Will Fastly's management disclose that price erosion has reverted to double-digit levels by Q2 2026 earnings?

The committee identifies price erosion causation (structural vs cyclical) as the single highest-impact unknown. However, several factors weigh against a rapid reversion to double digits within just 2 quarters. First, 50% of customers are on 2+ products and 75% of revenue comes from multi-product relationships -- this creates genuine switching costs that provide structural pricing support even if AI demand normalizes. Second, the resolution requires management to explicitly disclose double-digit erosion, and this metric was only disclosed for the first time in Q4 2025. Management may be reluctant to disclose deterioration of a metric they just introduced. Third, even if the improvement was partially cyclical, the transition from mid-teens to mid-single digits likely reflects both structural and cyclical components -- a partial reversion to high-single digits is more probable than a full reversion to 10%+.

The meta-synthesis ranked 'price erosion improvement: structural vs cyclical' as the highest-impact unknown. The cyclical case rests on AI traffic demand tightness creating temporary supply-demand imbalance. Several factors support the cyclical thesis: CDN pricing has historically trended downward over time, the 25-75% memory cost inflation creates pressure on Fastly's cost structure, and the CapEx step-up to 10-12% of revenue suggests Fastly is investing to meet demand that may normalize. However, the resolution window is tight -- only Q1 2026 and Q2 2026 earnings calls. Even if cyclical forces are reasserting, the lag between market pricing shifts and earnings disclosure is typically 1-2 quarters. A reversion from mid-single digits to double digits in such a short window would require a dramatic market shift. More likely, any deterioration would show as movement from mid-single to high-single digits first.

I note an important asymmetry in the resolution criteria: resolves NO if 'no updated disclosure is made (status quo assumed mid-single digits).' This means the YES case requires both (a) price erosion actually reverting AND (b) management choosing to disclose it. The Q4 2025 disclosure was framed positively as an improvement. If conditions deteriorate, management has strong incentive to avoid quantifying the deterioration, especially given this was their first-ever voluntary disclosure. They could simply stop disclosing the metric. Additionally, the Atomic Auditor noted volume leverage (spreading fixed costs over more traffic) as a parallel explanation for margin improvement. Even if price erosion worsens, gross margins could remain stable due to mid-20s% traffic growth, making it less likely management would highlight pricing pressure. The committee noted Network Services revenue grew 19% YoY with traffic growing mid-20s% -- the gap already implies some price erosion, but the volume growth may mask deterioration in per-unit pricing.

The base case is NO. Price erosion just improved from mid-teens to mid-single digits as of Q4 2025. For it to revert all the way back to double digits within 1-2 quarters requires a dramatic deterioration. The committee identified multi-product switching costs as a structural support. Even the bear case -- where AI demand tightness was the primary driver -- would likely produce a gradual reversion over 3-4 quarters, not a sudden snap-back. The CapEx step-up to 10-12% industry-wide could add supply, but new capacity takes time to come online. The most probable outcome is that price erosion stays in mid-to-high single digits through Q2 2026.

Low confidence because the committee itself flagged this as an E1-E2 evidence level question -- meaning the structural vs cyclical determination lacks sufficient data. CDN is a commodity market historically characterized by persistent price erosion. The improvement to mid-single digits was unprecedented and only lasted one disclosed quarter. Memory costs up 25-75% create real pressure -- if Fastly absorbs these costs rather than passing them through, effective real price erosion could already be higher than nominal. However, the disclosure risk cuts both ways: management explicitly introduced this metric as a positive, and analysts will now track it every quarter, creating pressure to address it even if news is bad. I assign higher probability than my base case would suggest because of this measurement uncertainty.

The question asks specifically about management disclosure of double-digit price erosion. Three scenarios: (1) Price erosion stays mid-single digits -- NO (most likely, ~50% probability). (2) Price erosion worsens to high-single digits -- management likely characterizes as 'modestly higher' without quantifying double digits -- NO (~25% probability). (3) Price erosion reverts to double digits AND management discloses -- YES (~25% probability). But scenario 3 requires both conditions: actual reversion AND disclosure. Even if CDN market dynamics shift, the timeline is compressed. Network Services growing 19% with traffic mid-20s% suggests current erosion is already ~5-7%, consistent with mid-single digits disclosure. A jump from ~6% to 10%+ in two quarters is a significant move.

Price erosion improved from mid-teens to mid-single digits per Q4 2025 disclosure. Multi-product customers (50% of base, 75% of revenue) create switching costs. Resolution window is only 2 quarters. Reversion to double digits within this timeframe is unlikely absent a major competitive disruption. Base case is erosion stays in single digits.

CDN pricing historically trends downward. This was first-ever disclosure with no trend data. Memory costs up 25-75% create cost pressure. AI demand tightness that may have supported pricing could normalize. But double-digit reversion in 2 quarters is a large move from mid-single digits. Low confidence due to limited data history on this metric.

Mid-single digits as of Q4 2025. Resolution requires explicit management disclosure of 10%+ erosion by Q2 2026 earnings. Even if conditions worsen, management may avoid quantifying deterioration. Traffic growth at mid-20s% provides volume cushion that could offset some pricing pressure. Probability below 30% reflects compressed timeline and disclosure hurdle.