Something is shifting underneath the revenue operations function, and it is moving faster than most teams anticipated. In the span of a single week in July 2026, a Tel Aviv-based AI startup raised $25 million to rethink how go-to-market (GTM) teams are structured, analysts upgraded Salesforce on the basis that AI bear cases look overdone, and a broader warning emerged from Apollo Global suggesting AI profits remain largely invisible outside of core tech. Taken together, these signals paint a complicated but important picture for RevOps leaders: the tools are evolving rapidly, but the returns are not yet evenly distributed – and the teams that close that gap first will have a meaningful advantage.
The Case for Rebuilding GTM Architecture From Scratch
Alta’s $25 million Series A is notable not just for the size of the round, but for the framing behind it. The company describes itself as an “AI System of Actions” for revenue teams – a phrase worth unpacking carefully. Rather than layering AI onto existing workflows as a productivity add-on, Alta’s pitch is that the entire RevOps architecture should be rebuilt around AI-native decision-making. That means AI does not simply assist a rep in drafting an email – it actively participates in sequencing, prioritisation, and pipeline movement.
This is a meaningful philosophical departure from how most organisations currently use AI in their GTM stack. Most teams have adopted AI incrementally: a forecasting assistant here, a call intelligence tool there. Alta and companies like it are arguing that incremental adoption leaves most of the value on the table. If the underlying architecture still assumes a human-first, tool-assisted model, AI cannot operate at the speed or scale it is capable of.
For RevOps leaders, the practical question this raises is straightforward: does your current sales pipeline management process actually benefit from AI, or is AI simply being layered on top of processes that were designed for a different era?
Salesforce, ServiceNow, and the Strategic Fork in the Road
The contrast between Salesforce and ServiceNow that emerged in analyst commentary this week is instructive for anyone thinking about vendor strategy. One is leaning into share buybacks – returning capital to investors – while the other is acquiring companies to expand its platform surface area. Neither approach is wrong, but they reflect meaningfully different bets about where value in enterprise software will be created over the next three to five years.
For Salesforce, the analyst upgrade – driven by the view that AI bear cases are overdone – suggests the market is beginning to price in real AI-driven revenue uplift rather than just potential. That matters for RevOps teams evaluating their CRM investments. If AI capabilities inside platforms like Salesforce are genuinely maturing, the calculus around Customer Acquisition Cost (CAC) and long-term platform value shifts. Teams that locked into multi-year contracts expecting slow AI progress may find themselves sitting on more capability than they realised.
If you are evaluating or re-evaluating your core CRM platform, the CRM Tools Directory offers a structured comparison across the major vendors to help you cut through marketing claims and focus on what each platform actually delivers for revenue teams today.
The Profit Gap – Why AI Returns Are Still Uneven
Apollo Global’s warning that AI profits are largely absent outside of core tech is the most sobering data point from this week’s news cycle, and it deserves serious attention from RevOps professionals who are being asked to justify AI tool spend to their CFOs.
The pattern that is emerging looks something like this: AI tools improve activity metrics – emails sent, calls logged, sequences launched – but the translation into improved win rate or Net Revenue Retention (NRR) is inconsistent and often slower than expected. This is not an argument against AI investment, but it is a strong argument for measuring it more rigorously.
The teams seeing real returns from AI GTM tools tend to share a few characteristics:
- Clean, structured CRM data – AI tools perform dramatically better when the underlying data is reliable and consistently maintained.
- A clearly defined Ideal Customer Profile (ICP) – AI-driven prospecting and sequencing tools need a sharp ICP to work from. Vague targeting produces vague results, with or without AI.
- RevOps ownership of AI tooling decisions – When AI tool selection is left to individual reps or siloed teams, adoption is patchy and measurement is nearly impossible.
- Instrumented feedback loops – The highest-performing teams are treating AI tools like any other growth experiment: hypothesis, measurement, iteration.
The compliance side of the GTM stack offers a useful parallel here. Kintsugi’s new integration with SHOPLINE for automated sales tax compliance is a relatively unglamorous example of AI doing exactly what it should: eliminating a low-value, high-risk manual process so that revenue teams can focus on work that actually moves the number. This is where AI ROI is clearest – not in replacing human judgment on complex deals, but in removing friction from predictable, rule-based workflows.
What RevOps Leaders Should Do Next
The convergence of these signals – new AI-native GTM platforms raising serious capital, established CRM vendors getting credit for AI maturity, and macro-level warnings about uneven AI profit distribution – creates a specific set of decisions for RevOps leaders heading into the second half of 2026.
First, audit your current AI tool stack against actual pipeline outcomes rather than activity metrics. If a tool is improving emails sent but not improving sales forecast accuracy or deal velocity, that is a signal worth acting on.
Second, take the “AI-native architecture” argument seriously, even if you are not ready to act on it immediately. Alta’s framing – that the GTM stack needs to be rebuilt rather than retrofitted – is going to become a more common conversation in the next 12 to 18 months. RevOps leaders who have thought through their position will be better prepared when that conversation reaches the boardroom.
Third, invest in the fundamentals that make AI work. Data quality, ICP clarity, and process discipline are not glamorous investments, but they are the difference between AI tools that deliver and AI tools that disappoint.
For a deeper grounding in how to structure your revenue operations function for this next phase, the CRM Guides library covers everything from pipeline architecture to AI tool evaluation frameworks. And if you want to stay current as this space moves quickly, the CRM Daily Newsletter delivers the most relevant developments directly to your inbox each week.
The AI GTM architecture debate is no longer theoretical. The capital is flowing, the analysts are paying attention, and the early movers are pulling ahead. The question for RevOps teams is not whether to engage with this shift – it is how fast and how deliberately to do so.