Threat level 3 AI readiness 3.5/5 Updated Jul 5, 2026

Close

The inside-sales CRM with calling/SMS/email built in, not bolted on — and as of June 2026 a genuinely real voice AI agent (Chloe), running on top of a 12-year-old, rigid data model.

The bet

Close is betting that the phone-first outbound team, not the data model, is the durable moat — win the "CRM built for action" niche outright, then extend it with AI rather than redesigning the core around AI. Founded 2013 by Steli Efti (CEO), Anthony Nemitz and Thomas Steinacher; bootstrapped and profitable off a single ~$250K–$500K 2013 seed round (SV Angel, Spark Capital), no institutional capital since. CEO-stated revenue crossed $40M in 2025 and "just surpassed $45M" by June 2025, trending toward $50M by year-end. ~11,500+ businesses run on Close as of the Chloe launch, with a ~110-person fully-remote team per the CEO's own account.

Why they're a credible #3

Close just shipped a real, production-scale voice AI (818K+ beta calls) into a base of 11,500+ paying businesses — this is not a roadmap slide. But it's an AI layer added to a 2013-era leads/contacts/opportunities schema that reviewers already call rigid. That gap is the opening.

Snapshot

$250–500K

total raised · 2013 seed only

$40–50M

ARR range, CEO-stated 2025

11,500+

businesses on Close (Jun 2026)

$990/mo

10-seat Growth (annual)

~110–156

headcount (conflicting)

Review-site scorecard click a card → source

Strong, high-volume proof on G2 (2,000+ reviews, 85% five-star, Support 9.3/10) and solid Capterra/TrustRadius — but zero Gartner presence, so enterprise-leaning buyers who anchor on Gartner have nothing to check Close against.

Verification gaps (flagged, not asserted): Crunchbase, PitchBook, G2, and Trustpilot all returned HTTP 403 on direct fetch this session — funding, headcount, and review figures above are relayed via search-result snippets, not confirmed against the primary pages themselves. Gartner coverage: searched specifically, found none — treat as a gap, not a confirmed absence. Reddit/HN Close-specific discussion: none surfaced via general web search this pass.

What's genuinely good

Native calling UX, no bolt-on dialer stack — calling/SMS/email built directly into the CRM core, the #1 praise theme on G2/Capterra ("intuitive navigation," "ready to jump into right out of the box").
Best-in-class support — G2 Quality of Support scored 9.3/10, with reviewers repeatedly citing fast response times.
A real, usage-validated voice AI agent — Chloe calls leads, qualifies, books meetings, transcribes/summarizes, and auto-enriches CRM fields; beta alone ran 818,787 calls across 300+ businesses before general launch.
Fast time-to-value — vendor claims most teams are "up and running in under an hour," and the product is shipping adjacent surface area fast: native scheduling (May 7 2026), an official ChatGPT app (Apr 30 2026), reworked Opportunities reporting (Mar 2026), and free-tier web forms (Jan 2026).
Bootstrapped, profitable, not going anywhere — 12 years, no down-round or acquisition risk, a durable business model competitors can't easily out-wait.

What's weak

Call reliability bugs — the single most-repeated G2 complaint: "missed notifications, delays, and the inconvenience of needing to reopen the app frequently" (239 mentions).
Rigid data model, thin reporting — "limited customization options... especially in account hierarchy and reporting features" (111 mentions); no custom objects; pipeline opportunity reports "not fully customizable".
No native file upload — support reportedly directs users to link Google shared drives instead.
Can't track account-level signals — third-party comparison sites independently converge: Close can't natively track funding events, org charts, or competitive/technographic signals — strong on activity volume, weak on account intelligence.
Opaque usage-based total cost — calling, SMS, phone numbers, and AI credits beyond the pooled allotment all bill separately on top of seat price; independent reviewers say real spend for active-calling teams can run 2–3x the sticker price.

AI capability lens

Graded on shipped, usage-validated capability — not the launch-day press release.

Close · AI readiness3.5

Shipped & verifiable

Chloe Voice (general launch Jun 9 2026) — outbound AI voice agent: calls leads, qualifies, handles objections, leaves voicemails, books meetings; mandatory non-overridable AI-disclosure + recording notice on every call. Beta alone: 300+ businesses, 818,787 calls, 111,915 prospects reached, 6,424+ hours of conversation.
Chloe Notetaker — automatic call transcription + summarization on every call.
CRM auto-enrichment — Chloe fills fields (company size, business model, etc.) directly from call content.
Chloe Chat — plain-English conversational interface over leads/deals/pipeline activity, plus next-best-action and at-risk-deal surfacing.
Official ChatGPT app (Apr 30 2026) — generate reports, build lead lists, summarize interactions, create Workflows from inside ChatGPT.

