Estimate what you spend each year maintaining point-to-point integrations and reconciling data by hand, and compare it with a FHIR layer architecture. Every assumption is yours and editable.
Parameters are entered in USD
What it costs to not be able to USE the data: ETL pipelines, preparing information before it can be analysed, and the tax every piece of development pays. Softer assumptions than the ones above, which is why they sit apart.
62,100 USD – 103,000 USD
over 3 years · 20,700 USD – 34,200 USD per year
28,800 USD – 57,600 USD
over 3 years · 6,270 USD – 12,500 USD per year
4,460 USD – 73,900 USD
30 months
Computed on the conservative scenario: most expensive implementation, smallest saving.
6
possible point-to-point pairs
4
connections in a hub topology
21 possible pairs at the end of the horizon
11,100 USD – 18,600 USD
per year
Total difference (direct + second-order)
15,400 USD – 106,000 USD
The figure above is the one that holds up in a negotiation; this is the ceiling if the data is actually put to use.
See the ETL disappear: SQL on FHIR in Ema Dash →This is a parametric estimate, not a quote. Its most fragile assumption is the reduction in manual work: if your team does not reassign those hours, the saving never materialises. Ranges are conservative–optimistic, and no single figure from this model should be treated as a commitment.
Optional. You already have the full result; the email is just so you keep it in writing.