IT Procurement as been a hot button issue as one of the largest civilian IT projects – DoD IT-sized in a way – was “bungled” in almost every phase: Cost went 6x original bid, architecture was overly complicated, no system integration concept, no end to end software or data lifecycle management end, and no quality acceptance procedures, criteria, incentive or the like.
We all know some or all those problems. But as the FCW article cites: “Bungled launches didn’t start with HealthCare.gov.” Its everywhere. Its government, commercial, non-profit, everywhere.
Of course, at Xentity we are biased – we believe in take your medicine now approach – design upfront, register and buy back the risks, and then move into agile design, rapid development, and iterative launch as relevancy and the market allows.
1. Do a better job on defining desired outcomes upfront
2. Improve the training options for the federal acquisition workforce to put them on an even footing with vendors
3. Give agency CIOs more budget authority
4. Avoid lowest price, technically acceptable contracts on large innovation-heavy projects.
5. Use agile development strategically and mainly when a project does not require a log of interaction with legacy systems.
To too our own horn, these are many of the fundamental goals that Xentity staff and solution focuses on. In reply to those five items:
1. This is our main emphasis on the design the concept of operations, and requirements for the SOW, and registering the risks and knowing them ahead before procurement. Still allow vendors the flexibility for logical and technical design, but know upfront the various concepts that may come back and know how to score them
2. our fedbiz.xentity.com business management specialists service integrated with our architecture practice allows us to help bring contracting and procurement specialization into helping understand how vendors who may respond based on market analysis results will respond to certain requirements or solicitation frameworks
3. We are setup to help advise CIOs on enterprise portfolio, architecture, capital planning, and segment adoption of CIO services and solutions
4. We agree that the LPTA method does not work for acquiring design, planning, creative, and solution management services. LPTA beltway experts tend to game the systems by replying to architect positions with application developer rates for architects on initial task order and technically it is acceptable, and the rates are 30% better. But its a lot like asking a cook, “can you farm?” Technically, the cook probably could, but wouldn’t you want someone familiar with the subject matter expertise of agriculture economics, farming lifecycles, key risk and success factors or someone who knows food. The solution ends up costing more as MODS occur, the app developer gets replaced, and government pays the cost for missing deadlines and scope creep.
5. This is a biggie! Marketing hyped up Agile as the “it slices, it dices, it julienne’s!”. It is great for new transaction systems on abstracted solutions. It can be good for some feed ETL or integration. But, when you have an immovable object, it doesn’t matter how agile you are. In those cases, you need to conduct architecture design, concept of operation alternative analysis, business case evalution, requirements definition, and register and buy back risk.
All this said, we are very happy to see this attention back onto better design and define up front. Of course, being in this field, it is always nice to be noticed and knowing we are on the right track. More importantly, we believe it is what is needed and the right thing to do for investing wisely with our citizens or customers money/assets.