Defensibility Grade: The A–F Score of the Finished Study
The defensibility grade in Quartyl: the A–F letter score (0–100) that rates a finished study on sample size, comparability, range tightness, margin placement and acceptance rate.
Definition
The defensibility grade is Quartyl’s letter grade (A–F) on a 0–100 scale that rates how defensible a finished study is — the score the Predictive Risk Engine attaches to each study, beside the audit-probability and litigation-risk scores, that answers the TPO’s first question: “why should I believe this comparables set?” The grade is computed, not assigned — a weighted read across the study’s own statistics: the sample size (how many comparables survive the screens), the comparability (the coefficient of variation of the PLI distribution), the range tightness (the IQR relative to the median), the tested party margin placement (where the studied margin sits against the interquartile range), and the acceptance rate (the share of screened companies accepted). A study with a tight range, a sized set and a margin inside the IQR grades high; a study built on a small, scattered set with the tested party outside the range grades low — the grade is the study’s own audit-readiness, measured.
The defensibility grade, in one computation:
1. The criteria (per study, from the study's own statistics)
a. The sample size (the accepted comparables — 20% weight)
b. The comparability (the CV of the PLI distribution — 25% weight)
c. The range tightness (the IQR / the median — 20% weight)
d. The margin placement (the tested party's PLI vs the IQR — 20% weight)
e. The acceptance rate (the accepted / the screened — 15% weight)
2. The overall score (the weighted sum, 0–100)
3. The letter grade (the band the score falls in)
→ A (85+, Excellent) · B (70–84, Strong) · C (55–69, Adequate)
→ D (35–54, Weak) · F (below 35, Inadequate)
→ the portfolio view (the average grade across the firm's studies)| The criterion | The weight | The read |
|---|---|---|
| The sample size | 20% | The accepted comparables: 20+ full marks, 12+ strong, 8+ adequate, 4+ weak, below 4 inadequate — the set the range is built on |
| The comparability (CV) | 25% | The coefficient of variation of the PLI values: below 15% tight, below 25% good, below 40% fair, above 55% scattered — the tightest band, the heaviest weight |
| The range tightness (IQR/median) | 20% | The interquartile range relative to the median: below 0.3 well-bounded, 0.3–0.5 tight, above 1.0 wide — the range the authority tests the price against |
| The margin placement | 20% | The tested party’s PLI inside the IQR (full credit), within a 20% buffer outside the quartiles (partial), outside that (low) — the position the adjustment theory turns on |
| The acceptance rate | 15% | The share of screened companies accepted: 70%+ strong, 50%+ good, below 30% weak — the screening discipline visible in the numbers |
The working read (the TP audit defense and the statistical significance rule): the grade is the pre-emptive answer to the examination’s challenges — the small set the TPO will call “insufficient” (the statistical significance threshold), the wide range the TPO will call “inconsistent” (the interquartile range the price is tested against), and the margin outside the range the TPO will convert into the adjustment. The grade does not defend the study — the documentation, the accept-reject matrix and the evidence do — but it measures the study’s exposure in advance, so the team strengthens the weak criterion (the re-benchmark for the set, the screen review for the acceptance rate, the adjustment for the placement) before the examination finds it.
Example
The finished study (the TNMM, OP/OC, the Indian services tested party): the defensibility grade reads the study’s statistics — (1) the sample size (14 accepted comparables — 80 points, the 12+ band); (2) the comparability (CV 18.2% — 85 points, the below-25% band); (3) the range tightness (IQR/median 0.42 — 80 points, the below-0.5 band); (4) the margin placement (the tested party’s 11.8% inside the IQR’s 9.4%–14.1% — 95 points, full credit); (5) the acceptance rate (62% of 31 screened — 75 points, the 50%+ band). The overall score: the weighted sum (80×0.20 + 85×0.25 + 80×0.20 + 95×0.20 + 75×0.15) = 82.8 — the B (Strong) band. The reading: the study is defensible as built (the sized set, the tight range, the in-range margin) with one visible soft spot (the acceptance rate — the 38% screened out needs the documented accept-reject matrix and the per-company evidence behind it). The portfolio view (the firm’s studies, the average defensibility score) shows the firm-level pattern — the studies grading D or F are the re-benchmark queue, the ones to strengthen before the assessment year, not after the TPO’s notice.
See also
- TP Audit Defense (what the grade pre-empts)
- Statistical Significance (the sample-size criterion)
- Interquartile Range (the range the placement is tested against)
FAQ
Is the grade a substitute for the TP documentation? No — the grade is the measure (the study’s audit-readiness, computed from the study’s own statistics) and the prioritisation (the portfolio’s weak studies, the re-benchmark queue). The documentation (the Rule 10D local file, the master file, the CbCR, the contemporaneous record — the TP audit defense file) is the defence; the grade is the pre-flight check on the file’s numbers. A B-grade study with a thin file is still a thin file — the grade measures the statistics (the set, the range, the placement), not the paper.
Can the grade change after the study is signed off? The grade is computed from the study’s persisted analysis data (the accepted set, the statistics, the margin, the screening counts) — so it is stable for a finished study, and it moves when the underlying work moves: a re-benchmark (the new set, the new range), an override batch (the new acceptance rate) or a re-run with new data recomputes the score and the letter. The portfolio view (the average across the firm’s studies) updates with each — the firm’s defensibility, tracked study by study.
What does a low grade mean for the audit? A low grade (D or F) means the study carries the patterns the TPO challenges — the small set (insufficient, per the statistical significance rule), the wide or scattered range (inconsistent comparables), or the margin outside the interquartile range (the adjustment’s starting point). It is not a finding and not a prediction of adjustment — the audit-probability score carries the risk side — but it is the internal signal that the study should be strengthened (the re-benchmark, the screen review, the documented adjustments) before it is relied on, not after it is examined.
Run the screens as a study, not a spreadsheet
Quartyl applies the method, PLI and screening steps above as a pipeline — and keeps a documented reason for every exclusion.
Related docs
TP Audit Defense: TPO Proceedings, Scrutiny and Appeals
The Indian TP examination end to end: how the TPO proceeding runs, the adjustment theories, the AO interplay, the appeal route, and the documentation readiness that decides the outcome.
Read docStatistical Significance: Is the Comparables Set Big Enough?
Statistical significance in benchmarking: the set-size adequacy question — the minimum set, the TPO’s first ask, and the documentation that answers it.
Read doc