Is Decagon losing to Sierra?

Is Sierra already beating Decagon in AI customer service?

Yes. Sierra is currently ahead of Decagon in revenue, major enterprise customers, funding, and valuation. But Decagon is growing quickly, and the gap between the two companies is smaller than their valuations suggest.

Both startups were founded in 2023 to build AI agents that can handle customer conversations, access business systems, and resolve problems without human support. Sierra was founded by former Salesforce co-CEO Bret Taylor and former Google executive Clay Bavor. Decagon was founded by Jesse Zhang and Ashwin Sreenivas.

The rivalry became harder to ignore after Sierra raised $950 million at a $15.8 billion valuation in May 2026. Just four months earlier, Decagon had raised $250 million at a $4.5 billion valuation. Two huge rounds only months apart say a lot about who is funding whom, which we trace in our agentic AI report.

Their revenue figures tell a more interesting story. According to Sacra, Sierra reached approximately $200 million in annual recurring revenue in May 2026, while Decagon reached $100 million in annualized revenue in July.

Sierra has roughly twice the revenue and 3.5 times the valuation. Yet Decagon's estimated revenue more than doubled in seven months, and its latest enterprise wins show that it can compete for customers that would previously have seemed natural Sierra targets.

The question now is whether Sierra can turn its early commercial lead into a lasting advantage before Decagon closes more of the gap.

Is Sierra growing faster than Decagon?

Sierra is adding more revenue in absolute terms, but Decagon has been growing faster in percentage terms. Decagon's recent growth makes it difficult to argue that Sierra is pulling away across the board.

Sacra estimates that Sierra increased annual recurring revenue from $26 million at the end of 2024 to $130 million at the end of 2025 and $200 million by May 2026.

Decagon grew from approximately $10 million at the end of 2024 to $44 million at the end of 2025 and $100 million by July 2026.

We calculated the changes from their December 2025 baselines. Sierra added approximately $70 million in annual recurring revenue over five months, a 54% increase. Decagon added $56 million over seven months, equivalent to 127% growth.

The periods differ, so the percentages should not be treated as a direct comparison of quarterly growth rates. They do, however, reveal two businesses still expanding very quickly.

Sierra has the stronger absolute growth record. Decagon's achievement is reaching $100 million annualized revenue while maintaining triple-digit growth from its previous year-end level.

There is also a useful historical comparison. Sierra crossed $100 million in annual recurring revenue in November 2025, seven quarters after launching in February 2024. Decagon reached a similar revenue scale approximately eight months later. Its growth has been remarkable, although Sierra reached that milestone first.

Revenue metric	Sierra	Decagon
End-2024 annualized revenue	~$26M	~$10M
End-2025 annualized revenue	~$130M	~$44M
Latest 2026 estimate	~$200M	~$100M
Latest measurement	May	July
Increase since end-2025	~$70M	~$56M
Growth since end-2025	~54%	~127%
Latest revenue ratio	2× Decagon	0.5× Sierra

Is Decagon closing the revenue gap with Sierra?

Decagon has recently narrowed Sierra's relative revenue lead, although Sierra's absolute advantage has continued to grow.

At the end of 2024, Sierra generated approximately 2.6 times Decagon's estimated annualized revenue. By December 2025, that ratio had widened to about three times.

Using their latest available estimates, Sierra's revenue is now approximately twice Decagon's.

In other words, Decagon went from generating around 34% of Sierra's revenue at the end of 2025 to approximately 50% at the latest reported levels.

There is a catch. Sierra's absolute lead increased from approximately $86 million at the end of 2025 to $100 million when comparing the latest estimates. Also, the $200 million Sierra figure is measured two months earlier than Decagon's $100 million figure.

Decagon has improved its relative position. We cannot yet say that the dollar gap is narrowing.

The next comparable revenue figures will be particularly revealing. Decagon needs to generate larger absolute gains while Sierra continues expanding from a much bigger base.

Does Sierra have more major enterprise customers than Decagon?

