DECIDE · STEP 2 OF 4
Build vs buy
Set your spend, channel count, and whether a data scientist already exists on payroll. The board below prices both paths with sourced numbers only: BLS salary data times a stated loading factor on the build side, published vendor prices on the buy side, and a verdict sentence written to be pasted into a budget slide.
By Oliver Wakefield-Smith · Updated 2 August 2026 · Every number on this page resolves to a primary source in the ledger below.
Build in-house
- Median DS salary1
- $120,230
- Loading factor2
- x1.46
- FTE fraction
- 1.0
- Open-source license
- $0
- Annual total
- $175,852
Buy managed
- Published floor3
- $35,880/yr
- Published ceiling4
- $220,000+/yr
- Quote-only vendors
- 9 of 12
- Published range
- $35,880 to $220,000+
At $10M paid spend, in-house runs about $176k/yr fully loaded versus a published managed range of $36k to $220k/yr. Get two managed quotes before committing either way; most vendors do not publish pricing.
The 1.0 / 0.5 FTE fraction is a stated assumption, not a measured figure. Salary and loading factor are sourced in the ledger below; the managed range uses published prices only, and 9 of the 12 managed vendors in the registry publish none.
The arithmetic, shown
In-house: median US data-scientist salary $120,2301 (BLS OEWS, May 2025, SOC 15-2051) times 1.462 loading (BLS ECEC, March 2026: total compensation $49.32/hr against wages $33.72/hr) = $175,852 fully loaded, times an FTE fraction of 1.0 for a first build without an existing data scientist, or 0.5 where MMM joins an existing scientist's load. The FTE fraction is our stated assumption; the salary and loading numbers are not.
Managed: the floor is Sellforte's published Growth tier, $2,990/mo3 = $35,880/yr; the ceiling is Paramark's published Enterprise from-price, $220,000/yr4, both verified 2 August 2026. Nine of twelve managed vendors publish nothing, so your quote may land outside this bracket; the bracket is simply everything any vendor was willing to print.
What the calculator does not price
Time to first defensible number (typically a quarter in-house, less with a vendor whose pipeline your data fits), key-person risk on the build side, black-box risk on the buy side, and the calibration experiments both sides need. Those trade-offs are qualitative and covered in open-source vs managed.
Next: step 3, the full cost anatomy.
Source ledger
Every numeric claim on this page resolves to one of these primary sources.
- [1]BLS OEWS, Data Scientists (SOC 15-2051), May 2025 national estimateshttps://www.bls.gov/oes/current/oes152051.htmretrieved 2026-08-02
- [2]BLS Employer Costs for Employee Compensation, March 2026https://www.bls.gov/news.release/ecec.nr0.htmretrieved 2026-08-02
- [3]Sellforte pricing pagehttps://sellforte.com/pricingretrieved 2026-08-02
- [4]Paramark pricing pagehttps://paramark.com/pricingretrieved 2026-08-02