Plain English
There are five recognised methods. None is automatically better. You pick the one that fits what the parties actually do and where you have the most reliable data — then explain in writing why you rejected the others.
Technical definition
Selection of the method that, having regard to the respective strengths and weaknesses of the OECD methods, the appropriateness of the method in view of the nature of the controlled transaction, the availability of reliable information, and the degree of comparability, provides the most reliable measure of an arm's length result.
Why it matters
Method selection determines the answer far more than fine-tuning the comparables. It is also the first thing an examiner challenges.
How it works in practice
- 01Complete the functional analysis and delineation first.
- 02Identify which party performs routine functions — usually the tested party.
- 03Assess data availability for each candidate method.
- 04Prefer CUP where a reliable internal comparable exists.
- 05Prefer a profit split where both parties make unique and valuable contributions.
- 06Document the reasoning, including rejected methods.
Worked example
Why not CUP?
A group sells a patented compound both to a related distributor and to an unrelated one. On paper the internal CUP looks perfect. But the third-party deal covers a different territory, half the volume, and no regulatory support obligation. Rather than force unreliable adjustments, the analysis moves to TNMM on the distributor, and the CUP is documented as considered and rejected with reasons.
Common mistakes
- Defaulting to TNMM because the data is easy.
- Selecting a method before completing the functional analysis.
- Not documenting rejected methods at all.
Audit red flags
- The same method applied to every transaction type in the group.
- Profit split used with no evidence of unique contributions on both sides.
Documentation & data
Documents to hold
- Method selection memo with reasons for acceptance and rejection.
- Functional analysis supporting the tested party choice.
Data you need
- Internal third-party transaction data.
- Reliable segmented financials for the tested party.
Who owns this internally: Group tax, advised by the benchmarking team.
Jurisdiction notes
- United States
- Framed as the 'best method rule' under the Section 482 regulations, with comparable profits method playing the TNMM role.
- OECD
- Traditional transaction methods are preferred where they can be applied equally reliably.
Notes by role
Advisors & consultants
The rejected-methods paragraph is disproportionately valuable in controversy. Never delete it to save space.
Frequently asked
- Can two methods be used together?
- Yes, one can corroborate another. The Guidelines do not require multiple methods but accept corroborative use.
Sources & status
- Primary source
OECD Transfer Pricing Guidelines, Chapter II, Part I
OECD, 2022
- Primary source
Treas. Reg. §1.482-1(c) best method rule
US Treasury, 1994
Reference material only, not advice on a specific fact pattern. Reviewed 2026-06-30.
Careers
How this shows up in the job
Expect a case question: 'you have an internal CUP and a good TNMM set — which do you use and why?'
Careers in transfer pricing