Will GitLab disclose Duo Agent revenue or adoption metrics showing >5% penetration of Ultimate customers by Q2 FY2027?

The original prediction of 0.44 was made when Duo Agent was pre-GA with maximum uncertainty. Now we have concrete data that materially reduces the probability. Three binding constraints: (1) ~70% of customers are self-managed and need v18.8+ upgrade, with management estimating 6 months for 50% adoption — meaning by the September 2026 resolution date, perhaps 50-60% of the self-managed base can even access Duo Agent. (2) Management explicitly guided 'not significant revenue contribution' in FY2027, which is inconsistent with disclosing a >5% penetration metric that would imply meaningful traction. (3) The resolution requires DISCLOSURE, not just attainment — if management is framing expectations low, they have no incentive to highlight a specific penetration number unless it dramatically exceeds expectations. The SaaS cohort (~30% of Ultimate) can adopt immediately, but even strong SaaS adoption alone may not reach 5% of ALL Ultimate customers. The calibration data showing we were underconfident on GitLab execution (revenue beat Brier 0.34) introduces modest upward pressure, but the structural barriers here are qualitatively different from a revenue beat.

Taking a more pessimistic stance on the conjunction of requirements. The market needs: (a) >5% of Ultimate customers actually adopting Duo Agent, AND (b) management choosing to disclose this metric publicly, all by September 30, 2026. On (a): with 70% self-managed needing version upgrades over ~6 months, only the SaaS cohort (~30% of Ultimate) plus early self-managed upgraders can adopt. If SaaS Ultimate customers show 15% adoption (optimistic for 8 months post-GA), that is roughly 4.5% of total Ultimate — still below the 5% threshold. On (b): management's 'not significant' framing creates an expectation anchor that makes detailed metric disclosure counterproductive unless numbers dramatically exceed expectations. The qualitative anecdotes (airline, semiconductor, insurance) are classic early-adopter stories, not evidence of broad penetration. The Q1 FY2027 earnings call (June 2026) is the last disclosure opportunity before resolution, and will have only ~5 months of post-GA data — insufficient for enterprise-scale penetration.

Adopting a more optimistic frame: the calibration data shows we were underconfident on GitLab execution (revenue beat Brier 0.34). Could we be underestimating Duo Agent adoption velocity? Several positive signals: (1) the airline achieving 90% autonomous component updates suggests genuine product-market fit in specific use cases, (2) the semiconductor competitive win with 5K users is a substantial early deployment, (3) GitLab has strong strategic incentive to disclose AI metrics given the MODEST market expectations — even partial data could serve as a catalyst. However, the structural barriers remain. The self-managed upgrade cycle is a hard constraint that cannot be overcome by enthusiasm alone. Even accounting for potential positive surprise, management's own guidance language creates an expectation ceiling. I weight the upside case at ~33% — acknowledging real adoption momentum among SaaS customers and enterprise early movers, but recognizing the structural barriers and disclosure dynamics work against YES resolution by September 2026.

The earnings update materially shifts this market toward NO. The original 0.44 reflected maximum uncertainty pre-GA. Now three concrete data points compress the probability: self-managed upgrade barrier (structural), management's 'not significant' guidance (intentional), and zero aggregate metrics disclosed on the Q4 call despite 7 weeks of post-launch data (revealed preference). If management had strong early adoption data, the Q4 call was the natural place to highlight it — they chose anecdotes instead. The SaaS segment (~30%) is the only viable adoption channel for Q2 FY2027 resolution, and even there, enterprise procurement and security review cycles mean many SaaS Ultimate customers will not adopt within 8 months. I assign 30% — lower than original but acknowledging a non-trivial path if SaaS adoption proves rapid and management decides to highlight it as a positive catalyst on the Q1 FY2027 call.

Focusing on the disclosure requirement specifically. Management has two remaining opportunities to disclose: the Q1 FY2027 call (~June 2026) and potentially an investor day or product announcement before September 30. The disclosure incentive analysis is nuanced: with MODEST market expectations, GitLab would benefit from disclosing strong AI metrics. But management's explicit 'not significant revenue contribution' guidance for FY2027 sets up a contradiction — disclosing >5% penetration would undermine their own cautious framing and could create an expectations trap they cannot sustain. The more likely scenario is management continues qualitative commentary ('growing customer interest,' 'expanding pilots to production') without crossing the quantitative disclosure threshold this market requires. Even the optimistic case of 5-7% SaaS penetration may not be disclosed as a specific metric if management is managing expectations down.

Taking a moderately bullish perspective within the updated constraint set. Key consideration: the OR condition in the resolution criteria. The market resolves YES on either revenue disclosure OR adoption metrics showing >5%. This broadens the paths. GitLab might not disclose revenue but could disclose adoption-adjacent metrics — 'X% of our Ultimate customers have enabled Duo Agent,' 'Y thousand active Duo Agent users' — that an analyst could interpret as showing >5% penetration. The competitive dynamics also matter: GitHub Copilot's market share narrative creates pressure for GitLab to demonstrate AI traction. If a competitor claims AI-driven customer wins, GitLab has strategic incentive to counter with its own metrics. The airline and semiconductor examples suggest at least a few dozen early customers — if the Ultimate base is ~5,000 customers, 250 would be 5%. Among SaaS-deployed Ultimate customers (perhaps 1,500), even 75 active customers would be 5% of that cohort. These are achievable numbers even in early adoption.

The self-managed upgrade barrier is the decisive new fact. With ~70% of customers on self-managed deployments needing v18.8+ and management estimating 6 months for 50% upgrade completion, the addressable market for Duo Agent by September 2026 is structurally limited to SaaS customers (~30%) plus early self-managed upgraders. Management's 'not significant' guidance makes disclosure of a >5% metric unlikely. The calibration adjustment from the revenue beat (we were underconfident) is insufficient to overcome these structural barriers.

Taking the most pessimistic reasonable position. The conjunction problem from the original analysis has become even more binding: GA + meaningful adoption + quantified disclosure all required, AND now with a confirmed version upgrade barrier on top. Management could have disclosed early metrics on the Q4 call but chose not to — this is the strongest signal that >5% penetration is not on track. Enterprise AI tool procurement cycles run 3-6 months, meaning many customers who started evaluating Duo Agent at GA (Jan 14) will still be in evaluation by September 2026. The 20% probability reflects a tail scenario where SaaS adoption surprises strongly upward.

Weighting the base case against the upside scenario. Base case (~70% weight): Duo Agent adoption grows steadily through FY2027 but does not reach the 5% threshold for the full Ultimate base by September, and management maintains qualitative-only commentary. Upside case (~30% weight): SaaS Ultimate customers adopt rapidly (>15% penetration by mid-2026), GitLab faces competitive pressure to disclose AI metrics, and the Q1 FY2027 call includes a specific penetration or 'customers enabled' metric that meets the resolution criteria. Low confidence because the post-GA data is still very limited and the next earnings call (Q1 FY2027 in June) will be the pivotal data point we do not yet have.