Skip to main content
Methodology

How we get to the number.

Ten methods, a routing rule that picks between them, market data from brokers reporting closed deals, and a discount rate built from its components. Below is all of it — including what we deliberately do not do.

One method is chosen, not averaged

A valuation that blends every method into an average hides the one decision that matters. AltosIQ selects a single primary method for each business and states why, then cross-checks the result against a second approach and shows both figures side by side.

The choice is driven by the shape of the business, not by preference: whether it is distressed, whether its assets outweigh its earnings, whether it is growing fast enough that revenue is the better measure, and how large it is. The method chosen and the reason for it are computed together in one pass and saved with the valuation — so a report cannot later explain a method it did not use.

The methods

Market approach

What businesses like this one actually sold for.

SDE Multiple
Seller’s Discretionary Earnings times a market multiple. The Main Street standard, and the most common outcome for owner-operated businesses.
Adjusted EBITDA Multiple
Normalised EBITDA times a market multiple. Takes over as businesses grow past the point where the owner’s own labour dominates earnings.
Revenue Multiple
Used where earnings are not yet the right measure — high-growth software being the usual case.
Hybrid Weighted
A blend of the SDE and EBITDA results, for the band where neither alone is right.

Income approach

What the future cash is worth today.

Discounted Cash Flow
Projected cash flows discounted at a rate built up from its components rather than assumed. The build-up is shown below and printed in every report.
First Chicago
Three scenarios — downside, base, upside — probability-weighted. Used where the range of outcomes matters more than any single forecast.

Asset approach

What the balance sheet supports.

Asset-Based
Adjusted net asset value. Routed to when the assets are worth materially more than the earnings they produce.
Liquidation Value
Orderly liquidation, net of recovery discounts. Reserved for genuine distress, and never applied under a going-concern premise — a stated premise that contradicts the method disables the valuation rather than printing a contradiction.

Early-stage methods

For businesses without meaningful earnings history.

Berkus
Criterion-based ceiling for pre-revenue companies.
Scorecard
Comparison against regional early-stage norms.

Where the multiple comes from

The multiple applied to earnings is drawn from current market transaction data — a survey in which business brokers report the deals they actually closed. Not asking prices. The distinction matters: most of what is freely available online is what a seller wanted, and the gap between that and what a buyer paid is large and systematic.

Where enough transactions exist in a business's own sector, the multiple comes from that sector: 137 sectors, drawn from 9,574 reported closed sales. A sector is only used when it carries at least ten reported sales — below that the figure is one or two transactions, which is an anecdote rather than a benchmark, and we fall back rather than quote it.

Where a sector has no usable sample, the multiple is applied by deal size instead, because at this end of the market size predicts the multiple more strongly than sector does. A business with $400,000 of earnings and one with $1.5m do not trade at the same multiple even in the same industry.

Every report names the source and the period it reflects, so the figure can be checked rather than taken on trust.

The discount rate is built, not assumed

Where an income approach is used, the discount rate is assembled from five components and each is printed with its own source. Two are real external data. Three are derived from the business, and are labelled as derived — never dressed up as published research.

Risk-free rate
Live
The 10-Year US Treasury Constant Maturity yield, fetched from the Federal Reserve’s FRED service at valuation time.
Equity risk premium
Published
Aswath Damodaran’s implied US equity risk premium (NYU Stern), the standard citable reference. Republished annually.
Size premium
Derived
Scaled from the same deal-size bands used for the market multiple, so size is treated consistently across the valuation. Labelled as derived, not as a published study.
Industry risk
Derived
A coarse tiering by sector. Labelled as derived.
Company-specific risk
Derived
Built from the risk factors recorded for this business.

Then the business is adjusted for its own risk

A market multiple describes a category. It does not describe this business. Two companies with identical earnings are not worth the same amount, so the concluded value is adjusted for factors specific to the company. Each adjustment applied is itemised in the report with its direction and size.

Customer concentration
Discount
Revenue leaning on too few customers is the single most common reason a buyer re-trades a price.
Owner dependence
Discount
Earnings that walk out with the seller are worth less than earnings that stay.
Intellectual property
Premium
Defensible IP raises the multiple where it genuinely exists.
SBA lending readiness
Premium
A business that clears SBA structural requirements has a wider buyer pool.

What we do not do

A methodology is defined as much by its limits as by its methods. These are ours, stated plainly, because you will find them out eventually and it is better that you hear them here.

We do not present individual comparable transactions

Sector multiples are aggregates — the average across every reported sale in that sector, not a median, and the report says which. We do not claim to have matched your business against named transactions, because we have not. The number of sales behind each figure is stated so you can judge how much weight it carries.

We do not audit the financials

The valuation is built on figures supplied by the business and its adviser. Reports classify the evidence those figures rest on, but classification is not verification and an audit is a different engagement entirely.

We do not use asking prices

Listing data is abundant and free, and it measures what sellers hoped for. It has no place in a valuation a lender will read.

We do not flatter the business

Declining revenue is described as declining revenue. Where a figure is not known, the report records that it is not known rather than filling the gap with something plausible.

Everything is recorded, not recomputed

The method, the reason it was chosen, the discount-rate build-up, the risk adjustments and the triangulation are all saved with the valuation at the moment it is issued. Opening a report a year later shows the same figures it showed on the day it was signed — an issued opinion cannot move under the client who relied on it.