Скачать презентацию Changing HMIS Software What Do I Need Скачать презентацию Changing HMIS Software What Do I Need

3f2727e24ba94ab7c749a60dce50ef5e.ppt

  • Количество слайдов: 49

Changing HMIS Software – What Do I Need to Know? When is the Right Changing HMIS Software – What Do I Need to Know? When is the Right Time? What are the Cost Implications? Jan Marcason, Mid America Assistance Coalition Birgit Pauli-Haack, Pauli Systems, LC, Co. C Collier County September 18 -19, 2006 – Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development

Overview • Learning Objectives • Switching HMIS Software – Collier County Experience – Kansas Overview • Learning Objectives • Switching HMIS Software – Collier County Experience – Kansas City’s Experience September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 2

Learning Objectives • To discuss the warning signs that a change might be in Learning Objectives • To discuss the warning signs that a change might be in order • To identify lessons learned from a community that has made a software change • To develop a criteria for deciding when/if to change software September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 3

The Collier County Experience Birgit Pauli-Haack Project Manager and System Administrator Pauli Systems, LC The Collier County Experience Birgit Pauli-Haack Project Manager and System Administrator Pauli Systems, LC Collier County September 18 -19, 2006 – Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development

Scope of Operation • Pilot Group: 6 Agencies, 8 Locations • 55 Users, of Scope of Operation • Pilot Group: 6 Agencies, 8 Locations • 55 Users, of which 6 are Agency Administrators. (=Super. Users) • 1. 0 FTE Project Management & System Administration • Start June 2004 • First agency online Sept 2004 September 18 -19, 2006 - Denver, Colorado • Vendor switch June 2006 • Contract new vendor signed April 2006 • Added. 5 FTE • First agency online July 3 rd 2006 • Expansion plans: 4 more agencies by Spring 2007 Sponsored by the U. S. Department of Housing and Urban Development 5

Warning Signs September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Warning Signs September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 6

Warning Signs • Failure by vendor to communicate server maintenance and IP address changes. Warning Signs • Failure by vendor to communicate server maintenance and IP address changes. • After eight months an agency gave up waiting for a printing solution and quit using the system. • A customized application form delayed an agency from beginning to use the system. • After a ten month period of communication and meetings aimed at solving important agency issues, little progress could be reported. • Detailed tracking of issues had to be implemented to serve as supporting documentation in the effort to achieve resolution of issues. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 7

Warning Signs (Cont. ) • A non-mandated partner agency withdrew from participation when their Warning Signs (Cont. ) • A non-mandated partner agency withdrew from participation when their deadline for resolution of essential issues was not met. • There was not a viable system to demonstrate to prospective partner agencies. • Reporting requirements could not be met and were not addressed by the system. • User Group meeting attendance fell as issues remained unresolved. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 8

Decision Making Process • Changing vendors was discussed and dismissed for reasons such as: Decision Making Process • Changing vendors was discussed and dismissed for reasons such as: – – – Invested time Invested money Not going through the whole start-up again Concern for loss of collected data Cost to begin again. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 9

Decision Making Process • Over time we realized: – Our HMIS was not progressing; Decision Making Process • Over time we realized: – Our HMIS was not progressing; we were not progressing. – Critical issues had not been resolved, an agency had withdrawn. – Morale was getting low. – The lack of trust in a productive partnership with our vendor became a motivating factor to again consider changing vendors. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 10

Decision Making Process (Cont. ) • With the renewal of our current software vendor Decision Making Process (Cont. ) • With the renewal of our current software vendor contract six months away, it was felt we should look at different vendor solutions now in order to keep our options open. • Four weeks were spent talking with agencies, emphasizing the skills and knowledge acquired in our implementation and answering questions and discussing worse case scenarios in a potential switch-over process. • All agencies supported change, with several saying, “This is way overdue. ” September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 11

Decision Making Process (Cont. ) • We discussed the situation with our HUD representative Decision Making Process (Cont. ) • We discussed the situation with our HUD representative when we met to review our goals for awarded grants and to ask about changing the technical submission if we made the decision to change vendors. • An outline of the necessary steps that would have to be taken and a draft timeline were submitted to the steering committee. • In Week One of our scenario, the Board passed a motion to explore alternative software vendor options to possibly replace our current vendor. At this point we still hadn’t made an absolute decision to change. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 12

Assessment • Vendor Workgroup charge by steering committee: – Complete thorough investigation in a Assessment • Vendor Workgroup charge by steering committee: – Complete thorough investigation in a very short time. – Only consider vendors with multiple implementations. – HMIS experience a must. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 13

