Archetype Router
Different businesses get different math.
Before AXIOM discounts a single cash flow, it decides what kind of company it is looking at. Nine archetypes, each with its own formula, decay schedule, and terminal growth. A hyper-growth name gets revenue multiples; a bank gets P/B; a cyclical gets normalized through the cycle. The interesting part is not the nine buckets. It is the calls at the boundary between two of them.
The nine
One router, nine frameworks.
Pattern in, framework out. Terminal growth shown per archetype, the single assumption that moves a DCF most.
Hyper-Growth
g 4.0%Growth > 30%, margins still negative.
Revenue multiples, not DCF. Earnings are too early to discount.
Growth
g 3.5%Growth 15-30%, margins > 5%.
Standard DCF with a growth premium and slow decay.
Stable Growth
g 3.0%Growth 8-15%, strong margins > 15%.
Traditional DCF, the textbook compounder case.
Mature
g 2.0%Growth < 8%, profitable > 10% margin.
FCF focus, lower growth assumptions, faster decay.
Cyclical
g 2.0%Cyclical sector with volatile through-cycle margins.
Normalize to through-cycle metrics, never on the peak.
High CapEx
g 3.0%CapEx > 25% of revenue.
Normalize to steady-state maintenance CapEx (~60% of current).
Distressed
g 2.0%Negative / very low margins, declining revenue.
Normalize to industry averages; discount current earnings entirely.
Turnaround
g 2.0%Depressed metrics, but a credible recovery path.
Normalized margins plus an explicit turnaround discount.
Financial
g P/B · ROEFinancial-services business model.
P/B and ROE framework. A bank has no free cash flow to DCF.
Boundary calls
Where the classification actually matters.
A clean case routes itself. The names worth arguing about sit between two archetypes, and the pick changes the answer by more than the discount rate ever will. Three of them:
GFL
CYCLICAL, not defensive-stable.
LOOKS LIKEa defensive waste hauler: contracted municipal volumes, 6%+ pricing power, recession-resistant tonnage.
GFL trades on defensive volumes but carries a cyclical balance sheet: ~4.0x net leverage, an M&A-driven growth engine, capex-intensive operations. The router credits the pricing durability but discounts for the leverage and execution risk. It lands fair value at $41.44, right on top of the ~$40 price, which is the honest read: fundamentals are fairly valued, the Street’s extra upside is an event bet.
source · live AXIOM run, gfl_facts.md
A bank
FINANCIAL, not Mature.
LOOKS LIKEa Mature compounder: ~7% growth, a fat ~29% margin, steady dividends. Every screen wants to DCF it.
You cannot DCF a bank. Deposits are not debt, loans are not capex, and “free cash flow” has no clean meaning when the balance sheet is the business. Routing a bank to a DCF produces a confident, wrong number. FINANCIAL swaps in a P/B and ROE framework instead, the multiple that actually governs how banks are valued.
source · methodology doc, Financial archetype
Intel
DISTRESSED / TURNAROUND, not Mature.
LOOKS LIKEa broken Mature business: 2.3% margin, -5% growth. Value it on what it earns today and it screens as nearly worthless.
The margin is depressed by a fab buildout, not by permanent decline. Valuing on trough earnings understates the business; the router normalizes EBITDA toward the ~25% industry level and CapEx toward ~12% maintenance, then applies a turnaround discount. The whole call hinges on one question, whether the depression is temporary or structural, and the archetype is where you answer it out loud.
source · methodology doc, Intel example
See a boundary call run end to end.
GFL is the worked example: the full path from archetype decision to a bear / base / bull fair value.