When organic does NOT compound — six failure modes: no search demand, thin content, weak product/PMF, algorithm/AI-Overview shift, rebuild reset, entrenched incumbents

Claim: The compounding case for organic is conditional. Six identifiable failure modes where the J-curve never turns up — the trough is the whole story:

1. Zero search demand for the topic — no one searches for what the business does.
2. Thin / undifferentiated content that never earns authority or links.
3. Fundamentally weak site/product — no product-market fit, no conversion regardless of traffic.
4. Algorithm update or AI-Overview shift that erases traffic at stable rankings (Broad core updates are NOT penalties — Sullivan: "this doesn't mean all sites will go back up to wherever they were if they are down from a previous peak"; recovery requires substantive improvement + waiting for the next update; 2026 cadence: March 27–April 8 + May 21–June 2, AI-era verification verdict — older click-based ROI/payback studies should be treated as UPPER BOUNDS and partially obsolete as of 2026).
5. Abandonment or rebuild that resets authority (A rebuild that breaks URL structure can forfeit "the most valuable marketing channel" and take "months or years to recover" — discards accumulated link equity, resets authority).
6. Highly competitive heads where established incumbents hold the SERP (Ahrefs (May 2025) — for HIGH-VOLUME keywords, only 0.3% of newly published pages rank in the top 10 within a year, Ahrefs — 72.9% of top-10 pages are more than 3 years old).

Any one of these can turn the J-curve into a flat line; combinations can turn it into a permanent loss.

Source: Compass_artifact research document, June 2026.

Confidence: Industry-consensus.

Caveat: Diagnosing which failure mode is in play requires honest investigation, not just waiting longer. The diagnostic threshold is Rule: if there is no impression / keyword-footprint movement by ~6 months despite clean technical SEO and genuine content, re-evaluate the site itself — do NOT just wait longer — if there's no impression movement by month 6 despite clean technicals, the site is likely in failure-mode 1, 2, 3, or 6 and waiting is the sunk-cost trap (Sunk-cost honesty — the literature warns that "overemphasis on avoiding the sunk-cost fallacy can lead to premature abandonment of worthwhile projects"; but patience is rational ONLY when the underlying site is genuinely good).