The same AI wave is splitting software companies into winners and losers: Okta's beat-and-raise shows businesses paying up for security as AI amplifies cyber threats, while Adobe's end of an 18-year era signals AI is eroding a legacy franchise — and its s

Verdict: the divergence is real, but the pair's own rules say wait

The idea's core divergence shows up in the numbers: Okta's latest quarter (ended 2026-07-31) delivered revenue of $805M up 5.2% sequentially with operating margin jumping to 13.3% from 7.3%, while Adobe's quarter ended 2026-05-29 showed just 3.4% revenue growth with operating margin down 10.5% to 33.8% and free cash flow down 27.9% to $2.1B, on top of the September 4 surprise CEO-transition headline. The strongest point for the trade is that both legs are backed by real fundamental movement in opposite directions, not just narrative. The strongest point against it is that the entry logic cannot fire — both legs require the lower Bollinger band below 50, yet OKTA's sits at 151.4 and ADBE's at 274.1, meaning the trade needs a broad software selloff precisely while OKTA is rallying on its beat, and it could gate itself out if OKTA's RSI (now 66.1) crosses 75 before any entry. The backtest itself is thin: 3.7% total return across 22 trades over 60 months at a 50% win rate, with zero trades in the most recent 24 and 12 months, and no robust parameter setup was established because the sensitivity evaluation ran out of time. Insider filings for the period ended 2026-06-30 also cut awkwardly against the long leg: net open-market selling of roughly $11.5M across 3 OKTA holders versus only about $110K across 9 ADBE holders — those are as-filed disclosures for that reporting period, not current signals. The verdict flips to actionable if a software-wide drawdown pushes both lower Bollinger bands below 50 while the fundamental divergence holds.

Trade now: the pair is armed, but one entry condition is still far away

This is a rules-based long/long pair — long OKTA and long ADBE on the daily chart — with a backtested record over 60 months of 22 trades, a 50% win rate, a 3.7% total return, and a 4.2% maximum drawdown. The idea behind it is directional: Okta's beat-and-raise shows customers paying up for security as AI amplifies cyber threats, while Adobe faces AI erosion of a legacy franchise plus a surprise CEO change, per the idea's thesis. The entry logic requires, for each leg: daily RSI (14) above 20, the lower Bollinger band (20) below 50, price above zero, and a positive MACD histogram. Here is where things stand right now. On OKTA ($170.60): RSI is 66.1 (met), price is above zero (met), the MACD histogram is +7.5 (met) — but the lower Bollinger band sits at 151.40, far above the below-50 requirement. On ADBE ($266.51): RSI is 39.1 (met), MACD is +8.9 (met), price above zero (met) — and again the Bollinger condition is the blocker, with the band at 274.10 versus a below-50 trigger. That single condition is what 'wait' means concretely: neither leg can trigger until both names sell off hard enough to compress their lower bands under 50, i.e., a broad risk-off flush in software rather than the current OKTA strength (it sits just 1.4% below its range high). Risk framing when triggered: the strategy uses a fixed stop loss at a 2.5% loss per position and a take profit at a 4.9% gain — roughly 2:1 reward to risk — with secondary stops below the second support level and an overbought RSI exit above 75 after 60 bars. With OKTA's nearest support at $170.00 and ADBE's at $250.88, both legs are near the top of their local ranges, which is exactly why no entry is live today. Do not chase; let the rules pull you in.

The gap between Okta's operating leverage and Adobe's margin compression is the thesis, and it's visible in the numbers

The fundamental divergence the idea describes shows up directly in the numbers. Okta's latest quarter, ending July 31, 2026, delivered revenue of $805M, up 5.2% sequentially and up from $728M a year prior — consistent with the Yahoo Finance report of a 20% stock jump on a beat-and-raise quarter driven by AI-amplified security demand. Operating margin more than doubled quarter-over-quarter to 13.3% from 7.3%, net income climbed 57% to $116M, and net margin expanded to 14.4% from 9.7%. Those are operating-leverage signatures, not one-time items, and they support the idea's claim that the AI security spending wave is translating into profit, not just revenue. Adobe, by contrast, shows the exact profile the idea argues is under pressure. While its full-year FY2025 numbers remain formidable — 89% gross margin and a 99th-percentile free cash flow of $9.85B within its sector — the most recent sequential read, through May 29, 2026, shows revenue growth slowing to just 3.4% quarter-over-quarter while operating margin fell 10.5% to 33.8% and net margin fell 12.4% to 25.9%. Free cash flow dropped 27.9% to $2.1B, and net income fell 9.4%. That is a decelerating, margin-eroding franchise rather than a structurally healthy one, and it aligns with the Yahoo Finance headline that Adobe is ending an 18-year era as AI pressure mounts. The surprise CEO transition reported September 4 adds the leadership-vacuum uncertainty the idea cites as a compounding catalyst. Critically, both legs of the pair are supported by the data rather than just the narrative — Okta is improving on the metrics that matter most for a growth re-rating (operating margin, net income, free cash flow trajectory), while Adobe is deteriorating on the same metrics. The pair structure itself is also supported: the backtest produced a 3.7% return over 60 months with a maximum drawdown of just 4.2% across 22 trades, suggesting the divergence-capture approach has not required taking on large drawdown risk to generate returns. The most recent execution, on a 1-day timeframe, aligns with the idea's stated horizon of weeks, and the 60-month window spanning multiple…

Scores

• Conviction score breakdown: 50
• Thesis support: 70
• Trade readiness: 20
• Risk quality: 55
• Backtest evidence: 40
• Fundamentals trend: 65

Watch items

• OKTA — Lower Bollinger band (20)
• ADBE — Lower Bollinger band (20)
• OKTA — RSI (14)
• ADBE — Next quarterly XBRL fundamentals (quarter ending ~2026-08-29)
• OKTA — Insider net open-market activity (next ownership filing)
• ADBE — Insider net open-market activity (next ownership filing)