What H1 2026 Revealed About the CPQ Market

June 16, 2026

Looking back at the first half of 2026, I would not describe the CPQ market as slow. I would describe it as more selective, more complex, and more architecture-driven.

The most interesting change is not simply that CPQ vendors are adding AI or that buyers are reviewing more options. The bigger shift is that CPQ is increasingly becoming a question of commercial architecture: where product, price, quote, contract, order, billing, and revenue logic should live — and how much of that logic can be safely exposed to AI, APIs, portals, and other front-end experiences.

By commercial architecture, I mean the system and operating model that governs how product rules, pricing logic, approvals, contracts, orders, billing, renewals, and revenue data work together. The quote is only one output. The more important question is which systems control the rules behind that output.

These observations are based on Novus CPQ’s H1 2026 market work, including 56 CPQ Briefing documents, ongoing vendor briefing updates across a broad CPQ vendor base, regular conversations with CPQ vendors and approximately 17–20 system integrators, more than ten customer and prospect discussions, and several CPQ Sales Report updates and customer-reference conversations, including work around camos, Revalize, Engineering Intent, and Salesforce ARM. This is not a formal statistical survey, but it provides a useful cross-section of what CPQ stakeholders are discussing right now.

What felt different in H1 2026 was the combination of trends. Salesforce CPQ disruption is forcing architecture reviews. AI is changing how buyers research vendors and how system integrators deliver work. Usage-based and hybrid pricing models are pushing CPQ closer to billing and revenue operations. At the same time, buyers appear less willing to accept broad transformation messaging without clearer evidence of implementation scope, production readiness, and measurable value.

The first thing that stands out is that CPQ is becoming broader than “configure, price, quote.” That phrase still matters, but it no longer captures the full market conversation. More vendors are positioning around revenue lifecycle management, quote-to-cash, lead-to-cash, product-to-cash, pricing, billing, contracts, subscriptions, usage commitments, and downstream revenue operations.

This does not mean core CPQ capabilities are becoming less important. Product configuration, pricing logic, quote creation, approvals, proposals, and integrations remain essential. But the question is shifting from “Can this system create the quote?” to “Can this architecture govern the commercial outcome?”

That distinction matters because the quote is increasingly only one expression of logic that may also drive partner selling, e-commerce, renewals, amendments, usage commitments, approvals, billing events, and revenue reporting. A CPQ decision can no longer be treated only as a sales operations decision in many companies. It can affect pricing governance, contract structure, order management, billing accuracy, renewals, usage commitments, and revenue visibility.

AI is the second major theme, and this is where I would be careful not to overstate what is happening. AI is clearly influencing CPQ, but the most important H1 2026 observation is that AI is creating more architecture and governance questions than it is currently solving in production.

The more credible AI use cases in H1 2026 were practical: deal desk support, proposal assistance, discovery documentation, meeting-summary-to-action workflows, implementation scoping, product modeling assistance, controlled quote generation, and selective seller guidance. These can save time and improve consistency, but they are not the same as autonomous CPQ.

In most real-world CPQ environments, there is still a need for governed product data, pricing logic, discount controls, approval policies, security, auditability, and reliable downstream execution. Large organizations remain cautious about broad AI agents because of autonomy, data access, compliance, cost predictability, and ownership questions. The useful question is not whether a vendor “has AI.” Almost everyone can now say that. The better question is where AI is already used in live customer environments, what data it can access, what actions it can take, and where human validation remains required.

In other words, the market is not simply moving from “old CPQ” to “AI CPQ.” It is moving toward CPQ environments where governed commercial logic may be accessed through many more interfaces, including AI.

One AI-related area that I find especially interesting is product modeling. Many companies still do not have their full product portfolio in CPQ because product rules, dependencies, constraints, pricing logic, exceptions, and outputs are difficult to model and maintain. If AI can help turn spreadsheets, technical specifications, product documentation, or existing configuration logic into usable model structures or rule suggestions, CPQ coverage could expand to more products and use cases.

This could be especially relevant in manufacturing and industrial environments, where configuration knowledge is often spread across sales, engineering, product management, spreadsheets, ERP, and experienced employees. But the practical value of AI will be judged not by how quickly it creates a model, but by whether the model can survive engineering review, pricing validation, sales use, and future maintenance.

AI is also changing vendor discovery. Buyers, vendors, analysts, and consultants are increasingly using AI search and answer engines to compare vendors and summarize market options. This may become more important than many vendors realize. A vendor that is difficult for AI tools to understand may increasingly be difficult for early-stage buyers to find. For buyers, AI search can support early research, but it should not replace vendor briefings, customer references, implementation due diligence, and careful scope validation.

