How often should benchmarks be refreshed?

Common practice is a full search every three years with annual financial updates, provided the business has not materially changed.

Is the interquartile range required?

It is a widely used statistical tool for improving reliability where comparability defects remain; it is not mandated in every jurisdiction.

Comparability analysis — Definition, Example & Sources

Key facts about Comparability analysis
Term	Comparability analysis
Also called	Benchmarking study; Comparables search
Primary authority	OECD Transfer Pricing Guidelines, Chapter III (OECD, 2022)
Source status	Primary source
Topics	Core principles; Pricing methods
Most relevant to	In-house tax teams; Advisors & consultants; Students & job seekers
Most common audit trigger	Benchmark last refreshed more than three years ago.
Who owns it internally	Usually an external advisor or an internal benchmarking team of specialists.
Last reviewed	2026-06-30

Plain English

You cannot say a price is right without something to compare it to. A comparability analysis finds that something — either a deal your own group did with a third party, or the published results of independent companies doing similar work — and makes adjustments for the differences that matter.

Technical definition

A comparison of a controlled transaction with an uncontrolled transaction, where none of the differences between the situations being compared could materially affect the condition being examined, or where reasonably accurate adjustments can be made to eliminate the effect of such differences.

Why it matters

It converts a principle into a number. It is also the single most contested deliverable in an audit: search strategy, rejection reasons, and comparability adjustments are all attackable.

How it works in practice

01Determine the years covered and review the taxpayer's circumstances.
02Understand the controlled transaction and select the tested party.
03Look for internal comparables before external ones.
04Run a database search with a documented strategy and quantitative screens.
05Apply qualitative review and record why each company was accepted or rejected.
06Make comparability adjustments — working capital being the most common.
07Interpret the resulting range and set or test the price.

Worked example

Interquartile range on a routine service provider

A shared service centre is tested on full cost mark-up. After screening, eleven independent comparables produce a three-year weighted mark-up range with a lower quartile of 3.8%, median 5.6% and upper quartile 7.2%. The tested party's 5.1% sits inside the range and needs no adjustment. Had it been 2.0%, most authorities would adjust to the median.

Common mistakes

Accepting database output without qualitative review.
Rolling forward a study for years without checking the business has not changed.
Using a pan-regional set where reliable local comparables exist.
Failing to keep the rejection log that supports the search.

Audit red flags

Benchmark last refreshed more than three years ago.
Comparable set including entities with related-party revenue above the independence threshold.

Documentation & data

Documents to hold

Search strategy with database, version, date and screens.
Accept/reject log with reasons per company.
Financial data and any adjustment computations.

Data you need

Tested party segmented financials.
Commercial database access.
Working capital balances for adjustment calculations.

Who owns this internally: Usually an external advisor or an internal benchmarking team of specialists.

Jurisdiction notes

European Union: Local comparables are frequently required or preferred; some authorities reject pan-European sets for local distributors.
United States: The regulations emphasise the arm's length range derived from uncontrolled comparables and permit statistical narrowing where data quality is limited.

Notes by role

Advisors & consultants

The rejection log is the deliverable that survives audit. Write it for a reader five years from now.

Students & job seekers

Learn one database properly. Being able to describe a full search strategy is a differentiator at graduate level.

Frequently asked

How often should benchmarks be refreshed?: Common practice is a full search every three years with annual financial updates, provided the business has not materially changed.
Is the interquartile range required?: It is a widely used statistical tool for improving reliability where comparability defects remain; it is not mandated in every jurisdiction.

Related terms

Arm's length principle: The rule that transactions between related companies should be priced as if the parties were independent.
Most appropriate method: The requirement to choose the transfer pricing method best suited to the facts, rather than following a fixed hierarchy.
Transactional net margin method (TNMM): A method that tests the net profit margin a party earns from a controlled transaction against margins earned by independent companies.
Arm's length range: A range of figures produced by applying a method to multiple comparable data points, any of which may be arm's length.

Sources & status

Primary source
OECD Transfer Pricing Guidelines, Chapter III
OECD, 2022
Primary source
Nine-step comparability process
OECD, 2022 — Described as a good practice, not a compulsory sequence.

Reference material only, not advice on a specific fact pattern. Reviewed 2026-06-30.

Careers

How this shows up in the job

Most graduate transfer pricing work starts here. Screening, rejection notes, and financial normalisation are the daily craft.

Careers in transfer pricing

Key facts