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 that personalises sales outreach messages based on prospect research, company signals, and engagement history. Includes automated research-to-email pipelines and personalisation at scale; distinct from email campaign generation in marketing which targets segments rather than individuals.
AI-personalised prospecting has a proven toolchain, documented ROI, and a mature vendor ecosystem -- but a stubborn execution gap separates the vanguard from the field. The core capability -- tailoring outreach messages using prospect research, intent signals, and engagement history -- is well understood, and platforms like Outreach, Apollo, and Gong now ship agentic features that automate the research-to-send pipeline end to end. Organisations with clean data and disciplined workflows report transformative results: 10x productivity gains, 62% revenue lifts, and conversion rates that dwarf manual baselines. The question is no longer whether the technology works. It is whether a given sales team can operationalise it. Only about 5% of organisations have integrated AI prospecting at scale, even as 88% experiment with it. Most pilots stall on data quality, deliverability constraints, and the persistent need for human judgement in high-stakes outreach. The practice is firmly good-practice -- accessible, proven, and commercially supported -- yet the gap between what the best teams achieve and what the median team manages remains wide.
The market is undergoing a fundamental shift from list-based to signal-based prospecting. Inbox saturation (30-50 cold emails per decision-maker weekly), AI-commoditised generic messaging, and enforced email authentication (SPF/DKIM/DMARC) have collapsed traditional cold-outreach reply rates from 5-8% (2021) to 1-2% (2026). Signal-based teams are pulling ahead decisively: independent testing shows signal-triggered personalisation achieves 3.4x more meetings with 60% fewer touches (11.2% vs 2.1% reply rates) than volume-based approaches, and trigger-based emails (funding, new hire, tech adoption) achieve 4-8% reply rates vs 1-2% from static lists. A Tolly Group evaluation of Apollo's integrated platform achieved 2.37% cold-to-meeting conversion (vs 0.5-1.5% industry benchmark) without email warming.
Agentic adoption is accelerating: 37% of B2B organisations deployed agentic prospecting systems in the past 12 months (vs 8% in the prior two years combined). High-performing teams synthesise real-time signals—company announcements, funding rounds, hiring surges, technographic shifts—and reach out within 24-48 hours, operationalising what leading practitioners call "Signal Stacking." Successful deployments apply deep personalization at scale, not templated variations: research-over-messaging wins because context-rich outreach with mediocre copy outperforms perfect copy with no context. Early adopters report 10x productivity gains and 62% revenue increases; Outreach and Apollo's 2026 releases (Research Agent, agentic GTM platforms) confirm continued platform maturity.
However, the market has bifurcated sharply. Results come from teams with clean CRM data, GTM engineering expertise, and mature signal infrastructure (Clay, n8n, attribution tools add $1-1.5K/month cost). The broader reality: 95% of AI prospecting pilots fail to deliver revenue acceleration, 73% of reps struggle with effective personalisation despite tools, and 50-70% annual churn on AI SDR platforms signals user dissatisfaction. Mechanical personalization—inserting company names and recent news into templates—is now broadly recognized as spam; one leader's 1,400 cold emails yielded zero responses, while hybrid teams (AI research + human writing) converted at 38% vs AI-only 11%. Email deliverability remains constrained (15-25% bounce rates), vendor costs run $35-60K annually, and the transition from list-based to signal-based systems requires architectural changes beyond tool swaps. The bifurcation continues: data-ready organisations with signal orchestration extract measurable value; the majority remain stalled between promising pilots and the infrastructure and discipline required for production adoption.
— DevCommX comparative analysis: AI achieves 4-12% reply rate vs manual 3-8%, $150-500 cost per meeting vs $400-900, 500-5K accounts/day vs 15-25; McKinsey: 10-15% revenue uplift, 20-30% efficiency gains; AI learns from reply feedback.
— Apollo GA embedded in HubSpot Breeze Prospecting Agent (April 2026); 230M verified contacts, signal monitoring, auto-research, personalized sequences; early testing: 2x higher response rates, 95% research time reduction, 70% ICP alignment.
— Martal Group analysis: only 25% of B2B companies leverage signal data; organizations using signal-qualified leads achieve 47% better conversion; AI platforms enable relevance at scale via signal monitoring, micro-segmentation, multi-channel execution.
— Practitioner analysis: signal-specific personalization achieved 18% vs 3% reply rate (5x differential); generic templates now detectable within 3 seconds; optimal: 75-125 word emails with {{detail}} reference to recent action.
— R[AI]SING SUN research consolidates 2026 B2B adoption (87% of sales orgs use AI, only 24% agentic); 5x more likely to hit targets; 53% cite poor data quality as top barrier; documents that adoption-to-impact gap is strategic divide.