The Salesforce CPQ transition is another major market signal. Salesforce CPQ End of Sale is not only a Salesforce ecosystem issue. It is a broader CPQ market catalyst because it is forcing many companies to reconsider their future quote-to-cash architecture.

Some organizations are evaluating Revenue Cloud / Agentforce Revenue Management. Some are staying on legacy Salesforce CPQ for now. Some are looking at non-Salesforce CPQ alternatives. Others are considering hybrid models where Salesforce remains central for CRM, while CPQ or commercial logic may sit closer to ERP, billing, pricing, product data, or industry-specific systems.

A Novus CPQ LinkedIn poll on this topic received 84 votes and 6,380 impressions. I had hoped for more responses, so I would not position the results as representative customer research. Also, the responses came from a broader group of CPQ stakeholders, not only Salesforce CPQ customers. Still, the result is directionally interesting: 40% selected “evaluating non-Salesforce CPQ,” 28% selected “moving to Revenue Cloud / ARM,” 22% selected “staying on legacy CPQ for now,” and 8% selected “no decision yet.”

The takeaway is not that one path will dominate. The more important point is that Salesforce CPQ disruption is creating a rare market-opening moment. It is generating migration activity inside the Salesforce ecosystem, but it is also giving non-Salesforce CPQ vendors, ERP-centered approaches, and hybrid architectures a new reason to be considered.

Another interesting shift is that the traditional CPQ user interface may become less central. A quote may now start in CRM, an e-commerce flow, a partner portal, a spreadsheet, an email, a Slack interaction, a self-service portal, an API call, or an AI assistant.

That does not make CPQ logic less important. It makes it more important. Product rules, pricing controls, discount policies, approvals, subscription changes, usage commitments, billing triggers, and downstream order processes still need to be applied consistently. The interface may change, but the rules cannot become optional.

This is why headless, API-first, composable, and MCP-style access concepts are getting more attention. Some of this is still early, and some of it may be over-marketed. But the direction is worth watching. The next CPQ architecture debate may be less about “which screen do sellers use?” and more about “which system owns the rules?”

AI may also expose weaknesses in traditional CPQ pricing and services models. Seat-based pricing is familiar and easy to budget, but AI introduces different economics. If AI reduces manual work or changes how many users actively interact with CPQ, vendors may need new packaging models. At the same time, pure consumption-based pricing can create cost uncertainty for customers. The near-term direction seems likely to be hybrid pricing: base subscriptions combined with token bundles, credit wallets, capped AI usage, modular AI add-ons, or selected usage-based elements.

System integrators face a related issue. AI can support requirements gathering, user story creation, solution summaries, documentation, testing, proposal drafts, and configuration support. That may improve delivery productivity before AI fundamentally changes CPQ software itself. But it may also create an expectation gap if customers assume AI-enabled delivery should automatically reduce project cost, while SIs use AI mainly to improve speed, consistency, utilization, and margin.

The final H1 signal is that the market is active, but disciplined. I do not see a weak CPQ market. But I also do not see a market expanding aggressively across all segments. Many vendors and SIs appear to be targeting controlled growth rather than broad hiring. European vendors are often focusing on selected regions such as DACH, Benelux, the Nordics, Western Europe, and the UK instead of broad global expansion. Manufacturing CPQ demand remains uneven, with some strong pockets of activity but also longer decision cycles, regional variation, tariff uncertainty, and more budget scrutiny.

This discipline may be a positive sign. CPQ has often been damaged less by lack of ambition than by overpromising, weak scoping, poor data readiness, and underestimating long-term ownership. In H1 2026, more buyers appeared to be asking harder questions about implementation realism, demo-to-production gaps, ownership, and measurable value.

The broader conclusion is that CPQ is not disappearing into AI, CRM, ERP, or billing. Instead, CPQ is becoming part of a larger commercial architecture conversation. For customers and prospects, that means CPQ should be evaluated as part of a broader operating model, not just as a quoting application. For vendors, it means proving production value rather than relying on broad AI or revenue transformation messaging. For system integrators, it means repeatable delivery, scope discipline, and practical use of AI. For investors, it means looking beyond “old CPQ” and understanding where commercial logic, monetization, industrial configuration, and AI-assisted workflows are converging.

The companies that do well in the second half of 2026 will likely not be the ones with the broadest AI claims or the longest feature lists. They will be the ones that can show how product, price, quote, contract, order, billing, and revenue processes connect in a way that is practical, governed, and proven in production.