This site uses cookies that are essential for our site to work. We would also like to use non-essential cookies to help us improve your browsing experience and help make this website better, by collecting and reporting information on how you use our site.

Cost considerations for build vs. buy: healthcare analytics platform

Healthcare Analytics

Approximately 30% of the world’s data volume is being generated by the healthcare industry — a single patient generates over 80 megabytes of data each year. This data holds incredible promise for improving the quality and cost of care, but making sense and full use of it places tremendous operational stress on Healthcare Organizations (HCOs). To take full advantage of their data assets, savvy HCOs are investing in data and analytics platforms that unlock use cases for success in both FFS and VBC payment models.

Consideration points for building vs. buying a healthcare data and analytics platform

Below are some of the decision vectors that go into building or buying a healthcare data analytics platform:

  1. Time to value and opportunity cost
  2. Managing capital and operational costs (CapEx / OpEx)
  3. Staffing
  4. Technology

1. Time to value and opportunity cost

HCOs have time-sensitive goals that can be put at risk if they choose to build their own healthcare data platform from scratch. In choosing the best path, timing is everything.

What is your expected timeline and how do you expect to derive value from your data platform?

Sample timeline of a phased approach for building a healthcare analytics platform

Consider the following in the time required to building the platform:

  • Choosing and partnering with a vendor: The vetting and selection process can take several months, and multiple vendors means more overhead
  • Organizational transformation: Building a fully staffed team means recruiting software engineers at 45-60 days per hire, plus onboarding and training
  • Construction: Integrating your existing infrastructure with new capabilities will take at least 9 months, not including unexpected maintenance and vendor solution upgrades
  • Training rollout and enablement: Once the platform is built, stakeholders and end users will need a minimum of 2-3 months’ training and certification
Build vs. Buy: Sample timeline of a phased approach for building a healthcare analytics platform

The total 'build' timeline can be well over a year and a half of time and monetary investment. That doesn’t account for the overhead of multiple vendors, which can pull your IT team away from important strategic work. Additionally, the work needs to be complete to the point that you can extract value while not implementing correctly as that can greatly inflate costs.

Sample timeline of a phased approach for buying a healthcare analytics platform

Consider the following in the time required to buying the platform:

  • Platform configuration: User setup, risk algorithms, quality measures, reports and dashboards, provider org anization structure, contract structure
  • High-priority data source integration: Access, installation, review, and go-live of data sources
  • High-priority module implementation: managing potential data bottle necks, data storage, and end-user training
  • Additional scope: Average client implementations can be between 6–12 months as large connector data volumes may increase the timeline
Build vs. Buy: Sample timeline of a phased approach for buying a healthcare analytics platform

The 'buy' option often goes faster than building from scratch, but it still takes time for a ready-made data platform to become fully operational within an organization. It's important to consider the length of initial set-up and configuration, plus anticipated or unanticipated delays like data bottle necks and user trainings.

2. Managing capital and operational costs (CapEx / OpEx)

Healthcare organizations often pay excessive costs for application development, vendors’ services, and legacy maintenance. Managing capital — how much can you allocate to R&D for initial procurement?

  • Managing multiple contracts: Working with multiple vendors can be time-consuming and riddled with unexpected costs
  • Staffing: A software engineer’s base salary alone can start at $150k, and the more complex your software, the higher your staffing costs
  • Storage: Each user on your platform will need ample storage space, and to avoid data volume charges, you’ll need to design and implement strict data governance practices
  • Maintenance: An ongoing expense that will trail your data platform every time you maintain or update it

By contrast, choosing to buy an option already on-market eliminates this potential waste of time and resources. Though 'buy' comes with a different set of considerations, outsourcing concerns like staffing and storage can lessen the burden on an organization.

3. Staffing

HCOs are vying for technical resources in a highly competitive market. Staffing raises a unique set of concerns, including:

  • Hiring: The average timeline for hiring a software engineer is 60-65 days (with a full recruiting team in place).
  • Compensation: Competing with a vast landscape of tech companies means providing high salaries, equity, and benefits.
  • Volume: You’ll need a full team of software developers, connectors, and operations experts with 1-2 people per area. That quickly amounts to millions of dollars per year for a functioning team. Contractors often have limited availability, and can command as much as $175-$210 per hour.

4. Technology

There are lots of technical decisions to make and vendors to evaluate when it comes to stitching together the healthcare data platform. 

  • Which cloud computing platform? Azure vs. AWS vs. on-premises hosted?
  • Which strategy is right for my data? Data lake, data warehouse, NoSQL?
  • What considerations do I need to make for my users? Do we need a semantic layer of data organization for end-users?
  • How do you stay on top of changing data standards and requirements? Master data management. Interoperability and a massive data asset requires a common set of data standards and ongoing management.
  • How will you manage extract, transform, load / extract, load, transform pipelines (ETL/ELT)? Should we rely on third parties to transform the data? What is the organizational cost of doing this ourselves?
  • How do you ensure proper use of the data and data governance? How do you actively manage storage costs?
  • Which power business intelligence tools do you use? How will they work with my data lake strategy?

[Building our own healthcare and data analytics platform] allowed us to be pretty advanced early on in the market from a CIN perspective. But over a period of time, it became increasingly more expensive because we weren’t really scaling. Every time we wanted to add a new practice or add a new feature or template, it required new investments and new staff. Although we had some success, we saw that the costs were not scaling effectively with us. [We started looking at platforms] and luckily we met Arcadia and they brought a strong value proposition to us which has allowed us to grow our network and perform better and do more without spending more.

Anthony Del Rio
Former Executive Director and President Rush Health

In the face of a costly and critical decision, information paves the path forward

Reach out to a member of Arcadia's team to discuss your organization's healthcare data analytics needs.