Picking Your Post-CPQ Path for Mid-Market RevOps: Revenue Cloud, Composable, or Wrap-with-Agents

Salesforce CPQ went End-of-Sale on March 19, 2025. End-of-Life is partner-projected for 2029–2030. That gap matters. You have runway, not an emergency, and you should use it to make a path decision your CFO will still respect two acquisitions from now.

The trouble is that most existing guidance is single-path advocacy in disguise. Salesforce partners point you to Revenue Cloud Advanced. Composable vendors point you to DealHub or Nue.io. Your AE will quietly leave the Growth tier off the proposal because Advanced pays them better. None of this is malicious, but it does mean the default choice fits roughly a third of the mid-market teams it gets pitched to. Revenue Cloud Advanced wins by inertia, not by fit. Nobody gets fired for choosing Salesforce, and that's the problem.

This is the framework we use at Simplementix when a RevOps leader at a 200–2,000-employee B2B SaaS company asks us which post-CPQ path is right for them. It's vendor-neutral. It uses company-shape signals rather than vendor preference, and it names the licensing trap that nobody on the seller side will surface for you.

If you haven't read our CPQ End-of-Sale migration checklist, start there. This piece picks up where that one stops. Once you know you need to move, the next question is where to.

The three paths, without the marketing spin

There are three credible paths off Salesforce CPQ in 2026. Renewing CPQ through ~2029 and migrating later is technically available, but the SI capacity model makes that a delay, not a path. We'll come back to why.

Path 1: Revenue Cloud Advanced / Agentforce Revenue Management (ARM). Salesforce's official successor. Same vendor, same platform, same data model in name only. It is not a managed upgrade. There is no codebase shared with legacy CPQ. Custom Apex, Visualforce templates, product rules, approval matrices, and pricing logic all get rewritten. SI partner benchmarks (Cloudely, Plative, Kugamon's public LinkedIn analysis) put reasonably configured orgs at 9–12 months. Heavily customised orgs land at 14–18 months and $400K–$750K of implementation services. Salesforce's public price book lists Revenue Cloud Advanced at $200 per user per month and Revenue Cloud Growth at $150.

Path 2: Composable best-of-breed. Replace CPQ with point tools, typically a unified vendor like DealHub (which acquired Subskribe in November 2025), or a stack across Conga, Nue.io, or Tackle. You keep Salesforce as CRM, you replace the CPQ layer entirely, and you integrate through documented APIs. Vendor case studies report 6–10 week deployments for disciplined onboarding programs, though those numbers come from vendor marketing rather than independent analyst research. The honest trade is single-vendor accountability for an architecture you have to govern yourself.

Path 3: Wrap-with-Agents. Keep legacy CPQ as the data layer, lay Agentforce Revenue Management on top to fix the worst friction points. The Summer '26 release explicitly targets legacy CPQ pain: full-height Sales Transaction Line Editor, guided ramp wizard with trial periods, live Browse Catalog with real-time pricing, a Generate Document LWC that no longer requires an OmniStudio licence, and Slack-native approvals that sync back into Salesforce. None of this changes the underlying data model. What it does is buy you 12–18 months of credible runway while you plan a real move.

The four axes that actually route the decision

Most decision frameworks for this question stop at "what does Salesforce recommend?" That's not a framework. It's a sales motion.

The four axes that actually predict which path will work for a mid-market RevOps team are CRM count, catalog complexity, M&A trajectory, and AI maturity. Each one shifts the answer.

Axis 1: How many CRMs will you run in three years?

This is the question servicePath (a quote-to-cash analyst firm that publishes mid-market RevOps research) surfaces, and most other frameworks miss. If the honest answer is one and it's Salesforce, Revenue Cloud Advanced is on the table. If the answer is "two or more, because we acquired a HubSpot shop last quarter," or "we don't know, because corp-dev moves faster than IT," then RCA's single-CRM consolidation thesis works against you.

Mid-market RevOps leaders rarely loop corp-dev into CPQ decisions, and CPQ decisions rarely show up in M&A diligence. Over a five-year horizon those two things are tightly coupled. A composable path absorbs CRM diversity. A consolidated Salesforce path resists it.

Axis 2: How complex is your product catalog?

If you have fewer than 500 SKUs, linear quote-to-cash, and approval chains a senior RevOps manager can describe on a whiteboard, your data model is not the problem. Your UX is the problem. That is exactly what wrap-with-agents solves.

If you have 1,000+ SKUs, multi-segment ramps, heavy product dependencies, and approval matrices that route through legal and finance differently per region, the wrapper layer cannot save you. You need a new data model. The choice narrows to RCA Advanced or composable.

Axis 3: How active is your M&A pipeline?

If you've closed two acquisitions in the past 24 months or expect to in the next 24, your CPQ choice is an M&A architecture choice. RCA Advanced wants you to consolidate onto one Salesforce org as the source of truth. Composable assumes you'll run multiple CRMs and orchestrate around them. The wrong answer here is expensive. The avoided-migration cost per acquisition we see in mid-market practice runs into six figures, and synergy capture slows materially when the acquiring company has consolidated onto a single-CRM architecture that resists the acquired CRM's data model. servicePath has published the most useful framing on this point.