Assessment • HMIS. info list of 34 vendors down to 8 to 5 to Assessment • HMIS. info list of 34 vendors down to 8 to 5 to 3 • Website, phone conversations with agency and COC users. • Weekly meetings to review findings, preliminary ranking. • Review of RFP drafts and other documentation. • Plan for local review of selected vendors. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 14

Assessment (Cont. ) • RFP Process for each Vendor. – Three week deadline for Assessment (Cont. ) • RFP Process for each Vendor. – Three week deadline for return of written proposal. – One on-site presentation to the HMIS Steering Committee, and others interested in HMIS. – One week availability of a demo test site for agency end users to look at the software unguided. – Note: Don’t hesitate to send additional questions. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 15

Tools For Selection • Computer Lab Testing – Use Case Scenarios – Survey • Tools For Selection • Computer Lab Testing – Use Case Scenarios – Survey • Features List Comparison • Cost Comparison Template (See Tools in accompanying workbook) September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 16

Tools for Selection: Computer Lab • Computer Lab Use Case Scenarios – Submitted use Tools for Selection: Computer Lab • Computer Lab Use Case Scenarios – Submitted use case scenarios from agencies tested by end users. – Be aware: with no formal training, you are asking your users to jump into the deep end of the pool and pretend they know how to swim. – Computer Lab Questionnaire from the HMIS Implementation Guide. (www. hmis. info) September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 17

Tools for Selection: Computer Lab • All of the vendors went out of their Tools for Selection: Computer Lab • All of the vendors went out of their way to work with us and to accommodate the use case scenarios in their demo test site that was made available in our computer lab. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 18

Tools For Selection: Computer Lab (Cont. ) • Computer Lab Survey Results – – Tools For Selection: Computer Lab (Cont. ) • Computer Lab Survey Results – – 18 End Users, all six agencies 2 hours, three vendors • Additional Outcomes – Provided great insight into vendor process and software usability. – Participation in decision making process by users. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 19

Tools for Selection: Features List • Features list provides comprehensive overview of each vendor’s Tools for Selection: Features List • Features list provides comprehensive overview of each vendor’s software features. – Shows what you will have. – Shows what you will not have. • Use Features list as a decision making tool. – Identify what you cannot live without. – Identify what you need. – Identify what you can pass up in relation to the rest of your decision making information. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 20

Tools for Selection: Features list (Cont. ) • Record information on the list accurately. Tools for Selection: Features list (Cont. ) • Record information on the list accurately. • Don’t check a feature as available until after you have tested it and made certain it works in the manner you expect and need. • Don’t check a feature as available when it is only planned for future enhancement. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 21

Tools for Selection: Cost Comparison Template • Emphasizes feature specific, not vendor specific. • Tools for Selection: Cost Comparison Template • Emphasizes feature specific, not vendor specific. • Compare apples to apples. • Ask a million questions. • Watch out for inconsistencies. • What about nonquantifiable costs? i. e. Customized Reports September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 22

Tools for Selection: Cost Comparison Non-Quantifiable Costs • Submit 2 to 5 reports from Tools for Selection: Cost Comparison Non-Quantifiable Costs • Submit 2 to 5 reports from your old system and • Request the prospective vendors to describe – The customization process, – Estimating the cost, – Estimate turn around time – Changes in Data collection forms. – Programming needed to include in database. – Customization in their reporting software. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 23

Tools for Selection: Cost Comparison Non-Quantifiable Costs • Gives good idea on costs to Tools for Selection: Cost Comparison Non-Quantifiable Costs • Gives good idea on costs to come, dependent on your needs. • You will understand if additional programming is likely to be necessary. • Shows how additional data collection fields are handled. • How long does it take for the vendor to create a specific report? • The process is as important as the price. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 24

Communication Protocols • HMIS Steering Committee meeting schedule changed to monthly meetings • Users Communication Protocols • HMIS Steering Committee meeting schedule changed to monthly meetings • Users were informed through their agency representative and during site visits • An outline of needed activities and estimated timeline was presented in every meeting (steering committee, user group meetings, trainings, Board meeting). • After full implementation of current participating agencies, public announcements and recruitment of new agencies will begin again. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 25

Data Migration: Never ending story • Plan for a lot of hours. • Coordinate Data Migration: Never ending story • Plan for a lot of hours. • Coordinate with former vendor—retrieval, costs, be prepared to work with minimal documentation. • Coordinate with your new vendor. Request Tools, forms and process documentation. • Reassure agencies who wonder where their old data are. • What should we include in our contract to avoid this dilemma ever again! • Vendors could be much more proactive. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 26

Time is Cash, Time is Money • Vendor Search & Assessment – Project Management Time is Cash, Time is Money • Vendor Search & Assessment – Project Management / System Administration – Actual time spent in vendor selection process, including analysis and attendance at all meetings. – Approx. 170 hrs. – Agency Staff – Time used for attendance at vendor presentations, demo site testing, vendor workgroup meetings, referral phone calls, research and reading material. – Approx. 300 hrs. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 27