Sierra has a clear lead among the largest corporations, although Decagon has already won customers big enough to challenge the idea that Sierra owns the enterprise market.

In its May 2026 funding announcement, Sierra reported serving more than 40% of the Fortune 50, meaning at least 21 of America's 50 largest companies.

Sierra provided another useful breakdown in February 2026. One-quarter of its customers generated more than $10 billion in annual revenue, while half generated more than $1 billion. A customer base that includes giants earning over $10 billion is hard to fake, and we tell who has real customers from who only promises in our report covering agentic AI.

That is an unusually enterprise-heavy customer portfolio for a company founded only three years earlier.

Decagon's customer list is also increasingly impressive. The company added more than 100 global enterprise customers during 2025, including Avis Budget Group, Block, and Deutsche Telekom. Its wider customer base includes Mercado Libre, Notion, Rippling, Duolingo, and Hertz.

Those companies span banking, telecommunications, travel, e-commerce, and software. Decagon has demonstrated that it can sell well beyond smaller technology startups.

Still, Sierra has published a much stronger breakdown of its Fortune 50 and billion-dollar customers. Decagon hasn't disclosed equivalent customer distribution figures.

For now, Sierra's lead in large-enterprise penetration is one of its clearest competitive advantages.

Is Sierra worth three times more than Decagon because its AI is better?

No. Sierra's much higher valuation reflects its larger business and stronger position among major enterprises, but we haven't found evidence that its AI consistently performs better than Decagon's.

Sierra raised $950 million in May 2026 at a $15.8 billion valuation, following a $350 million round at $10 billion in September 2025.

Decagon's valuation jumped from $1.5 billion in June 2025 to $4.5 billion in January 2026, when Coatue and Index Ventures led its $250 million financing.

Using Sacra's latest revenue estimates, Sierra is valued at roughly 79 times annual recurring revenue, compared with 45 times annualized revenue for Decagon. These are approximate ratios using revenue estimates and funding valuations from different dates.

That gives Sierra a valuation premium of roughly 76% relative to current estimated revenue.

Investors may be rewarding its Fortune 50 penetration, management experience, and established enterprise contracts. But none of those advantages tells us whether Sierra resolves customer problems more accurately or cheaply.

Decagon's valuation has also been rising much faster. Its January funding tripled its previous valuation in approximately seven months, while Sierra's May financing increased its valuation by 58% over roughly eight months.

The funding market has rewarded both companies. Sierra commands the higher price, while Decagon's recent valuation growth shows that investors remain confident in its prospects.

Can Decagon compete with Sierra's much larger funding?

Decagon can still compete, but Sierra has enough capital to make life considerably harder for its smaller rival.

Sierra's May financing brought its total funding to approximately $1.6 billion and left it with more than $1 billion available to invest. Decagon has raised approximately $481 million across its funding history, according to Sacra.

That gives Sierra substantially more financial flexibility to hire engineers, pursue international customers, and support complicated enterprise deployments.

AI customer service agents often need access to CRM software, billing systems, identity verification tools, payment platforms, and internal knowledge bases. The harder the workflow, the more work it takes to build and maintain a reliable deployment.

Sierra has leaned heavily into this approach, combining its software with engineers who work directly with major customers.

Decagon uses a similar deployment model but has invested in tools intended to reduce the amount of custom engineering required. Its Agent Operating Procedures, reusable templates, and Agent Workbench give customers more ways to configure and improve agents themselves.

If those tools genuinely reduce implementation costs, Decagon could serve more customers without increasing its engineering workforce at the same rate.

The problem is that neither company publicly reports implementation costs per customer or gross margins on comparable contracts. Sierra's funding advantage is measurable. How efficiently each company spends that money remains unclear.

Is Sierra's AI customer service actually better than Decagon's?

We found no reliable public benchmark proving that Sierra's AI agents outperform Decagon's across comparable customer service tasks.

Both companies publish strong results from real deployments.