AI strengths

Native voice AI, no third-party dialer/transcription stack to stitch together
Production-scale validation before general launch (818K+ beta calls), not a demo
Compliance baked in — mandatory AI-disclosure/recording notice reduces a real legal-risk surface
Close claims Chloe "operates entirely within Close" with data never leaving to a third-party AI vendor — notable, though unverified independently

AI weaknesses

English-only, US/Canada-only at launch — narrow footprint for a geographically diverse SMB base
~4 weeks old at general availability as of this research date — unproven at scale/long-term reliability
Usage-based AI credits stack on top of already usage-billed calling/SMS — real total cost opaque until deep in usage
Acts within a fixed leads/contacts/opportunities schema, no custom objects — a ceiling on how deep "autonomous CRM update" can go

So what

Chloe is real, not vaporware — but it's an agent bolted onto a 2013 pipeline tool whose own reviewers call the data model rigid. Our opening: build the pipeline and the agent together, so autonomy extends into flexible objects and reporting, not just phone calls. Close's usage-metered AI (credits + per-minute calling + SMS, reportedly 2–3x sticker price for active teams) is also a pricing-predictability angle to contrast against.

Customer voice

Market signal · 82 G2 reviews, analyzed

Close is the best-rated CRM in our entire export (4.56/5) — reps love the built-in dialer and speed-to-lead. The sharpest recurring complaint is an integration gap that maps to its own ICP: "The lack of a direct, native integration with Apollo is a significant drawback… we rely on workarounds to sync data." Plus a trust ding: "we joined at a certain level, then they rolled out a new membership system and we ended up paying a lot more." Reddit volume is low (13 posts) — a small, loyal, phone-first base.

→ Full voice-of-market analysis

Praise

"The sales pipeline and lead management was incredibly intuitive and ready to jump into right out of the box."Capterra reviewer · platform: Capterra

"Close's Quality of Support is outstanding, scoring 9.3, with many mentioning quick response times and helpful resources."G2 review synthesis · platform: G2

Complaints

"Users experience call issues including missed notifications, delays, and the inconvenience of needing to reopen the app frequently."G2 review synthesis (239 mentions) · platform: G2

"You can't upload files directly — support suggested linking to Google shared drives instead."Capterra reviewer · platform: Capterra

Reddit and Hacker News searches for Close-specific threads returned nothing substantive this session — a genuine gap in platform-native sentiment, not confirmed absence; general web search likely under-indexes both platforms for this vendor.

Pricing teardown

Per-user, tiered, with pooled AI credits and separate usage billing for calling/SMS. Annual billing shown; monthly runs higher:

Tier	Annual /user/mo	What unlocks	AI credits/mo
Solo	$9	1 user cap, 10,000 leads, no workflows	500
Essentials	$35	Unlimited contacts/leads, calling/SMS/email, no workflows	1,000 (pool to 10,000)
Growth ★	$99	Workflows w/ Chloe, Power Dialer, bulk email	1,500 (pool to 15,000)
Scale	$139	Predictive dialer, role-based permissions, unlimited call recording, live coaching	2,000 (pool to 20,000)

10-seat Growth = ~$990/mo ($11,880/yr) — the realistic "team CRM" tier since Essentials excludes Workflows/Chloe. At Scale (needed for predictive dialer + unlimited recording), 10 seats run ~$1,390/mo. Gotcha: outbound calling (~$0.02/min), phone numbers (from $1/mo), SMS, and AI credits beyond the pooled allotment all bill separately — multiple independent reviewers flag real spend for active-calling teams at 2–3x the sticker price.

Head-to-head (10-seat deal)

✓ We win when

The buyer wants a CRM that's AI-native from day one, not a voice agent retrofitted onto a 2013 leads/contacts/opportunities schema. We win on reporting and data-model depth (Close's own reviewers flag rigid account hierarchy and non-customizable reports, 111 mentions), on account-level intelligence (funding events, org charts, technographic signals — gaps third-party comparisons independently confirm Close can't cover), and on pricing predictability versus Close's stacked usage billing (seat + calling minutes + SMS + AI credits, reportedly 2–3x sticker price for active teams).

✕ We lose when

The buyer is a phone-first outbound team that wants the deepest native calling UX in the category, proven at 2,000+ G2 reviews and 4.7/5, from a profitable 12-year vendor with zero acquisition/down-round risk — and where Chloe's now-shipped, usage-validated voice AI (818K+ beta calls) lets their champion say "we already have a real AI agent, and it's cheaper to start than yours."

Watch

Chloe general-launched Jun 9 2026 — the changelog shows no further dated releases through end of the ~90-day window at time of research, either a quiet post-launch period or a changelog lag. Watch for: international/multilingual expansion of Chloe (US/Canada + English-only today), any move to custom objects or deeper reporting (would narrow our structural opening), and whether AI-credit pricing gets simplified or published per-action (would blunt our pricing-predictability wedge).

← Full matrix