PURPOSE

To authorize the BWP General Manager to approve a change order to the Indus Utility Systems, Inc. contract to provide additional implementation support services for the Customer Information and Billing System.

BACKGROUND

In September 2002, City Council approval was received to enter into an agreement with Systems and Computer Technology, Inc. (SCT) for a new Customer Information and Billing System (CIS).  SCT has been purchased by Indus Utility Systems, Inc. (Indus) who is now responsible, in cooperation with City staff, for fulfilling the agreement to license and implement the new system. 

In July 2001, prior to embarking on this project, Langham Consulting Services, Inc. (Langham) conducted a review of the City�s staff and infrastructure resources to determine our readiness to proceed.  They made several key assessments and recommendations:

           1.                  Dedicate the current staff person maintaining the legacy billing system to the CIS project full-time. 

           2.                  Provide back-up support for the legacy system operation.

           3.                  Hire a full-time Database Administrator (DBA) to support the new system.

           4.                  Upgrade the City�s network links to accommodate expected heavy use of the new system.

           5.                  Assign appropriate staff, dedicated to project implementation.  Hire temporary staff to back-fill critical staff functions.

           6.                  Provide adequate work space for the vendor and project implementation team.

           7.                  Hire an Application Specialist to provide end-user support.

           8.                  Ensure qualified staff is available in the areas of network, systems administration, operations, and Oracle systems.

All recommendations were implemented, or are in process.  We have in place a cross-functional implementation team, including the current legacy system programmer and the new DBA.  Back up for the team members has been provided to maintain all on-going functions.  The network has been upgraded and work space has been set up for the vendor and implementation team.  In addition, BWP has secured the services of Langham to provide expert assistance in managing the project.

An Application Specialist will be hired as we get closer to our �go-live� date.   And, to realize the last recommendation, discussions were held in 2001 with Public Works, Finance, Information Technology, and Treasurer to ensure all stakeholders understood the level of commitment required and had the resources available, or planned to have the resources available, to proceed.

Today we are halfway through project implementation, with 37 of 78 deliverables completed.  We have expended 43% of the total project budget of $4,678,300 and are on track to �go-live� with production on the new system on or before the end of June 2004.  However, as we have neared this halfway mark, we have found that we are missing critical skills within our implementation team to successfully fulfill our goals.  The critical skills are in the areas of information technology project management and technical competence.

ANALYSIS

When we began planning for this project, it was understood that Information Technology (IT) needed to redevelop its capabilities in some areas.  Significant progress has recently been made in the area of network upgrades and we have confidence that with the changes being made the network will support the new CIS as well as the balance of the City�s requirements.   However, progress in other IT areas, critical to project success, have taken longer than planned.  Halfway through the implementation, we are at a critical juncture and must act swiftly to reestablish the probable success of this project.

In light of the situation and project timing, staff considered three options to put the project back on track.  One, put the project on hold until the required staff resources have been sufficiently developed.  Two, completely outsource the implementation and operation of the CIS.  Or, three, �outsource to in-source�.

The first option, putting the project on hold, is not an optimum solution.  Staff time, effort, and project expenses expended to date have been put to maximum use and  a significant amount of very good work has been completed.  It would be a shame to lose this work and the momentum that has been built up over the past months.  This project can succeed with the addition of focused technical skills.

The second option, completely outsourcing the CIS implementation and support functions, seems, at first blush, to be a workable alternative.  There are several companies that provide these services.  However, staff decided it would be best to work directly with Indus rather than invite an unknown third party, at this point, to join the project and expose ourselves to another element of risk unnecessarily.  Since Indus owns and markets the software, having them provide the support places the risk of failure firmly on their doorstep, mitigates potential finger pointing, and increases the likelihood of Indus�s ownership and accountability for a successful implementation.  Indus was asked for an estimate to take over project implementation and they returned with a preliminary estimate of $2,000,000.  This price includes Indus assuming full responsibility for setting up a data center, complete management of the new data center, systems administration, conversion, database administration, and operations.  This level of support assumes 6.5 full-time equivalents assigned to the project from September through June 2004.

As we explored this option, we found from other cities that outsourcing CIS implementation and hosting was not the panacea it was hoped to be.  In the majority of cases, costs escalated year after year and service levels declined.  More often than not, the outsourced function was eventually brought back in-house.  Hence, while outsourcing in general can be a cost effective alternative and should be considered, outsourcing CIS implementation and hosting is an expensive option and does not appear to be a durable long-term solution for managing CIS technical support.

Given what we learned about outsourcing, staff focused on the last option, �outsource to in-source�.  This means that we would outsource sufficient technical staff activity to get us through implementation.  And, rather than displace IT staff completely, the outsource vendor would work in cooperation with IT staff, training them to eventually reassume responsibility for CIS administration and operation after the system is up and running.  For the same reasons mentioned above, we asked only Indus for an estimate.  Indus�s estimate to perform this level of work is $1,000,000.

The specific areas of support for the �outsource to in-source� option are similar to the full outsourcing option.  The primary difference is the level of responsibility assumed by Indus for each activity.   Initially, Indus will take the lead in almost all areas.  However, over the duration of the contract, IT staff will take on greater and greater levels of responsibility as they move toward total independence. This level of support requires Indus to provide 3 full-time equivalents assigned to the project from September through June 2004. 

The �outsource to in-source� option allows us to secure the skills needed to successfully implement CIS while simultaneously rebuilding IT staff competency.   Indus and IT staff would work hand in hand from now through �go-live�.  Indus would remain on-site after �go-live� to provide post production support for up to two months and to assist IT in handling the inevitable and quirky problems that are bound to arise after cut-over to the new system.    By that time, IT staff will have had sufficient time to achieve knowledge transfer and technical competency to handle the CIS operation independently. To further support the �outsource to in-source� model, staff recommends that IT reinforce existing staff and fill a vacant BWP position with systems project management and administration talent dedicated to the CIS effort.

In the final analysis, it would be unacceptable to go forward with the implementation of the new CIS without adequate assurance that municipal service bills will be accurately produced and efficient customer account maintenance and service will be in place.   Every month the project is delayed would cost a minimum of $150,000 for continuing vendor, consulting, and staff support.  Rather than wait until we push up against the �go-live� goal and then attempt to retrace where we�ve been, it is simply smart to supplement IT staff now rather than wait to catch up later with an unknown ending time and an open- ended cost.

By using Indus to help with the IT portion of implementation while we train and rebuild staff competency, the City wins in many ways.  The project�s success probability is restored, the additional project cost falls from $2,000,000 to $1,000,000, and IT can use this opportunity to develop in-house competency to support the City in successfully managing future IT systems projects.

FISCAL IMPACT

Funds budgeted within the existing BWP capital outlay program for the Oracle project will not be significantly used this fiscal year.  Therefore, these funds are available for other technology projects.

RECOMMENDATION

Authorize the BWP General Manager to approve a change order to the contract with Indus Utility Systems, Inc. to provide CIS implementation support services for an amount not to exceed $1,000,000.

go to the top