How the index is constructed

43 questions across 8 categories, each answered on a 6-point anchored scale. Every option carries its own written anchor, so two people answering from different positions are answering the same question. There are no free-text fields and no weighting applied by the respondent.

The instrument is identical for everyone. Leads and engagement clients answer the same 43 questions; what separates the populations is the channel, never the question set.

A category score is points earned divided by points available on the questions answered. The overall score is the unweighted mean of the 8 category scores.

That last point is the most misreadable thing about the index and is worth stating plainly: categories count equally regardless of how many questions each contains. Founder and owner independence holds three questions and organizational structure holds eight; they carry the same weight in the composite.

The index is the midpoint of the sell-side and buy-side means — not an average of all responses. Unequal sample sizes on the two sides therefore do not weight the reading.

Sides are assigned by economic position — not by transaction type, and not by who operates the business.

The buy side is the acquiring party: a third-party acquirer, a partner buying out a co-owner, or a spouse buying out a departing spouse’s interest. The sell side is the disposing party, including a non-operating spouse whose interest is being purchased. An owner-operator who is buying out a departing partner is on the buy side, because they are acquiring.

Assessments by parties holding no directional interest are recorded separately as a neutral control and excluded from the index. The current release excludes 7 on that basis.

Engagement assessments are analyst-facilitated. Public web assessments are self-directed. These are different populations and are never pooled. The published index is engagement-only until a documented decision widens it, and each release names the channels it includes.

The index is published quarterly. Each release is built into an immutable dated file and is never edited after publication. A correction is issued as a new release with the cause disclosed, and the superseded file is retained.

Side-adjustment multipliers are computed at release time and carried in the release file, so a figure stated against a release does not move between publications.

Every score is stamped with both an instrument version and a scoring version. The current release is full-v1 scored under svq-scoring-1.0.0.

Questions will change as the market evolves. When they do, the instrument version is bumped, the revision is published, and the series breaks at that point rather than continuing as though nothing changed — the same practice a price index follows when it reweights. Because a category scores as points earned over points available, revisions do not break the arithmetic. They do mean a score can move without the business moving, so the discontinuity is stated.

Raw responses are retained so historical assessments can be re-scored under a current algorithm before any trend is displayed.

The current release carries 35 assessments — 23 sell side and 12 buy side. That supports a directional reading of the perception gap and the ordering of the categories. It does not support subset claims.

The smaller buy-side sample affects precision rather than balance: the buy-side mean carries more uncertainty than the sell-side mean. The sample’s size and sector composition does not mirror the general business population, and composition is disclosed with each release.

The SVQ Index reports assessments made by participants, not properties of the businesses themselves. It is not a valuation, an appraisal, a fairness opinion, or an opinion of value, and it does not determine a standard of value in any proceeding.

Release

2026-baseline · cut 2026-08-31 · engagement channel only

Period

March 31, 2026 – August 27, 2026

Sample

35 assessments · 23 sell · 12 buy · 7 neutral excluded

Reading

67.3% · sell 72.4% · buy 62.2% · gap +10.2 points