Sierra reports an 80% AI resolution rate at Airtable. Decagon reports approximately 70% chat and voice resolution for Chime, an 80% chat deflection rate for the Duolingo English Test, and more than 90% support deflection for Flashfood.

These are substantial results. But comparing them directly would be misleading because the companies measure different workflows and sometimes use different definitions.

For example, deflection measures how often a conversation avoids human support, while resolution should indicate that the customer's problem was actually solved. A customer who gives up on a chatbot can count toward deflection under some measurement systems.

A September 2026 analysis by Agent Marketplace compared published customer case studies from both vendors. It found a median reported resolution rate of 70% across 14 qualifying Sierra studies and 70% across eight Decagon studies.

That result is interesting because Sierra's greater commercial scale hasn't translated into an obvious advantage in published resolution outcomes.

However, the analysis relies on vendor-selected case studies rather than controlled tests using identical customer inquiries and evaluation methods.

For now, Sierra looks stronger commercially, while the available product-performance evidence is much closer.

Can Decagon's AI handle as many customer calls as Sierra?

Decagon has already proved it can handle substantial production call volumes, although Sierra operates across a larger publicly documented enterprise customer base.

Chime's Decagon case study reports more than one million AI-handled voice calls per month, nearly 70% voice resolution, and more than 70% chat resolution. A million voice calls a month at one customer shows voice is catching on, and we follow rising technologies in our report on agentic AI.

Maintained over a full year, that call volume would exceed 12 million voice conversations at a single customer.

That is a significant production workload for a company that only recently entered the AI voice market.

Sierra reported in May 2026 that agents built on its platform had handled billions of customer interactions across its customer base. The examples included mortgage refinancing, insurance claims, order returns, and fundraising.

The measurements cover different scopes. Sierra's figure aggregates interactions across its platform, while Decagon's Chime number refers to monthly voice calls at a specific customer.

Nevertheless, the Chime deployment establishes something important: Decagon's smaller revenue base does not prevent it from operating high-volume AI voice services for a major financial technology company.

Sierra still has the stronger evidence of breadth across large organizations. Decagon has enough operational scale to be taken seriously in major enterprise evaluations.

Is Sierra still ahead of Decagon in AI voice agents?

Sierra appears to have the more established AI voice business, but Decagon has made voice a serious part of its offering.

According to Sacra, voice had overtaken text as Sierra's primary interaction channel by October 2025. That is an important milestone because phone support remains one of the largest and most expensive parts of customer service.

Sierra continued improving the experience in 2026, including the introduction of Voice Personas in August. These tools help enterprises create more natural and consistent spoken interactions.

Decagon has moved quickly in the same direction. Its March Proactive Agents release introduced outbound voice capabilities and customer memory. Its October Voice 3 release expanded the company's voice offering further.

We also have evidence of Decagon handling high-volume voice support through Chime, as discussed earlier.

The competition is now about more than whether an AI agent can answer a phone call. Enterprise customers need agents that recognize callers, handle interruptions, access account information, complete transactions, and transfer difficult cases without losing context.

Neither company provides a complete public breakdown of voice revenue, total calls, average response latency, or error rates across its entire customer base.

Sierra's earlier commercial traction gives it an advantage. Decagon's progress has made voice a much more competitive market than it was a year ago.

Can Decagon make AI agents easier to manage than Sierra?

Decagon has made customer control a major selling point, and its Agent Operating Procedures could help businesses update AI agents with less engineering work.

Agent Operating Procedures, or AOPs, allow teams to describe customer service workflows through natural-language instructions while retaining access to technical integrations and controls.

A support manager could change how an AI agent handles refunds, cancellations, or account restrictions without rebuilding the entire system.

In February 2026, Decagon introduced additional AOP templates and expanded its Watchtower evaluation system. It followed in March with Agent Workbench, which helps teams investigate agent behavior and make improvements.

Sierra is working on the same problem through its Agent SDK and Ghostwriter tools. Ghostwriter allows users to create and improve agents through natural-language interaction rather than relying entirely on traditional development processes.

