by Joe Hanousek
Attorneys’ Title Insurance Fund maintains the largest automated title information plant in Florida, providing title plant services to assist its independent attorney agents in the title examination process. The plant supports over 2,500 online users and stores data and images for more than 100 million public records. Approximately 45,000 documents a day are keyed into the repository. In just the last few years the number of documents processed has increased tremendously. In 1999 the average daily rate was just 26,000 documents.
Faced with the rising costs associated with the increased document volume, The Fund began exploring its options to reduce costs through technology. In 2004 the company launched a pilot project to evaluate the benefits of GIS and automated mapping. After only a few months, the results were evident. Costs went down, the plant work flow became more efficient, and accuracy improved. After accounting for hardware, software, and training costs associated with implementing the pilot, the time to realize return on investment was surprisingly short.
The Old Mapping Bottleneck
The existing paper mapping system had served the company’s title plant operations well for more than 20 years. Here’s how it worked prior to the GIS system.
The Fund’s plant is geographically based and designed to minimize the property records that must be searched to locate a particular piece of property. The Fund employs a unique grid system that allows most properties to be identified based on 2.5-acre sections of a 640-acre square mile. These 2.5-acre areas - called Acreage Codes - streamline the examining process because fewer documents are returned to the examiner during the search. This ultimately saves the examiner time on orders and gives The Fund a competitive advantage in title plant services it offers.
Part of the plant’s function is to index all of the recorded property documents by location. That location is most often listed in the legal description of the property. Legal descriptions come in two varieties, platted and unplatted. Platted descriptions refer to a property that is located according to its Lot, Block, and Subdivision information and make up the majority of the recoded documents that need to be indexed. This index information is keyed into The Fund’s title plant database system (ATIDS). On the other hand, unplatted properties cause a significant bottleneck in the indexing process. Unplatted properties are usually described by a metes and bounds (bearing and distance) description or fractions of a Section, Township, and Range. These properties must often be mapped in order to find the location information to index them in ATIDS. Mapping the legal description identifies the 2.5-acre part or parts (Acreage Codes) of a section that the property falls within.
The Fund maintains a geographic index for 34 Florida counties. Three separate data centers map about 20% of all legal descriptions the company processes on a daily basis. Documents that needed to be mapped took 300% longer to index than non-mapped documents because of the time-consuming and labor-intensive manual mapping process. These orders experienced a slower turnaround time and affected service levels to The Fund’s customers.
The maps are drawn by hand and stored in paper files, making it difficult to share between staff. Branch offices would often map the same documents as the title plant’s mapping staff because the branch offices had no way to access the paper files stored at the data centers. This duplication of effort was costly. And as recorded document volume continued to grow in Florida, The Fund was forced to hire more mapping staff to keep up with the steadily increasing workload. Between 2002 and 2005 the number of staff assigned to mapping increased 15% and currently stands at 45.
The data centers maintained drawers of paper base maps. Each map is 10 inches by 11 inches and represents a one-square-mile section of a county. The paper-based system was problematic. As the paper maps aged, they became brittle, worn, and sometimes difficult to read. Because they are updated constantly, copies of the original maps could not be effectively colocated. Disaster planning and recovery amounted to the mapping department supervisor instructing everyone to “grab a drawer of maps on their way out of the building” in case of fire or hurricane evacuation.
The Fund recognized the need to find an alternative to the aging system - one that employed the latest technology to reduce overall costs even as demand for their services continued to increase.
A GIS-based Automated Mapping Solution
The Fund’s IT Strategy Manager, John Sayers, figured there had to be a way to modernize the mapping process the same way the level of automation had already increased throughout the rest of the plant processes. Sayers’ vision was to use GIS.
The Fund contracted Extract Systems to review its current environment. After documenting the existing mapping workflow, we met with many staff members and stakeholders from corporate headquarters and the data centers. Working closely with the staff during the meeting process, problems were identified and goals were established. The diverse project team identified a pilot program that designated a month-long test of the GIS mapping approach. Because IT resources were scarce, The Fund would only commit to putting the system into production once the results of the pilot project were measured. The pilot project defined the following goals:
We prepared a report recommending off-the-shelf software that would be customized. This pilot was backed up with a customized training program and post-implementation consulting to refine the workflow process further. The solution included a mapping application called IcoMapTM as the main mapping tool because it works effectively in the platform used for the project — ESRI’s ArcGISTM. Extract Systems provided additional GIS services based on its combination of experience with title companies and parcel mapping. Services included workflow consulting, implementation, training, installation, and GIS service work related to the project. By relying on Extract Systems for the GIS service work, there were minimal disruptions to the critical production work flow at The Fund. The pilot was facilitated through The Fund’s Central Florida Data Center using two mappers and the mapping department supervisor. Orange County was chosen for the GIS project internally dubbed “NetMap.”
