Pegasystems Inc 2026 Q3 Earnings Call

Good morning, everyone, and welcome to Pegasystems' Q3 2026 earnings call. Before we begin, I'd like to read our safe harbor statement. Certain statements made during this call and in the accompanying materials may be construed as forward-looking statements as defined in the Private Securities Litigation Reform Act of 1995. Words such as expects, anticipates, intends, plans, believes, will, could, should, estimates, may, forecasts, and similar expressions are intended to identify forward-looking statements. These statements speak only as of the date this statement was made and are based on current expectations and assumptions. Because these statements relate to future events, they are subject to risks and uncertainties that could cause actual results to differ materially from our current expectations for fiscal year 2026 and beyond.

Important factors that could cause actual results to differ materially are described in the company's press release announcing our Q3 2026 results and in our filings with the Securities and Exchange Commission, including our annual report on Form 10-K for the year ended December 31, 2025, as well as other recent SEC filings. Investors are cautioned not to place undue reliance on these forward-looking statements. Except as required by law, we undertake no obligation to update or revise any forward-looking statements to reflect subsequent events or circumstances. In addition, non-GAAP financial measures discussed on this call should be considered in addition to, and not as a substitute for, financial measures prepared in accordance with GAAP. Constant currency measures are calculated by applying the September 30, 2025, foreign exchange rates to all periods presented. Reconciliations of GAAP to non-GAAP measures can be found in our earnings press release.

With that, I'll turn the call over to Alan Trefler, founder and CEO of Pegasystems.

Thank you, Peter, and thank you to everyone joining us today. Before Ken covers our results, I want to just touch on the AI market context and why it is a major long-term opportunity for Pega. Our technology is differentiated and very relevant to the times we are in. However, candidly, our execution this year has not kept up. We are making real changes to better capture the opportunity, and Ken will be providing more detail in a moment. I would say the market status is confusion and a search for value. Enterprise leaders are facing tsunamis of AI hype. This is the most chaotic and confused technology market I have seen in my career. It makes the internet bubble of 2000 look like it unfolded in slow motion. There is real promise in AI, and occasionally real delusion.

This quarter, Salesforce announced Claudeforce, essentially tying its future to Anthropic, a company that has said publicly it is going after all of Salesforce's TAM, their targeted addressable market. Then in September, Anthropic called for a pause in AI development, and leaders at OpenAI and other frontier labs chimed in to support the idea. But in the months since, Anthropic has released three major Claude model changes. OpenAI has also released three major GPT models. Google has released a new Gemini model. That is not counting the new models from Mistral, Qwen, DeepSeek, or model labs that are using things internally they haven't released yet. I guess this is what a pause looks like these days. No wonder everyone is confused. The leaders I spoke to at businesses have actually stopped chasing the latest model.

They want AI connected to business operations in ways that deliver measurable value while maintaining governance and controlling costs. The market is full of ingredients, and enterprises need a recipe. That is what Pega delivers. The changes that Ken will talk about are focused on getting that recipe directly to the customers and senior leaders who need it. Much of the AI industry is selling models and tokens. Pega is offering something different, a structure for AI that delivers predictable outcomes and predictable costs built on how enterprises really operate. Our architecture harnesses the power of LLMs and our own statistical AI to transform how enterprises design, build, and run the workflows and decisions that power their businesses. Models are an important ingredient, but the recipe for AI demands structure you can put around the model to provide operational context that tells it what to do.

Our workflows wrap AI in a control and governance structure that feeds exactly the knowledge and context it needs at the right time. As Forrester put it in its very recent report on enterprise AI platforms, "Workflows, not insights, are the value driver for agentic AI." We apply that structure at every stage of AI transformation. At design, we help clients reimagine their workflows and decisions for an AI world. We build to help create and maintain the applications in a way that is understandable and accessible. We run them to create agentic workflows that behave predictably at scale. Leaders I speak to know they shouldn't just bolt AI onto legacy systems and outdated processes. They want to redesign the way they work.

Blueprint harnesses AI to solve a problem that no one else has, giving business and IT teams a way to reimagine their operating model together. AI speed. It takes the raw power of generative AI and gives it a structure built on decades of Pega's workflow expertise and industry best practices. Blueprint combines the raw power of models like Claude, GPT, and Gemini with the decades of our knowledge from the work we have done with leaders in various industries. It helps establish the rules, workflows, policies, and operating model upfront before an application goes into production. The result is a repeatable recipe for how businesses should run, translated directly into working systems. Here is what much of the AI play misses, that 95% plus of work a business does each day repeats.

Processes themselves will be transformed by AI, but once you figure out how to process a dispute or best onboard a new customer, you want to do it the same way at scale. Letting AI reasoning run roughshod over that repeatable work is risky and expensive. Blueprint uses AI to capture those repeatable processes. We design them to be far more efficient, and then run them consistently at massive scale, without a massive token bill. Not only is this far more cost effective, it is exactly the right approach for AI governance. By creating the workflows you plug directly into the Claude or GPT or AI tool of your choice with MCP connectors, you are building workflows that tell agents exactly what they are allowed to do, what data they can access, and what rules they must follow.

We believe the right way to govern AI is to design it into the processes from the start. This is exactly what Blueprint does and the Pega technology does. While others are selling control towers or kill switches to manage problems after they happen, we consider that madness. It is addressing the problem too late and in the wrong place. Our clients can not afford to rely on a kill switch. With our approach, they never need one. The key here is structure. By helping clients finish design their recipes in days and load Blueprint into Infinity to build the application, they are able to run with agents like Copilot, Claude, and Codex. We also let clients bring their own model. If for some reason they have something, either an activity or they have a preference.

We have added into our development environment more than 50 agent skills. The developer's environment and the Pega environment operate in a collaborative and comprehensive fashion. Developers can get the speed of agentic development, but we provide the structure that makes it understandable, reusable, and scalable. We understand that one of the pitches AI companies are making is code it yourself. The reality, we have competed against code it yourself for decades. Of course, AI coding tools are powerful. We use them ourselves, and we have built them into the AI assistants in Infinity Studio. This does not change the underlying question. AI agents generate lots and lots of code. Millions of lines. More code than we have ever seen before. Much of that code becomes very hard to understand and even harder to change 6 or 12 months later.

That is at the heart of the Pega difference. In Pega, the logic stays visual and transparent, not buried in code so dense that only agents can read it. Business and IT users can look at the workflows, the rules, and the configuration in Pega and see exactly what their application does. This is what it means to build for change. At runtime, Pega applications run agentic workflows that orchestrate agents with control and governance. In Infinity, any agent, a customer service chatbot, an employee assistant, even one outside of Pega, can discover and run Pega workflows and decisions. They don't need to re-reason through every step. Blueprint has already done that for them at design time. We've put the human in the loop, blessing the design, as opposed to sticking them in every time a transaction runs.

Pega agentic workflows call for agents to perform specific tasks, like reading a document or researching a case or client history. The workflow sets the task constraints. So each agent is controlled in what it can do and what it can see. This small, focused agent approach obviates hallucination risk and uses far fewer tokens. That's how we can offer a fixed case price with no token charge. I'm getting a lot of enthusiasm from clients as to this approach. At PegaWorld, I compared this approach to a chef using a test kitchen to perfect the recipes before opening a restaurant. Rather than reinventing every dish for every customer. Blueprint is that test kitchen. At design time, clients create a collection of business recipes. The workflows, decisions, rules, data, and organize them so that they can be approved, understood, and most importantly, changed.