not peripheral.
- Human resource management staff feel empowered to suggest ways that the process can be streamlined and improved further.
- The rigour applied to the titling in the digitisation project had meant that human resource management staff are beginning to apply more rigorous to titling the other digital records in the EDRMS.
- Records staff now have a higher profile, a positive learning environment and considerably better skills and experience so they too are feeling empowered.
- There is now a more collaborative relationship between ICT and records staff has been established.
Both the project team, and the wider DEC EDRMS project team, are aware there are still challenges ahead. They want to continue the processes of continuous improvement and this requires being vigilant and responsive throughout the remainder of the project.
They are looking to improve other records creation practices as well.
Both teams are actively encouraging staff to put links to digital records in the EDRMS in their emails rather than attachments. They are also now investigating the automatic capture of voice mail and its metadata.
They have determined that an additional challenge they face is to 'make it digital and keep it digital', i.e. that staff need to be supported and monitored so they do not fall back into old paper-based practices and thinking.
The wider DEC EDRMS project team are aiming ultimately to make the records accessible through the EDRMS anywhere, e.g. via laptops or IPADs, so that staff do not ever have to print records out.
Records management staff will also need to be vigilant about the management of the records in digital form – both born digital records and those created through digitisation processes. They have recognised the need to ensure that the digital preservation is planned for (e.g. with migration strategies to new archival formats etc) and managed appropriately so the records are accessible and usable as long as they are required.
Note: State Records NSW would like to thank the Department of Education and Communities for their willingness to share the information in this case study with other organisations.
For copyright, see our copyright policy.
Published June 2012