THE COVID-19 VACCINATION TECH FAILURE

California’s Broken Tech State

Waiting in line to get my first shot, people grumbled about the difficulty of getting an appointment, the aggravation of waiting in long lines and the inconsistent requirements (this site asked you to bring ID, but then needed you to show proof that you had an appointment).  One woman defended the state, “They only had two weeks to get these systems up.”  But that’s simply not true. 

Both the Pfizer/BioNtech and Moderna vaccines had been developed in January, 2020, with concrete plans for testing, as per FDA requirements.  All states knew that the vaccine would be coming and that demand would be high, giving them almost a year to create an integrated appointment system.  But they didn’t.  Most didn’t create a well-thought-out plan to get every possible system involved in hosting vaccine distribution.  Until the emergency use authorization (EUA) was finalized, it seems little thought was given to how people were going to make appointments.

California, a major tech hub, did not shine.  Each provider created their own appointment system, so users had to set up accounts and keep checking each one.  To make matters worse, in Contra Costa County (and probably others), you would generally get dropped after a few weeks of “we’ll let you know when an appointment becomes available.”  For some of us, concerns about Covered California security (the Obamacare site) added to the burden of sharing information to get registered.  Savvy shoppers relied on NextDoor and other social media to provide tips and tricks, which is how I got appointments for myself and my husband.  But the process was arduous and much too challenging for anyone with limited internet access, experience or patience.

None of this made any sense.  The solutions are everywhere, in the private sector.  Let’s use Trivago as a good example of how it could have worked.  Confession:  I don’t know how Trivago implemented their system, but it could easily have worked this way or by using a similar strategy to connect disparate applications.

Start with a front-end where you enter your birthday, your zip code and the distance you are willing to travel (dropdown with fixed options).  Similar to Trivago, you tell them you want a hotel and give them your dates and city.  No complicated registration; not until they find a hotel you want.  This front-end could then poll all local vaccine sites matching your criteria for possible appointment slots through an API (Application Programming Interface).  You’d grab the appointment and then be asked to register yourself.  If no appointments were available, like many travel sites, you’d be asked if you wanted to change your date range and/or your willingness to travel further.  You could also ask to be contacted as new dates were posted. 

An API, for those outside IT, is a way to connect disparate applications together.  When you book a flight on Southwest, the airline often offers a hotel and/or car, which means the Southwest application takes you to hotel and car sites, most likely through an API.  We see this all the time even as we have no real visibility into how it works.  We don’t need to. It simply works.

Tech projects can be enormously difficult even when corporations implement them, but the State of California has a particularly notorious reputation for both wasting millions (and billions) on failed or many-years-delayed projects.  Examples include:

-          Fi$Cal – a budgeting, accounting, cash management and procurement system running at over $1BB spent but not ready after 15 years

-          21st Century Project – to overhaul the state payroll.  $900 MM spent before cancelling

-          Countless failed DMV projects.  And for those using their systems, the last implementation drove the DMV reps nuts, failing to manage appointments accurately.  In one case, 50+ people were offered an appointment for 4PM, which isn’t even an allowable time.  I know—we were lucky to get in under the wire, as we were there early that day.

-          CARES – a new child welfare system - $421MM wasted

Too often, this happens because the state IT management doesn’t look for something similar to copy, but instead thinks that their system had to be unique.  They also rarely consider the impact to users and customers.  In addition, our 50-state laboratory system doesn’t work well, as the winning software never gets beyond state lines.  The equivalent driver registration in Florida can give you a physical license, your plates and registration on a single visit.  Don’t we all want that?

We give our state and federal governments too much power, expecting them to provide instant gratification (like Amazon).  Instead, empower the private sector by copying President Trump in Project Warp Speed.  Offer a reward for the first viable appointment system to companies (and individuals) and let every state and county use the same system. And fire everyone in charge of managing governmental IT.  Let the experts do it, instead.Failed California Projects