In an April 2026 interview, Sierra co-founder Bret Taylor described implementations that previously took four weeks being reduced to as little as four days with Ghostwriter.

That is a meaningful claim if customers can reproduce those results consistently.

Decagon emphasizes workflows that business teams can manage themselves. Sierra has focused on combining automated agent development with its deeper enterprise implementation model.

We don't yet have independent figures comparing the engineering hours, update times, or production errors involved in making equivalent changes on both platforms.

This is one of the areas where Decagon could build a lasting advantage, but Sierra is clearly aware of the same opportunity.

Are Decagon's customers more valuable than Sierra's?

Sierra's customer portfolio is currently stronger in terms of documented company size, while Decagon has attracted many fast-growing digital businesses that could increase their AI spending substantially.

Decagon works with technology companies such as Notion, Rippling, Duolingo, Substack, and Perplexity. Their products often change quickly, creating customer support needs that can be difficult to manage with rigid automation.

Decagon has also expanded into more traditional industries through Mercado Libre, Deutsche Telekom, Avis Budget Group, and Hertz.

Sierra's customer portfolio includes Cigna, SiriusXM, Nordstrom, Rivian, ADT, Rocket Mortgage, and other major businesses. Its unusually high concentration of billion-dollar companies gives it access to potentially enormous support volumes.

However, the size of the customer's business only tells us part of the story.

A digital service with millions of customer conversations could spend more on AI support than a giant corporation deploying agents in only one department.

To compare the two portfolios properly, we would need revenue per customer, contract expansion rates, and retention figures. Neither company provides those metrics in enough detail.

Sierra has the stronger large-enterprise profile. We can't establish that its average customer is more profitable or likely to spend more over time.

Are Decagon's AI agents saving customers more money than Sierra's?

Decagon has published some striking customer cost reductions, but Sierra has also demonstrated large operational savings and additional revenue benefits.

Chime reports a 60% reduction in customer support costs following its AI transformation with Decagon. ClassPass reports a 95% reduction in the cost measure covered by its Decagon case study.

Those numbers are impressive, although they come from individual customer examples and should not be treated as typical savings across the entire platform. Savings quoted by single customers are a shaky guide to pricing, so we compare which business models are winning in our agentic AI report.

Sierra's Airtable case study reports an 80% AI resolution rate and automated work equivalent to approximately 95 full-time support agents.

Other Sierra deployments focus on different economic outcomes. Cigna has reported improvements in patient authentication, while Rocket Mortgage has used conversational AI to support customer engagement and conversion.

Decagon's customer examples demonstrate that AI agents can cut support expenses considerably. Sierra's examples cover cost savings alongside tasks that can directly affect customer conversion and business revenue.

A proper financial comparison would require the total cost of each platform, engineering expenses, model usage, and verified savings for similar customers.

Those figures aren't publicly available. Both companies have demonstrated clear business value, but neither has established a consistent financial advantage across comparable deployments.

Could Sierra's enterprise strategy become a problem?

Sierra's focus on major enterprises has helped it grow remarkably fast, but servicing complicated customers could become expensive if every deployment requires too much engineering work.

Sierra's February 2026 review highlighted its concentration of large customers, with half generating more than $1 billion in annual revenue.

The company has also shown it can launch sophisticated projects quickly. One major healthcare customer went live approximately seven weeks after kickoff. Sierra has published other fast deployments involving Nordstrom and Singtel.

In April 2026, Bret Taylor described how Ghostwriter had shortened certain agent implementation timelines from weeks to days.

These results suggest Sierra is making progress in reducing setup time. What we don't know is how much engineering work goes into each deployment or how those costs change after launch.

A seven-week project might be highly profitable if most of the integration work is reusable. It could also require a large team of specialists behind the scenes.

Decagon faces similar challenges, but its AOP templates and customer-facing management tools are designed to make more of that work repeatable.

This could become one of the most important economic differences between the companies.

