The AI landscape doesn't move in one direction — it lurches. Some techniques leap from experiment to table stakes in a single quarter; others stall against regulatory walls, technical ceilings, or organisational inertia that no amount of hype can dislodge. Knowing which is which is the hard part. The State of Play cuts through the noise with a rigorously maintained index of AI techniques across every major business domain — classified by maturity, evidenced by real-world adoption, and updated daily so you always know where you stand relative to the field. Stop guessing. Start knowing.
A daily newsletter distilling the past two weeks of movement in a domain or two — delivered to your inbox while the index updates in the background.
Each dot marks the weighted maturity of practices within a domain — hover for a brief summary, click for more detail
AI across the revenue cycle from lead identification to closed deal. The most consistently mature domain: three-quarters of practices are good practice, including lead scoring, pipeline forecasting, and conversation intelligence. CRM copilots are mainstream. The few leading-edge practices involve autonomous prospecting and deal-coaching agents. Momentum is moderate — most gains are incremental rather than transformative.
The headline: Sales AI works and almost everyone owns it, but only one company in five makes money from it. The winners are not the ones with the best software — they have clean data, the discipline to act on buying signals within minutes, and a person checking the AI's work.
AI now runs through every part of selling — scoring leads, forecasting revenue, analyzing sales calls, writing proposals, setting prices — and the tools are mature and proven. The problem is execution, not technology. Sixty percent of workforces now have access to AI tools, but only 20 percent of companies generate revenue from them. A small group of disciplined teams captures real gains — 3 to 15 percent more revenue, by McKinsey's measure — while the majority get little or nothing. The dividing line is not the software you buy; it is whether your customer data is clean, your sales and marketing teams agree on the rules, and you act on buying signals fast. If you are deciding where to invest, the lesson is blunt: fix your data and your process before you buy another tool.
Three major studies converged on the same uncomfortable number. Deloitte found 60 percent of staff have AI tools but only 20 percent of companies make money from them; MIT found 95 percent of corporate AI projects showed no financial return within six months; and the share of teams that can even prove their content AI pays off fell year on year, to 41 percent. Owning the tools is not the same as getting results — demand proof on your own numbers before you renew.
Pre-meeting AI briefings have gone mainstream. Three-quarters of enterprise salespeople now read an AI-written brief on the account before a first meeting, up from one in five two years ago, cutting research from hours to minutes. This is now table stakes for competitive selling — if your reps still prep by hand, they are slower than the field.
Pricing AI became a courtroom risk, not just a reputational one. California's new pricing law drew its first big class action, naming BP, Walmart, Albertsons and others over 1,700 petrol stations accused of coordinating prices through a shared AI tool, while regulators documented ride-hailing apps charging 42 percent more for identical trips. If you use any third-party pricing tool, have legal confirm it is not feeding off competitors' live data.
The "let it run by itself" approach keeps failing. Fully autonomous sales AI — software that acts on its own without a person prompting it — performed far worse than AI with a human in the loop (a person reviewing each output): autonomous outreach got almost no replies, and autonomous deal-chasing cut win rates by nearly a third. Keep a person on the send button.
Two EU regulations land this year. The EU AI Act's main rules apply in August 2026 and a new Data Act follows in December, both creating hard requirements for how customer data is governed. Customer data management shifts from a nice-to-have to a compliance obligation — assign an owner and start the audit now.
The "autonomous agent" wave will force a build-or-wait decision. Analysts expect more than 40 percent of these self-running AI projects to be canceled by 2027 as costs and risks become clear. Resist pressure to deploy self-acting AI across sales this year; pilot narrowly, keep a person in the loop, and let the early movers' failure rate inform your timing.
Buyers are turning against obvious AI content. Trust in AI-generated material has fallen from 60 to 26 percent in three years, and half of business buyers avoid materials they spot as machine-written. Make human review of customer-facing AI content a standing rule — it roughly doubles engagement versus pure-AI output.
The constraint is your data, not the tool. Most companies cannot make AI pay because the deal information never makes it cleanly into the system the AI reads. Three-quarters of organizations have customer records that are less than half accurate, and records go stale at over 20 percent a year. No model fixes a dirty database.
Speed beats sophistication. A buying signal loses about 60 percent of its value within four hours, yet most teams take 8 to 14 days to act. Responding within five minutes produces 21 times the qualification rate of responding within thirty — this is an operations and routing problem, not a software one.
Autonomous AI is fragile when you chain steps together. That is why "set it and forget it" pricing or outreach keeps backfiring — one documented case won 8 percent margin in three weeks, then lost 40 percent of sales over three months. The reliable pattern everywhere is AI doing the research and drafting while a person owns the decision.
Go deeper: the full Sales & Revenue briefing — the longer analytical write-up, plus every practice we track in this domain with its maturity rating, the tools to consider, and the evidence behind our assessment.