Why RingCentral Doubters Were Wrong Before A 204% Gain

If you only glanced at RingCentral’s headlines in 2026, the story looked mixed. Quarterly updates mentioned steady mid single digit revenue gains, soft billings and rising competition from bundled AI suites. Holding RingCentral from the start of the year would have returned 203.9%, including dividends. If you had been weighing those conflicting signals back in January, how could you tell whether the reward justified the real risk around AI, partners and margins?

Narratives are how investors here put a case on the record, with explicit assumptions about revenue, margins and the multiple. Those assumptions imply an estimated Fair Value.

The move put RingCentral in the middle of this trade. Scan 38 AI small caps for other companies exposed to it.

The Two RingCentral Narratives Investors Had To Weigh

The shares cost US$28.88 at the start of the period, and RingCentral investors were effectively choosing between two very different stories about where AI and hybrid work might leave unified communications.

On the bullish side, the Narrative put Fair Value at US$47.08, 63% above the start price. It leaned on assumed revenue growth of 5.8% and a profit margin reaching 15.0%, supported by AI products like RingSense and RingCX and deep carrier partnerships.

The bearish Narrative anchored Fair Value at US$27, 7% below the start price. It assumed revenue growth of 3.8% and a 7.3% margin, with pressure from bundled competitors, rising security and compliance costs, and the risk that AI suites would squeeze UCaaS pricing.

What The Results Changed For The RingCentral Case

RingCentral’s Q3 2026 filing showed revenue of US$676.0m and net income of US$564.7m, with net margin reported at 83.5%. That profitability, on the numbers, leaned toward the bullish margin story, although such an unusually high margin invites questions about one off items. Prior periods of softer billings and customer acquisition costs still kept the cautious case alive. Overall, the evidence cut both ways.

The main assumption this period turned on was margin durability. When you judge any other software stock, compare narrative margins against the reported net margin and check whether one quarter’s spike comes from repeatable operations or exceptional items.

What RingCentral’s Price Now Asks You To Believe

RingCentral now trades at US$83.39, after a 203.9% total return from the start of the year. The selected Narrative’s Fair Value sits below that level and frames the move as rich for a business still guided to mid single digit revenue growth and reliant on cash generation to carry the AI story.

Instead of questioning survival, the focus has shifted to whether AI products and telephony strengths can support that richer price if billings stay soft. Anyone paying today’s valuation is effectively assuming AI upsell and cash conversion hold up, so the key question is how the risk of heavy AI spend and modest growth in the Narrative weighs on that bet.

"While AI products such as AIR, AVA and ACE are seeing rapid early ARR traction, multiple reports of slowing billings and only mid single digit revenue guidance suggest that heavy AI R&D spend of more than US$250m each year may not translate into enough incremental top line to offset this spending, which could limit future operating margin expansion and net income growth."

Not everyone reads the same price the same way. → See the lower figure this Narrative lands on, and how it gets there

Where RingCentral’s AI Needs Run Next

RingCentral’s story centers on software and call quality. Your AI tools still depend on something physical behind the scenes.

Every conversation, transcription and agent handoff needs fast access to stored information. Without enough memory and storage, even polished apps can lag.

Another business focuses on supplying that capacity for AI servers and data centers. Its products help keep accelerators working instead of waiting on data.

As AI workloads spread into cars, factories and devices, that supplier’s role could shift again. Changing memory content per system may redefine which bottlenecks matter most.

The argument comes with a number attached. → Uncover the AI infrastructure company trading 30% below one Narrative's Fair Value

Have feedback on this article? Concerned about the content? Get in touch with us directly. Alternatively, email editorial-team@simplywallst.com