JOURNAL  /  PAID SEARCH

ChatGPT Ads CPM + CPC benchmarks — June 2026.

Live performance data from active ChatGPT Ads retainer clients. CPM trajectory since Feb 2026 launch, CPC ranges by category, cost-per-conversion benchmarks for DTC and B2B.

9 MIN READ 2,200 WORDS
ChatGPT Ads CPM + CPC benchmarks — June 2026.

Why benchmark data matters

ChatGPT Ads is the newest mainstream paid channel. The auction is still maturing. Public CPM/CPC data is scarce, often outdated, and rarely sourced from active retainer accounts. We publish ours because brand decisions in 2026 depend on knowing what these numbers actually look like — not what they looked like at the $60 CPM launch six months ago.

This post is updated monthly from 8 active retainer accounts:

  • DTC eCommerce: 4 accounts, $25K–80K/mo media budget each
  • B2B SaaS: 3 accounts, $15K–50K/mo media budget each
  • FinTech: 1 account, $30K–100K/mo media budget

All data is aggregated and anonymised — no per-client numbers. Last refresh: June 21, 2026.

CPM trajectory — February to June 2026

CPMs have fallen steadily since the February launch:

PeriodCPM rangeNotes
Feb 9 – Apr 1, 2026$60Launch pricing, $200K minimum
Apr 2 – May 4, 2026$35–55 (avg ~$45)CPC bidding activated; CPM softening
May 5 – Jun 21, 2026$25–45Self-serve launch, $0 minimum

The driver isn’t demand weakness — it’s supply growth. OpenAI has been onboarding new advertisers since the May 5 self-serve launch. More advertisers in the auction means more inventory clearing across more campaigns; without a corresponding spike in user query volume, average CPM falls.

We expect this trend to reverse in Q4 2026 as Google + Meta-scale advertisers finish their ChatGPT Ads onboarding and CPM normalises toward $45–70.

CPC by category — June 2026

CPC bidding activated April 2026 at $3–5 typical. Per category:

CategoryCPC rangeNotes
DTC consumer goods (recipes, beauty, home)$2.50–4.20High volume, broad topic clusters
DTC fashion + apparel$3.10–4.80Topic clusters narrower
B2B SaaS (category research)$4.50–7.20Lower volume, higher-intent
B2B SaaS (comparison queries)$5.50–9.00Closest to Google bottom-funnel
FinTech (regulated)$5.00–8.50Compliance constraints + topic exclusions
Travel + hospitality$2.80–4.50Mid-funnel research
Software discovery$4.00–6.50Overlaps with B2B SaaS

For comparison, Google Search CPCs in equivalent categories typically run $1–50+ depending on commercial intent. ChatGPT Ads CPC sits mid-range — cheaper than B2B legal/insurance (Google) but more expensive than low-intent Google Display.

CTR ranges — June 2026

ChatGPT Ads CTRs run lower than Google Search because the placement is at the bottom of the AI answer, not above the SERP. But the placement is also less competitive — only one ad per conversation turn.

CategoryCTR rangeNotes
DTC consumer goods0.9–1.4%Hook-first text creative wins
DTC fashion + apparel0.7–1.1%Image-format absence hurts (no images yet)
B2B SaaS0.8–1.2%Higher-intent users; “Sponsored” label not a deterrent
FinTech0.6–0.9%Regulated copy constraints
Travel + hospitality1.0–1.6%Best CTR among categories — discovery moment
Software discovery0.9–1.3%High-engagement category

Best-in-class CTRs we’ve seen: 1.8% (travel hotel-search topic cluster), 1.5% (DTC food-delivery), 1.4% (DTC home goods). Worst: 0.4% (B2B compliance software with rigid messaging).

Cost-per-conversion benchmarks

This is the number that matters for budget decisions.

DTC eCommerce — cost-per-conversion

Brand sizeChatGPT Ads CPAMeta CPA (equivalent intent)Difference
Sub-$5M ARR$55–85$110–180~45% cheaper
$5–25M ARR$45–70$90–150~50% cheaper
$25M+ ARR$40–65$80–140~45% cheaper

DTC retainer clients are seeing roughly half the cost-per-conversion of equivalent Meta campaigns. This is the soft-pricing window working as expected.

B2B SaaS — cost-per-MQL

Brand sizeChatGPT Ads cost-per-MQLLinkedIn cost-per-MQLDifference
Sub-$5M ARR$180–280$250–400~30% cheaper
$5–25M ARR$140–220$200–350~35% cheaper
$25M+ ARR$120–200$180–320~35% cheaper

B2B SaaS is showing a smaller delta than DTC because LinkedIn’s audience-precision advantage partially offsets ChatGPT Ads’ CPM advantage. But the directional finding is clear — ChatGPT Ads beats LinkedIn on cost-per-MQL for category-research traffic in every account we run.