Today, Sierra's larger customers and greater funding support its engineering-heavy approach. Over time, the vendor that delivers reliable agents with fewer implementation hours could have a significant margin advantage.

Are Sierra and Decagon becoming direct competitors in sales as well as support?

Yes. Sierra and Decagon are now expanding into customer engagement, retention, and sales, making their products increasingly similar.

Sierra initially built its reputation around automating customer inquiries across text and voice. It has since moved into mortgage workflows, insurance claims, customer outreach, and transactions that go beyond conventional support.

Its July 2026 Horizon announcement introduced agents intended to maintain customer relationships over days or weeks rather than treating every interaction separately.

Decagon has followed a similar direction. Its March Proactive Agents release added memory and outbound communication. In May, Guided Discovery extended its agents into product recommendations, purchasing decisions, and customer retention.

Decagon also introduced Assist in August to help human support agents manage more difficult cases. Its October product releases expanded voice capabilities and tools for configuring specialized agents.

They increasingly want to manage the full customer journey, from a first product question to a purchase, a service problem, and a possible renewal.

This expands the market they can address, but also increases direct competition between them.

Their customers may eventually use the same AI platform for support, sales, account management, and customer retention. Winning that platform decision could be worth considerably more than automating a company's support tickets.

Could Salesforce, Zendesk, or Intercom beat both Sierra and Decagon?

Yes. Salesforce, Zendesk, and Intercom could become an even bigger competitive problem for Sierra and Decagon because they already have relationships with the businesses buying AI customer service tools.

Salesforce controls a major part of the enterprise CRM market and has been integrating AI agents through Agentforce. Zendesk already handles customer support workflows for a large installed base, while Intercom offers its Fin AI agent within an established customer-service platform.

Those existing relationships provide an obvious advantage. Installed customer bases are a real edge, so we look at who is ahead, and who is falling behind, in our report on who's ahead in agentic AI.

A company using Zendesk or Salesforce might prefer adding AI automation through its current vendor rather than buying another standalone platform, especially if switching requires additional integrations, security reviews, and training.

However, Sierra and Decagon have built their businesses around replacing large amounts of manual customer service work. Their enterprise deployments suggest that specialist platforms can still persuade customers to buy a separate AI product when the results justify it.

The incumbents have distribution and existing workflows, while the startups can concentrate their engineering resources on improving AI automation.

Sierra's Fortune 50 penetration shows that specialist providers can already win large customers. Decagon's recent enterprise additions reinforce the point.

Still, Sierra's current lead over Decagon should not be confused with dominance of the wider market. Both face established software companies with much larger customer bases and significant resources.

Is Sierra's $15.8 billion valuation harder to justify than Decagon's $4.5 billion?

Sierra has a much bigger valuation to support, while Decagon needs less additional revenue to reach the same hypothetical valuation multiple.

Using Sacra's latest revenue estimates, Sierra trades at an implied equity valuation of approximately 79 times annual recurring revenue. Decagon's equivalent ratio is around 45 times annualized revenue. Sierra at 79 times revenue and Decagon at 45 invite a closer look, so we compare valuations in our agentic AI market report.

Both are expensive compared with conventional enterprise software companies, although their growth rates are also unusually high.

We can calculate how much revenue each would need at different hypothetical valuation multiples.

At 20 times annual revenue, Sierra would need approximately $790 million to support a $15.8 billion valuation. Decagon would need around $225 million to support its $4.5 billion valuation.

At 10 times revenue, Sierra would need $1.58 billion and Decagon $450 million.

These calculations use equity valuations divided by revenue rather than enterprise-value multiples, and they assume unchanged valuations. They are useful scenarios, not predictions of fair value.

The difference becomes particularly interesting when compared with their recent growth. Sierra must add hundreds of millions in annual recurring revenue to move toward conventional high-growth software multiples. Decagon has a smaller absolute target, though it also needs to prove that its recent growth can continue.

Sierra's premium suggests investors expect it to retain a stronger competitive position. Decagon has more room to surprise positively if it can continue closing the revenue gap.