Time is Cash, Time is Money • After Selection – Training (conducting and attending) Time is Cash, Time is Money • After Selection – Training (conducting and attending) – Actual time spent in training or preparation for training – Approx. 160 hrs. – Configuration & Customization – Time spent to configure new software for agencies including meeting with agency staff and discussions (per agency ca. 30 hrs. ) – Approx. 180 hours – Project Management – Time required specifically because of implementing new software, including contract review, organizing meetings, coordination – Approx. 122 hours September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 28

Time is Cash, Time is Money • Data Conversion – Coordination with old vendor Time is Cash, Time is Money • Data Conversion – Coordination with old vendor – Data Extraction hours estimated by old vendor – Approx. 30 hours – Data Conversion research and documentation by PM – Data Import Assessment and Scripting new vendor – Data Mapping and Matching PM – Data Review and Clean-up before import – Approx. 42 hours – Approx. 56 hours. – Approx. 40 hours. – Approx. 30 hours. – Approx. 40 hours Total Data conversion approx. 250 hours. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 29

Lessons Learned • Try using all means available to avoid a switch. Beg, growl, Lessons Learned • Try using all means available to avoid a switch. Beg, growl, apologize, get new people on it, ask for new people, spend some money if you can. • From the first warning sign, document all issues and move aggressively to counteract problems before it becomes too late. Be certain to involve your committee chairperson and Board from the beginning. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 30

Lessons Learned • Don’t plan for or work on data conversion until you know Lessons Learned • Don’t plan for or work on data conversion until you know the new system inside and out, as you won’t know what you are doing until you learn to walk the new system. • Even then: It takes twice as long as you imagined and you have to work twice as hard on it to actually succeed. • We had only six months from the Board decision to investigate other vendors to contract end with current vendor. Six months is not enough time. We were only a small group. Our process doesn’t scale into the hundreds of agencies or hundreds of users. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 31

There is Hope, again. • For us, in our particular situation, we were faced There is Hope, again. • For us, in our particular situation, we were faced with either abandoning the HMIS initiative or switching the vendor. • This process would not have been undertaken without the confidence and trust, support and good will of all the people/agencies involved. We are not done yet, and still have some clean-up and plenty of training issues remaining. • Our agencies express satisfaction that they went through the effort a second time, and • We are now able to move forward and recruit new agencies. We are now bragging about our HMIS and its possibilities. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 32

Switching HMIS Software Kansas City’s Experience Jan Marcason Executive Director Mid America Assistance Coalition Switching HMIS Software Kansas City’s Experience Jan Marcason Executive Director Mid America Assistance Coalition September 18 -19, 2006 – Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development

Warning Signs of Needing to Switch HMIS Software • Agencies begin using other software Warning Signs of Needing to Switch HMIS Software • Agencies begin using other software to supplement your deficiencies • HUD or other funders’ requirements force significant modification • Board is concerned about investing in “old” technology September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 34

Warning Signs of Needing to Switch HMIS Software (continued) • Issues on tracking list Warning Signs of Needing to Switch HMIS Software (continued) • Issues on tracking list too long (vendor or project management issue? ) • Unable to get reports out • Availability of funding for change or upgrade • Contract renewal is looming, had to change on the contract year timetable September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 35

Evaluation of Development Options for Mid America Assistance Coalition • Stay with current product Evaluation of Development Options for Mid America Assistance Coalition • Stay with current product and upgrade • Partner with other non-profit software vendor • License commercially designed software September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 36

Factors to Consider • Total cost of ownership (includes one-time development cost, annual license Factors to Consider • Total cost of ownership (includes one-time development cost, annual license fees, equipment, staffing, etc. ) • Ease of use and administration (determined by current users doing testing and MAACLink staff analysis) • Financial and organizational stability (D&B report on vendors, analysis of company finances) • Overall flexibility (ability to change with changing HUD and other funder requirements and user needs) September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 37

Cost of Ownership Budget (two years) • One-time set up fee: • Annual Operating Cost of Ownership Budget (two years) • One-time set up fee: • Annual Operating Expenses – – – Salaries and Benefits Direct Support Administration Annual Software Development Costs Hardware and Network Maintenance Other Program Expenses September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 38

High availability Failover Mode Transaction level High-availability architecture Architecture designed in a highavailability mode High availability Failover Mode Transaction level High-availability architecture Architecture designed in a highavailability mode with at least n+1, preferable 2(n+1) design. Load balances Active transaction-level load balancing Enhancement Request Procedures Clear process for quickly implementing requested enhancements Custom Report Requests User-driven method of easily producing custom reports without needing to submit enhancement requests. Bug Tracking An automatic systematic approach to managing trouble tickets through to resolution including a defined escalation process. Release Process A predictable and known process in which changes to the application are published and communicated. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 39