If your M&A activity is genuinely zero and the cap table makes that durable, this axis collapses and the other three drive the decision.

Axis 4: How mature is your agentic AI practice?

AI maturity is the newest axis. It barely existed 18 months ago. Agentforce Revenue Management ships agentic capabilities that depend on data hygiene, governance, and topic boundaries. If your team has already shipped an Agentforce use case to production, in service or sales or anywhere else, you have organisational muscle for the wrap-with-agents path. If you have not yet deployed Agentforce anywhere, starting with a Revenue Management wrapper is starting two hard projects at once.

Our Agentforce governance framework covers the readiness signals here in more detail. If you're not ready for Agentforce generally, you're not ready for ARM as a wrapper specifically.

The decision matrix

Run your team through the four axes and you land in one of three zones.

Zone A. Revenue Cloud Advanced fits when: you operate one CRM and intend to keep it that way, your catalog has real complexity (1,000+ SKUs, multi-segment ramps, region-specific approval logic), your M&A activity is low or absorbed inside the existing Salesforce footprint, and your AI maturity is high enough to absorb ARM as part of a planned 12-month implementation. This is the right path for a small portion of mid-market teams. It is rarely the right path simply because "we're already on Salesforce."

Zone B. Composable fits when: you run multiple CRMs today or expect to within 24 months, your M&A pipeline is active, you have engineering capacity to govern API contracts, and you'd rather invest in orchestration discipline than in vendor lock-in. The DealHub–Subskribe consolidation reduced the integration overhead here significantly. The trade-off is that you own the architecture, and that ownership doesn't disappear.

Zone C. Wrap-with-agents fits when: your catalog is simple enough to live inside the legacy CPQ data model for another 12–18 months, you've already shipped Agentforce somewhere, and you need to buy time to either build internal consensus for a real migration or wait for your corp-dev situation to clarify. This is a bridge. Treat it as a permanent stay and it accumulates wrapper debt that complicates whichever migration eventually arrives.

The licensing trap nobody will surface

Independent procurement advisors estimate that roughly one in three customers proposed Revenue Cloud Advanced have requirements fully served by Revenue Cloud Growth. The list-price gap is $50 per user per month. On a 200-seat deployment that's $120,000 a year of unnecessary licence cost, plus harder renewal leverage because you're now locked into Advanced's roadmap dependencies.

The reason this happens is structural. The per-seat list-price gap is 33% in Advanced's favour, which carries straight through to the AE's quota credit. There's no incentive on their side to surface Growth unless you ask for it explicitly.

The honest test is whether your CPQ Standard implementation required heavy Apex customisation. If your quote-to-cash is relatively linear and your CPQ has held up without significant code, Growth is very likely a fit. Run the licensing arithmetic at your desk before your next AE conversation. Walking in with the math already done is the only real leverage you have.

The mistake most teams make

The default mid-market mistake is treating this as a CRM vendor choice instead of an architecture decision. "We're on Salesforce, so we're going to Revenue Cloud" is not a decision. It's a deferral.

The second most common mistake is treating "wrap-with-agents" as a way to avoid the decision entirely. Summer '26 is a credible 12–18 month bridge for the right team. It is not permanent. The teams that get hurt are the ones that wrap, declare victory, and find themselves making the same RCA-or-composable choice 18 months later with an additional layer of wrapper debt to unwind.

The third mistake is renewing CPQ through End-of-Life without modelling the 2027 talent crunch. RCA-trained partners are still building capacity, and consulting rates are rising as deadlines compress. Mid-market teams that defer evaluation past Q1 2026 will pay materially more in 2027 for less-senior implementation help.

Where to start

Before you talk to any vendor, run your team through the four axes. Document your answers. The conversation with Salesforce or DealHub or anyone else gets dramatically more useful when you walk in with "we're a Zone C right now, transitioning to Zone A in 18 months" instead of "what do you recommend?"

If you want a vendor-neutral run at the framework with your numbers, that's a workshop we run. We map your team to a zone, model the Growth-versus-Advanced licensing arithmetic with your seat count, and hand you a written recommendation you can take into your next AE conversation.

→ Book a Post-CPQ Path Assessment

You have runway. Spend the next sixty days on the decision, not the migration. The decision is where the leverage lives.

Related articles

Lightning Sync to Einstein Activity Capture: a 4-sprint migration playbook for mid-market RevOps

Salesforce retires Lightning Sync in August 2026. Migrate to Einstein Activity Capture without breaking dashboards: a 4-sprint playbook for mid-market RevOps.

Read Article →

Who Has Access to Your AI Agent? A Governance Framework for Agentforce in Production

A six-layer governance framework for Agentforce in production: who runs the agent, what it can touch, and why permission set hygiene comes first.

Read Article →