SoundHound AI (SOUN) Q1 2026 Earnings Transcript

SoundHound AI reported Q1 revenue of $44.2 million, up 52% year over year, with automotive and IoT AI organic growth of 88% excluding acquisitions. Management also announced the LivePerson acquisition, new OASIS agentic AI platform, and 2026 revenue guidance of $225 million-$260 million, while projecting $350 million-$400 million in 2027 post-acquisition revenue. Despite a $25 million GAAP net loss and $26.7 million adjusted EBITDA loss, the strong growth, new product launch, and large strategic deal are likely to support sentiment.

Analysis

The market should read this as a story of accelerating top-line quality, not just headline growth. The deeper signal is that management is trying to turn the business from a collection of point solutions into an operating system for enterprise voice/agentic workflows; if OASIS works, the economic model shifts from project-like deployments to higher-retention, multi-vertical account expansion with materially lower servicing cost over time. That creates a second-order beneficiary set: customers with dense voice/contact-center usage get the fastest ROI, while API-only “wrapper” competitors and model-agnostic integrators are structurally disadvantaged because they lack proprietary runtime economics and embedded distribution.

The key near-term catalyst is not the current quarter but the migration cadence over the next 2-4 quarters. If existing large accounts adopt OASIS without meaningful churn, the company can re-rate on proof of gross-margin expansion and lower customer acquisition cost; if migration stalls, the whole thesis collapses into expensive growth with weak operating leverage. LivePerson is the swing factor: even modest attrition is already baked into the framing, so the real question is whether post-close cross-sell offsets integration leakage fast enough to keep 2027 targets credible.

From a risk standpoint, the biggest hidden issue is execution complexity disguised as platform simplicity. They are simultaneously integrating acquisitions, building proprietary models, and asking customers to migrate mission-critical workflows; any one of those steps slipping could widen losses before synergy benefits arrive. The contrarian read is that the stock may be less about current revenue growth and more about whether the market is underestimating the value of a vertically specialized, owned-model stack in enterprise AI—if runtime economics matter more than model size, the moat could be real, but only after 6-12 months of evidence.

Key Decisions for Investors

• Long SOUNW into any post-earnings pullback over the next 1-3 weeks; use a staggered entry because the stock likely trades on migration/cross-sell proof, not the quarter itself. Risk/reward favors upside if OASIS adoption data starts to show, but size modestly because execution risk is still high.
• Buy SOUN 6-12 month call spreads instead of outright shares to express a re-rating on OASIS and LivePerson synergy realization. The cleaner setup is a defined-risk upside bet into the next two reporting cycles, when margin and retention proof should surface.
• Pair trade: long SOUN / short a basket of generic enterprise AI middleware or contact-center software names over 3-6 months. The thesis is that proprietary runtime + vertical distribution should outperform pure integration-layer stories if customers prioritize ROI over feature breadth.
• If LPSN weakness persists, consider a relative-value long SOUN / short LPSN pair into the close of the acquisition process. The spread should tighten if the market starts pricing in asset-stripping plus cross-sell optionality rather than standalone business decay.
• Avoid chasing WMT off the headline unless subsequent disclosures show broad deployment expansion; this is more of a small optionality beneficiary than a primary earnings driver right now.