Release Log A documented log of all changes made to the application. Version Control Release Log A documented log of all changes made to the application. Version Control A controlled process for managing changes to the base application source code. Back-up Procedures A repeatable and tested process for making copies of critical data to facilitate rapid restoration of data should a system outage or crash occur. Physical Server Environment A clean, secure, and environmentally controlled facility to house the hardware necessary to support the system that includes a redundant source of connectivity to the internet and power grids. Remote Server Management A secure method for remotely accessing the hardware necessary to support the system to conduct software changes, system enhancement, optimizations and trouble-shooting Network Security A method for protecting the hardware and software systems from internal and external attacks Server Monitoring A system for monitoring the status and performance of the hardware and software systems that includes alarm thresholds and notification systems. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 40

Areas of Comparison Best Practice Application Architecture Multi-tiered client-server with web-based front-end Application Structure Areas of Comparison Best Practice Application Architecture Multi-tiered client-server with web-based front-end Application Structure User customizable object-oriented application architecture Application Security Robust internal software security based around user access domains and resources Software Library Usage Leveraged open-source libraries to provide complex application functionality without “reinvention” of the technology needed for that functionality. Operating System While similar to questions about “religion”, one should use operating systems that are robust and easily supported. Database Structure Third-normal form database with strategic de-normalizations used for code tables. Database designed by individuals with an appropriate background. Database Integrity Referential integrity maintained by the database through triggers and stored procedures. Limited usage of cascading deletions and no non-cascading deletions allowed. Time and user stamped modification tracking fields included in records. September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 41

HMIS Vendor Contract Sections • Purpose of Agreement • Appointment (right to market product) HMIS Vendor Contract Sections • Purpose of Agreement • Appointment (right to market product) • Initial conversion and implementation for existing agencies/users • Payment and pricing: (initial conversion and implementation fees, annual maintenance and support fees, price increases, notice of price increases) • Modifications and Enhancements • Proprietary Rights (use of trademarks, copyrights, intellectual property) • Maintenance and Support (hosting, customer and user support, software maintenance and support services) September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 42

HMIS Vendor Contract Sections (cont. ) • • • Term/Termination Release License and Source HMIS Vendor Contract Sections (cont. ) • • • Term/Termination Release License and Source Code Confidentiality Warranties Indemnification Limitation of Liability Insurance Rights in Bankruptcy Mediation-Arbitration General Provisions September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 43

Schedules in Contract 1. 2. 3. 4. 5. 6. Features and Functionality Initial Conversion Schedules in Contract 1. 2. 3. 4. 5. 6. Features and Functionality Initial Conversion and Implementation Services Existing Agency and User Lists Pricing Standard Reports Maintenance and Support Services September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 44

Conversion Issues (Technology) • • Language/name for data elements Unusual Concepts (orphaned households) Transience Conversion Issues (Technology) • • Language/name for data elements Unusual Concepts (orphaned households) Transience of individuals and families Need to connect service/program with other data elements (such as budget, housing status, family composition) • Additional data elements in new system • Size of database ($80 M services in MAACLink) September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 45

Conversion Issues (Personnel) • Training users (initial and follow-up) • Re-training users • Users Conversion Issues (Personnel) • Training users (initial and follow-up) • Re-training users • Users are unfamiliar with extraneous issues that impact performance (tool bars that automatically install, pop-up blockers, etc. ) September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 46

Agency User Financial Implications • Upgraded hardware • Upgraded Internet access • More highly Agency User Financial Implications • Upgraded hardware • Upgraded Internet access • More highly skilled intake staff September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 47

Change is Hard • Agencies rely on software for conducting business and reporting to Change is Hard • Agencies rely on software for conducting business and reporting to funders. • Unrealistic expectations of new system • Blaming system for unrelated issues (poor work habits, lack of administrative oversight, inadequate record keeping, etc. ) • Diminishment of control by lead agency • Unrealistic timetables September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 48

Timetable • • • September 2004 engaged consultant to study options November 2004 MAAC Timetable • • • September 2004 engaged consultant to study options November 2004 MAAC Board voted on vendor March 2005, contract signed with vendor August-October 2005 trained 600 users October 24, 2005 conversion to new software January 20, 2006 software upgrade to new release July 21, 2006 software upgrade to new release Fourteen out of 18 reports are operational Continuing increase in participation No drop-off due to new software MAAC obtained grant to assist agencies with Internet access (high-speed lines) • Price increase scheduled for January 2007 September 18 -19, 2006 - Denver, Colorado Sponsored by the U. S. Department of Housing and Urban Development 49