— UnifyGTM documents three structural shifts: static→signal-triggered, single→multi-channel, manual→AI-native workflows; Salesforce: 92% with AI agents benefit prospecting; HubSpot: 70% report AI increases reply rates; AI-native platforms deliver compounding gains.
— Callbox case study: context-first adaptive engagement achieves 27% lead-to-appointment conversion vs templated scripts; personalized subject lines 26% higher open rate; demonstrates conversion uplift through contextualization.
— Outreach 2026 analysis: AI reduces research time by 90%, saves 10 hours per week, improves reply rates 3x, doubles conversion, achieves 90% demo contact within 24 hours; demonstrates AI productivity gains at scale.
2023-H1: Outreach and Gong introduced generative AI into sales engagement platforms. Market maturity for personalisation reached 98% organisational belief but consumer acceptance remained at 41%, creating implementation and trust barriers.
2023-H2: Apollo, Outreach, and Gong shipped production-grade AI outreach features with documented deployment gains (5x meetings, 3x reply rates). Sales team adoption accelerated to 71% impact on planning, but implementation gap widened: 60% of enterprises rarely used generative AI due to integration complexity and vendor lock-in concerns.
2024-Q1: Apollo demonstrated 3x meetings and 42% conversion rates; Gong's 1M+ opportunity analysis showed 35% win rate increases and vendor ecosystem expansion. Email spam policies tightened, raising deliverability barriers. Enterprise vendor lock-in concerns intensified, pushing demand for multi-vendor flexibility in AI sales stacks. Critical analysis highlighted AI's limits: effective for data organization but insufficient alone for high-touch early-stage prospecting.
2024-Q2: Outreach expanded AI feature set with Smart Email Assist GA (first/follow-up emails), Smart Account Assist, and Sequence Engagement Score. Apollo customer wins (RevBoss: 350 meetings, Maropost 60/40 split) demonstrated mid-market adoption. Outreach disclosed $69.9M pipeline and $14.9M closed-won revenue attributed to AI sequences, quantifying ROI. Critical signal emerged: AI-powered email filters and tightening spam policies commoditized cold outreach, with practitioner analysis arguing efficacy had approached zero and net ROI was negative—signaling eventual disruption of SDR/outbound model.
2024-Q3: Vendor ecosystem maturity accelerated with Forrester and IDC analyst rankings (Gong named leader in Revenue Orchestration). Outreach published benchmark data from 5,000+ customers showing 44% increase in meetings booked and 15% deal size increase, providing quantified ROI evidence. Market sentiment showed AI complementing (not replacing) SDRs, with 74% of GTM professionals managing 21+ SDRs. Practitioner discourse evolved to emphasize multi-channel signal integration and quality of personalization over template-based approaches, though concerns about lazy personalization and email deliverability headwinds persisted.
2024-Q4: Vendor ecosystem continued acceleration with Outreach announcing AI Prospecting Agent and Gong launching Gong Anywhere (AI insights in Gmail/Salesforce/Outlook). Gong survey of 600+ revenue leaders showed organizations using AI achieved 29% higher revenue growth, with email/engagement automation as top use case (63%). Named customer deployment (Aligned: 40%+ cold call conversion, 80% contact match) confirmed production adoption ROI. Critical perspective emerged: over-automation and generic templating had reduced efficacy; practitioner analysis emphasized contextual engagement over volume-based outreach as necessary for continued effectiveness.
2025-Q1: Outreach released AI agents for revenue teams with core prospecting capabilities; vendor ecosystem consolidation continued. Industry reports cited 10% higher open rates and 24% improved closing rates with AI personalization. Critical analysis highlighted implementation barriers: tool integration complexity, workflow fragmentation (53% rep time lost to tool-switching per Forrester), and need for manual effort to achieve effective personalization. Market sentiment shifted toward quality over volume, with practitioners identifying over-automation as the key constraint on further adoption rather than technology capability.
2025-Q2: Apollo achieved $150M ARR with 50k weekly users and demonstrated 46% improvement in meetings booked via AI Research Agent; customer Smartling reported 10x productivity gains using hyper-personalized emails. Outreach continued GA releases of AI agents for prospecting and engagement. However, independent field research documented persistent execution barriers: AI-generated emails remained templated and required manual customization; annual vendor commitments ($35-60k) offered limited flexibility; handoff quality on prospect replies remained problematic. Market sentiment reflected a plateau: vendors claiming sustained ROI (25-30%) and productivity gains (60%), but practitioners increasingly emphasizing quality-over-volume approaches and questioning feasibility of fully-automated personalization without significant manual effort.
