IBM Software Growth Slows: Will Recurring Revenue Support Recovery?

IBM's Q2 2026 software revenue rose 5% year over year to $7.8 billion, but organic growth was flat as Transaction Processing revenue fell 8% and several large deals were delayed. Offsetting the slowdown, Software segment profit increased 9% to $2.5 billion and margin expanded 110bps to 32.2%; Hybrid Cloud revenue grew 11%, Data revenue rose 19%, and OpenShift ARR reached $2.2 billion. IBM shares have declined 10.3% over the past year despite a 171.2% industry gain, while 2026 EPS estimates edged up 0.7% to $12.33 and 2027 estimates fell 0.6% to $13.22.

Analysis

The key inference is that IBM's weakness is more a capital-allocation displacement than clear demand destruction: enterprise budgets are being pulled forward into constrained infrastructure, delaying higher-margin software commitments. That creates a potential 1-2 quarter booking catch-up if server and memory availability normalizes, but it also exposes IBM to a more troubling possibility: delayed transaction-processing deals can be re-scoped toward cloud-native alternatives rather than simply deferred. The investment debate should therefore center on software backlog/conversion and renewal rates, not reported recurring-revenue mix alone.

IBM's productivity-led margin defense limits near-term EPS downside, but it cannot sustainably offset flat organic growth without raising the risk of multiple compression. Red Hat, HashiCorp and Confluent improve the strategic stack, yet acquisition-led data and automation growth needs to translate into cross-sell bookings; otherwise IBM remains a lower-growth incumbent competing against vendors with materially stronger cloud distribution. MSFT is best positioned to capture displaced enterprise software budgets through bundled productivity, security and Azure commitments; ORCL remains a beneficiary where customers are specifically funding AI/cloud capacity, although its valuation is more sensitive to infrastructure execution.

The non-obvious near-term beneficiary is the enterprise hardware supply chain: delayed software spending implies customers are prioritizing compute deployment, supporting demand visibility for DELL, HPE, SMCI and MU over the next 1-3 months. Contrarian upside for IBM emerges only if the delayed deals close without incremental discounting, producing an unusual combination of reaccelerating revenue and already-expanded margins. Falsify that view if the next report shows another sequential deterioration in transaction-processing revenue, weaker software bookings/deferred revenue, or a cut to 2027 EPS expectations.

Key Decisions for Investors

• Maintain a 3-6 month relative underweight in IBM versus MSFT: long MSFT / short IBM in dollar-neutral size. MSFT has superior budget capture and bundle economics; cover the IBM short if software organic growth reaccelerates above mid-single digits or management demonstrates material backlog conversion.
• Do not buy IBM solely on its discount to sector sales multiples. Establish a watch trigger for a tactical long only after evidence that delayed transactions are closing, measured by improved bookings/deferred revenue and stable-to-higher software margin; absent that evidence, the valuation discount is more likely a growth-quality discount.
• For a 1-3 month infrastructure-spend expression, prefer a basket long DELL and HPE, with MU as the higher-beta memory exposure, rather than adding IBM. Size modestly because supply normalization or an enterprise capex pause would reverse the budget-shift signal quickly.
• Monitor ORCL for a relative long opportunity versus IBM on any pullback tied to capex concerns. The trade is invalidated if cloud backlog/RPO growth decelerates materially or data-center capacity constraints postpone revenue recognition rather than merely shift it.