Valuation scenario	Sierra required ARR	Decagon required ARR
Latest equity valuation	$15.8B	$4.5B
Latest estimated revenue run rate	$200M	$100M
Current rough multiple	79×	45×
Revenue needed at 40×	$395M	$112.5M
Revenue needed at 30×	$527M	$150M
Revenue needed at 20×	$790M	$225M
Revenue needed at 15×	$1.05B	$300M
Revenue needed at 10×	$1.58B	$450M

What would prove that Decagon is really losing to Sierra?

Decagon would be in genuine trouble if Sierra kept widening its revenue lead while winning more competitive deals and delivering better results for similar enterprise customers. The evidence available today doesn't show that pattern.

Sierra has a larger customer base among the Fortune 50, substantially more funding, and greater estimated revenue. Those are strong advantages.

Decagon, meanwhile, has continued adding major enterprise customers and more than doubled its estimated annualized revenue from its previous year-end level.

What we haven't seen is reliable evidence of customers systematically leaving Decagon for Sierra.

Neither company discloses detailed customer churn, net revenue retention, competitive win rates, or full contract economics.

Those numbers would reveal much more than another large funding round.

For example, if Sierra consistently won customers replacing Decagon, while Decagon's existing customers reduced spending, we would have strong evidence that Sierra's product or commercial strategy was superior.

We would also want to know whether Sierra's larger customers are expanding their contracts faster than Decagon's customers, and whether that additional revenue comes with healthy margins.

At present, Sierra leads commercially, but Decagon's growth gives us little reason to describe it as a failing competitor.

Can Decagon still catch Sierra?

Decagon can catch Sierra, but doing so will become harder as both companies grow. The next challenge is generating hundreds of millions in additional annualized revenue, rather than simply maintaining impressive percentage growth.

Decagon reached approximately $100 million in annualized revenue by July 2026. Sierra had already reached an estimated $200 million in May.

To close a $100 million revenue gap, Decagon would need to add $100 million even if Sierra stopped growing. Sierra's recent trajectory makes that scenario unrealistic as a working assumption.

Still, Decagon has several reasons to remain competitive.

It has moved into demanding enterprise environments, shown it can process large voice volumes, expanded into proactive customer engagement, and developed tools intended to make AI agents easier to manage.

The company also operates in a market where adoption is still changing quickly. Businesses are experimenting with how much customer support, sales, and account management they can automate, and vendor decisions are far from permanently settled.

Sierra's challenge is maintaining its growth while serving an increasingly large and complex customer portfolio. Decagon's challenge is turning rapid percentage growth into much larger absolute revenue gains.

Our judgment is that catching Sierra remains possible, particularly if Decagon's more configurable approach produces lower deployment costs. But Sierra's head start is substantial, and Decagon hasn't yet shown enough absolute growth to erase it.

Is Decagon losing to Sierra after all?

Partly true. Sierra is winning the commercial race today, but Decagon is still growing quickly enough that calling it a loser would be misleading.

Sierra has built a bigger business in a remarkably short time. Its estimated revenue, Fortune 50 penetration, funding, and enterprise relationships give it a strong lead.

Decagon has nevertheless become a serious competitor. Its revenue more than doubled from its previous year-end level, its enterprise customer base has expanded, and its product now competes with Sierra across voice, support, and proactive customer engagement.

Our analysis found an important difference between the two companies' financial positions and their demonstrated product results. Sierra's valuation is approximately 3.5 times Decagon's, while its latest estimated revenue is only about twice as large. Available customer case studies also fail to establish a consistent advantage in AI resolution performance.

There is good reason to believe Sierra deserves a commercial premium. Its ability to sell into the largest companies has already been demonstrated at scale.

But Decagon has been closing the relative revenue gap, and we have seen no convincing evidence that its product is losing customers or failing to handle complex enterprise workloads.

For now, Sierra is the clear commercial leader. Decagon is a credible challenger gaining ground proportionally, and the outcome of this competition remains open.