Implementing the GIS Pilot Project
GIS parcel data was purchased from Orange County, Florida. Extract Systems created the GIS framework and loaded the Orange County data into the GIS system. The underlying foundation of GIS is a database. The Fund chose to utilize a SQL database back end because its IT team already supported SQL databases for other projects. The pilot report and project team also identified that a terminal server environment would be the best infrastructure solution to host NetMap. Multiple users editing in GIS can consume considerable network bandwidth, but having the GIS applications running on servers in the corporate server environment minimized network traffic to the data center, which is at a different physical location than The Fund’s corporate headquarters. It was imperative that the pilot workflows and technology infrastructures be scalable and match what would be used in any future enterprise-wide rollout. This enabled The Fund’s management to accurately analyze the ability to scale the system in the future and minimize any surprises that might be encountered in a large implementation project.
Dual monitors were recommended for the mapping staff to maximize productivity. The work flow requires the users to use their document-management system, ATIDS, and the GIS application at the same time. This dual monitor system turned out to be an excellent low-cost way to boost efficiency. With two monitors, users spent less time toggling between the applications and windows used in the mapping process.
End user training for the mapping staff was crucial to NetMap’s success. Most software or technology solutions involve training that typically teaches users how to operate a particular piece of software. We recommended an in-depth training that included teaching the new workflow, from the time the PC boots up through to completing and filing the completed map. Also, the end users were now able to make decisions that previously only a supervisor would have made. Because many users had previously been mapping by hand, the learning curve of the new workflow varied greatly from person to person, based on their outside PC experience level. Training on how to deal with new choices and educating the users on the consequences of each decision point made the new technology more acceptable and the users more productive.
The Fund required that we come back for a week when the NetMap pilot was launched to address user questions that arose during everyday activities. That insured that users and stakeholders alike had a positive initial experience with the new system.
The Pilot Project Results
Measuring the results is crucial to any new pilot initiative. The NetMap team decided to run this pilot in parallel with their existing manual mapping. After only one week of parallel analysis, it became clear the results exceeded everyone’s expectations.
As a result, the data center immediately placed GIS into production, even though the pilot was only to last 30 days and was not intended to receive budgetary consideration until the following year. After 30 days NetMap was three times faster than the previous process. The Fund’s time study revealed that the traditional method of mapping in the data center required an average of 41 minutes per document. The GIS-based Netmap system reduced the average to 12 minutes. The data center also completes mapping requests for branch offices. When the branch offices are working on policies for large commercial developments, they need the data center to quickly produce maps for these high-value policies. While not part of the pilot, the data center mappers quickly realized that NetMap allowed them to turn these requests around much more quickly. The results were measured here as well. What they found was stunning. Commercial mapping orders that had previously averaged 19.25 hours were reduced to 2.35 hours. Their internal study further illustrated that The Fund could save more than 34 hours per day at the data center by using NetMap — the equivalent of four full-time staff.
There are other benefits that are not as easily measured in hours or dollars but are also significant. For example, when a mapper came across a document containing an incomplete or “unmappable” legal description that could not be located, he/she would post the document to a General Index (GI) for further research. On average the mappers would each post about 15 documents a day to the General Index. With GIS and IcoMap, the mapper can now locate most of these GI properties by searching the GIS system for a parcel number, a street address, or even an owner name and is able to post this document to its correct geographic location. Today, The Fund averages one General Index posting per day.
Another example of important improvements is accuracy. Mapping by hand on paper using a pencil, scale, and protractor does not always produce maps that are precise, and the paper maps assume that each section is a perfect one mile square. In reality, almost no sections are exactly perfect.
To compensate for these inaccuracies, any line drawn on the border of an acreage code line is posted to the codes on both sides of the line. Entering these extra codes is called overposting. Overpostings cause examiners to wade through more documents than necessary when researching the documents that affect a particular property. The GIS system shows every section in its true size and dimensions, and the properties the mappers create are as “survey accurate” as the calls spelled out in the legal description. The GIS-based system completely eliminates overposting. Error correction with the GIS system has a positive cascading effect; the improvements in accuracy cause the system to operate even more efficiently. As more of the parcel layer is created, the need to remap parcels is eliminated. And all of this has the effect of incrementally increasing productivity over time — the more the GIS method is used, the faster the system becomes.
Faced with increasing staffing demands and outdated mapping processes, The Fund found a cost-effective solution with GIS and automated mapping. By hiring a consultant with specific title plant GIS expertise, involving management and end users in the planning process, setting goals and measuring them in the form of a pilot rollout, The Fund has established what will likely become the emerging model for tomorrow’s GIS-driven title plant.
Joe Hanousek is vice president of Extract Systems, Middelton, WI, offering work flow and GIS consulting for title companies. He can be reached at firstname.lastname@example.org. This article is an excerpt from his presentation during the 2005 ALTA® Annual Convention in New York.