E&P parallels to the SaaSpocalypse

I spent a decade in equities covering North American E&P; I've worked in Canada (Calgary, Toronto) and NYC, and I've been thinking a lot about the parallels between these two sectors, going from loved to hated. Both stories follow the same psychological cycle: a supply/demand mismatch misread as structural, euphoric M&A and multiple expansion, a technological shock that inverts the core thesis, and then an identity crisis about whether the business model is even viable.

Phase 1: Peak Oil

In Canadian E&P's golden era, the dominant narrative was supply scarcity. Conventional oil was running out, Canada sat on one of the world's largest reserves, and $200 oil calls were serious. M&A was rampant because scale mattered enormously in a world where reserves were the scarce asset. SaaS had its analogue: from roughly 2019–2021, the narrative was software inevitability - every company was becoming a software company, recurring revenue was the most durable business model ever invented, and the per-seat model would compound indefinitely. Multiples peaked at 18x forward revenue on EMCLOUD in 2021. Both periods shared the defining feature of a consensus so strong it felt almost geological.

Phase 2: Technology Shock

For E&P, the shock was U.S. shale. Fracking technology - not demand destruction - turned a 2–3% global undersupply into a 2%+ oversupply almost overnight. Critically, Canadian oil sands producers were already not particularly profitable despite the boom: high upfront capex, $75/bbl+ breakevens, and a heavy oil discount meant the growth was real but margins were thin. The shale shock didn't just hurt prices - it exposed that the profit model was fragile beneath the surface.

SaaS is experiencing a similar disruption. AI is the shale - a technology that rapidly expanded the supply of software capability, collapsing the scarcity premium that justified those multiples. The per-seat model, analogous to the "reserve size = value" logic in E&P, breaks when AI agents reduce the number of users required. And just like oil sands, many SaaS companies that looked great on top-line growth were never that profitable - growth obscured FCF weakness, and now that growth itself is decelerating (median revenue growth fell to 12.2% by Q4 2025, from 25–30% pre-COVID).

Phase 3: The Oversupply + Multiple Collapse

The speed is the one real difference. The shale oversupply took ~3–4 years to fully reprice E&P. The SaaSpocalypse has fast: $1 trillion in market cap erased in a single week in February 2026, but it may not be over just yet. Public markets price AI disruption in real time, while physical oil production ramp-up was slower.

Phase 4: The Pipeline Problem

This is the most underappreciated parallel. Canadian E&P suffered a prolonged, existential question about infrastructure - Trans Mountain, Energy East, Keystone XL - that was independent of the commodity price question. Even if oil recovered, could you get it to market? The discount to WTI wasn't just about oil quality; it was a structural infrastructure tax.

SaaS has a quieter version of this: distribution and go-to-market are the pipeline. Enterprise sales cycles, implementation complexity, and switching costs were always the moat, but AI is now commoditizing the "last mile" of software delivery. Point-solution horizontal SaaS vendors (the equivalent of landlocked oil) face the harshest re-rating (eg. MNDY, SVMK), while vertical/domain-specific SaaS with proprietary data (the producers who had pipeline access) are proving more resilient (FDS, .

Phase 5: Peak Demand

The cruelest part of the Canadian E&P story was the 180-degree narrative flip: from "we'll run out of oil" to "we'll stop wanting oil" - peak consumption fears layered on top of the oversupply. It wasn't just that there was more oil; the endgame of the whole industry was being questioned.

SaaS is living this now. The debate is no longer just "AI will take market share" - it's "does the SaaS model itself survive?" . Here I see the same logical incoherence that existed in E&P - peak demand forecasts were often incompatible with the actual consumption trajectory.

Valuation Implications

I see a lot of posts about how much valuation multiples have compressed, with hopes of them returning to "historical norms". I believe this is the wrong thing to focus on. This day may never come.

There is a point of euphoria in many industries where proxy metrics substitute for NAV forecasts. This is largely because the industry is too immature to really build a NAV. The FCF isn't there, the companies aren't that profitable, so we default to things like debt-adjusted production per share, reserves growth, or in the case of SaaS : Annual Recurring Revenue, Net Retention Revenue. These become irrelevant quickly when the market starts valuing the business as a stream of free cash flows - the fate of any mature stock.

Peak pessimism hit E&P around 2020, by that point: balance sheets were fixed, debt was being retired, capex was sustainable, opex had been crushed to its lowest point, companies were generating walls of free cash flow, and still trading at an astounding low 2-4x DACF.

Today, SaaS has fallen from the sky, but under a FCF is king framework, there is really only one company that screams undervalued: ADBE (~10% FCF yield). Two other names I'm watching: PEGA (~7%), VEEV (~5%, healthcare moat, 40%+ FCF margin).