2025-Q3: Apollo reached 9,000+ G2 reviews with #1 rankings in AI Sales Assistant; released company lookalike and custom AI filter features for prospecting. Outreach data showed 45% of revenue teams using hybrid AI-SDR models with 90% time savings on research/personalization. Independent analysis revealed growing adoption headwinds: 60% of organizations reported unmet AI expectations, only 20% saw revenue increases, and 42% of AI initiatives abandoned before production. Regie.ai case study showed successful deployment (50% higher open rates, 35% more demos), but remained proof-of-concept rather than market-wide shift. Market stalled between vendor claims of capability and practitioner documentation of execution complexity.
2025-Q4: Apollo and Gong launched agentic platforms for end-to-end prospecting automation, signaling continued platform maturity. Industry survey of 250+ leaders confirmed 80% of teams using AI in prospecting with 48% adopting AI outreach agents as mainstream tool. Concrete 2025 deployment metrics showed 30% conversion increases, 40% time savings, and $3.50 ROI per dollar, validating adoption at scale for data-ready organizations. However, critical assessments documented persistent execution barriers: email deliverability (15-25% bounce rates), data quality issues, manual personalization effort, and high TCO (Gong $400-500/user monthly). Market bifurcated between successful early movers with clean data and struggling broader cohort facing unmet expectations and ROI barriers.
2026-Jan: Early 2026 deployments confirmed mixed market momentum. Case studies reported strong outcomes: Snov.io experiment achieved 30% conversion boost with 7% reply rate and 0.5% bounce; Hashmeta case study documented 847% response increase and 63% cost reduction in B2B SaaS. However, adoption barriers intensified: 88% of organizations experimenting with AI but only 5% integrating at scale, 95% of AI pilots failing to deliver revenue acceleration. Market sentiment cooled as enterprises struggled to move beyond pilot stage; 73% of B2B reps continued struggling with effective personalization despite AI tools, while practitioners cautioned that true personalization still required substantial manual effort. AI agent hype peaked and began cooling as ROI skepticism and deployment complexity barriers emerged.
2026-Feb: Outreach shipped agentic features (MCP Server, Meeting Prep Agent, Deal Agent, Research Agent) signaling continued platform maturity. Independent analysis of Outreach documented 10x productivity gains for early customers. Voice agent deployment (Pete & Gabi) unlocked $139K from dormant accounts with 3.25% conversion. Apollo maintained 550K+ user base at 96% email accuracy. Outreach case study showed 62% revenue increase from AI agent implementation. Despite gains, market sentiment reflected integration complexity and recognition that true personalization required manual customization beyond AI automation.
2026-Apr: Market bifurcation between signal-based leaders and volume-based laggards sharpened further. Independent Tolly Group evaluation of Apollo's integrated platform achieved 2.37% cold-to-meeting conversion (versus 0.5-1.5% industry benchmark) without email warming, while a benchmark of 847 B2B organisations (2.3M interactions, $18.7B pipeline) found 37% deployed agentic systems in the past 12 months, with a projected revenue productivity gap exceeding 40% between agentic leaders and laggards by 2027. At the same time, the quality threshold became explicit: one practitioner's 1,400 AI-generated cold emails yielded zero responses, and hybrid teams (AI research, human writing) converted at 38% versus AI-only 11%—confirming that mechanical personalisation is now broadly identified as spam and that signal orchestration, not message volume, is the differentiating factor.
2026-May: April-May scan confirms market consolidation around signal-based and hybrid models. Vendor updates (Outreach April release: 90% research time reduction, 3x reply lift, 11-day cycle acceleration) document continued productivity gains for data-ready teams. May scan adds: Apollo embedded in HubSpot Breeze (230M contacts, 2x response lift), industry reports showing 87% of sales orgs use AI but only 24% deploy agentic systems (R[AI]SING SUN), and 47% conversion lift with signal-qualified leads (Martal). However, new vendor and independent data surfaces critical adoption barriers: cold email deliverability collapsed 35-45% open rate (2022) to 12-18% (2026), with 1-3% reply rates on generic lists versus 5-12% on signal-seeded sequences; multi-vendor agentic SDR analysis (Belkins, Sopro, Woodpecker) revealed 50-70% churn within 90 days for full-replacement deployments and identified failure modes (drift, reward hacking); 88% of prospects now ignore suspected AI outreach; 53% of AI initiatives cite poor data quality as top adoption barrier. Hybrid outcomes outperform full automation 2.8x ($147K human vs $56K AI per 38,000 attempts), but require 40-60 setup hours, clean CRM data, and 10-15 hours weekly human oversight to achieve sustainable ROI. Practitioner analysis shows signal-specific personalization achieves 18% reply rates vs 3% for generic templates (5x differential). Market consensus: cold email volume outreach is functionally commoditized; value comes from research automation (cutting 20-minute tasks to 2 minutes), context-first frameworks (27% lead-to-appointment conversion), and multi-touch signal orchestration with human oversight.