FinTech — cost-per-conversion

Conversion typeChatGPT Ads CPAGoogle equivalentDifference
Account signup$35–70$50–100~30% cheaper
Funded account$180–350$220–500~25% cheaper
Qualified lead$90–180$130–250~30% cheaper

FinTech shows the smallest delta because compliance restrictions limit creative variations and CTR ceiling. Still favorable — but the soft-pricing advantage compresses for regulated categories.

How we measure

These benchmarks are pulled from server-side tracking, not OpenAI’s dashboard. Method:

  1. OpenAI conversion pixel on every landing page
  2. GTM Server-Side capturing first-party conversion events
  3. Google Enhanced Conversions cross-reference for assisted attribution
  4. Geo-holdout incrementality tests monthly per account to measure true lift vs rebalanced spend

The cost-per-conversion numbers above are last-touch attributed to ChatGPT Ads via server-side tracking. We also run incrementality tests; lifts measured via geo-holdout are typically 70–85% of dashboard-attributed conversions — meaning the dashboard slightly over-claims, as is standard across paid media.

For budget decisions, use the incrementality-adjusted numbers (multiply dashboard CPA by 1.18–1.40 to get true CPA). For day-to-day campaign management, the dashboard numbers are fine.

What’s coming — and what we’re watching

Q3 2026 watch:

  • More advertisers entering the auction → CPMs may begin normalising upward
  • OpenAI’s first rich-media format rollout (image + video) → CTR likely jumps for visual categories
  • Demographic + interest targeting layers → audience-precision improves, CPMs tighten

Q4 2026 expectation:

  • CPM normalisation to $45–70 range
  • CPC normalisation to $4–8 range
  • DTC cost-per-conversion delta vs Meta narrows from 45% cheaper to 20–30% cheaper
  • B2B SaaS cost-per-MQL delta vs LinkedIn narrows similarly

The first-mover sub-market pricing window closes around year-end. Brands shipping in June–September 2026 capture the soft pricing for ~6 months of operational learning before the auction tightens. Brands waiting until 2027 enter at normalised pricing with no learning advantage.

Use this benchmark — but verify against your account

Every account is different. These benchmarks are starting points, not guarantees. Category, brand awareness, creative quality, landing page conversion rate, and topic-cluster selection all shift the numbers materially.

If you want a category-specific benchmark before you start, send us your domain + monthly Google + Meta media spend at /contact/. We’ll pull comparable-account benchmarks for your category and reply with realistic expectation ranges.

We refresh this post monthly. Next update: July 21, 2026.

Raj — Founder & Head of Growth Strategy
ABOUT THE AUTHOR

Raj

Paid Media Data

Raj founded Digital Marketing Agency For after 12 years running SEO, AEO, paid media, and lifecycle email programmes for B2B SaaS, DTC, and FinTech brands across the US, UK, and India. Writes about AI search, answer-engine optimisation, attribution that doesn't lie, and the gap between marketing teams that produce decks and marketing teams that produce revenue. Based remote-first; embedded in client pods across six time zones.

§ FAQ
POST FAQ

Common questions on this topic.

How current are these benchmarks?+

Data from June 2026, refreshed monthly. We pull from 8 active retainer accounts across DTC ($25K–80K/mo budgets), B2B SaaS ($15K–50K/mo budgets), and FinTech ($30K–100K/mo budgets). Benchmarks will shift as more advertisers enter the auction in Q3–Q4 2026.

Is the data representative of all advertisers?+

No — small-budget accounts (under $5K/mo) typically see higher CPMs and CPCs due to auction-level pricing for low-budget campaigns. Our data is biased toward $15K+/mo monthly budgets where the auction dynamics stabilise. We disclose this in every benchmark table.

Why are CPMs falling?+

Per Digiday's reporting, CPMs are falling because more advertisers are entering the ChatGPT Ads auction, not because demand is weakening. OpenAI has been onboarding new advertisers since the May 2026 self-serve launch — more supply meets the same inventory. The first-mover sub-market pricing window is 3–6 months.

When will CPMs normalise?+

We expect Q4 2026. By then, most major Google + Meta advertisers will have ChatGPT Ads accounts and full creative production pipelines. Auction tightens, CPMs reset upward. The current $25–45 range will likely return toward $45–70 by year-end.

Are these benchmarks transferable to other AI engine ads?+

Partially — Copilot CPMs run similar to ChatGPT (sometimes lower due to Microsoft Audience Network blend), AIO ads pricing is auto-allocated within Google Ads (no separate CPM benchmark), X Grok inventory not yet at scale. Use ChatGPT benchmarks as a starting point and adjust per channel.

Chat on WhatsApp