Blog
Insights, tutorials, and updates from the Soku AI team.

ChatGPT Ads Reporting Delay: Clicks Every 15 Minutes, Spend in 7–8 Hours, Conversions in 24–48
OpenAI replaced the vague 'can appear within minutes' wording across the ChatGPT Ads help centre with explicit reporting freshness: impressions, clicks and CTR refresh about every 15 minutes, spend is delayed about 7–8 hours, CPC and CPM wait for settled spend, and attributed conversions take 24–48 hours. The exact before-and-after wording, and what it means for pacing checks, test reads, invoices and automated readbacks. Verified 2026-09-24.
13 min read · September 24, 2026

OAI-AdsBot: Why ChatGPT Ads Cannot Reach Your Landing Page
OAI-AdsBot is the only OpenAI crawler that decides whether your ChatGPT ad is allowed to run, and OpenAI's own published IP file for it covers 256 addresses — the smallest footprint of any OpenAI bot, and the one most sites forget to allowlist. The three layers that block it, what the help centre and the developer docs each say, the exact user-agent strings, and the crawler IP ranges as they stood on 2026-09-23.
12 min read · September 23, 2026

Best DSP Platforms for Advertisers in 2026: What Each One Actually Gives You Access To
Almost no demand-side platform publishes what it costs, so comparing DSPs on price is a dead end. Compare them on what they let you buy instead. A platform-by-platform guide to Amazon DSP, DV360, The Trade Desk, StackAdapt, Yahoo, Microsoft Invest, Viant, Basis and Adform — self-serve versus managed, the one published spend minimum in the category, and the spend level where each starts to make sense. Verified 2026-09-22.
14 min read · September 22, 2026

ChatGPT Ads Product-Feed Campaigns Are Country-Level Only — and custom_label Is Not Geo-Targeting
OpenAI's Ads developer documentation dropped its 'verify account support' hedge and now states flatly that new product-feed campaigns support geographic targeting and exclusions only at country level. In the same change it added a warning that a custom label in ads_metadata selects products, not audiences — a WA label does not keep the ad inside Washington. The exact before-and-after wording, why the Google Shopping habit breaks here, and what to do about regional catalogs. Verified 2026-09-22.
12 min read · September 22, 2026

Ad Fatigue: How to Diagnose and Fix It (2026)
Most teams call every performance decline ad fatigue and refresh the creative. Half the time the creative was fine. A diagnostic sequence that separates true frequency fatigue from audience exhaustion, auction shifts, tracking breakage and seasonality — and the fixes that match each cause.
12 min read · September 21, 2026

Best CTV Advertising Platforms, Ranked by How Easily You Can Actually Start (2026)
A ranking of connected TV advertising platforms in 2026 built around the question the listicles avoid: what is the real entry point? Verified published minimums, self-serve versus managed routes, measurement realities and the creative work CTV quietly demands.
13 min read · September 21, 2026

Best Dynamic Creative Optimization (DCO) Tools, Ranked (2026)
A scored ranking of the best dynamic creative optimization tools in 2026 — feed-driven DCO, AI-native creative generation, and platform-native asset combination — with a published-pricing audit showing how few DCO vendors will tell you what they cost before a sales call.
13 min read · September 21, 2026

Best Programmatic Advertising Tools: The Full Stack, Layer by Layer (2026)
Programmatic advertising is not one tool, it is seven layers. A practical map of the DSP, SSP, ad server, identity, verification, creative and measurement layers in 2026 — which ones you actually need at your spend level, what each costs, and where teams over-buy.
14 min read · September 21, 2026
ChatGPT Ads Pixel Cookie Expiry: The 30-Day and 365-Day Windows OpenAI Just Documented
OpenAI's Ads developer documentation gained a Cookie expiry section that, for the first time, names the two first-party cookies the ChatGPT Ads Measurement Pixel writes, their lifetimes, their refresh rules, and how consent removes them. What it means for your CMP, your attribution debugging, and your privacy notice. Verified 2026-09-21.
11 min read · September 21, 2026

Retail Media Networks, Explained for Advertisers (2026)
What a retail media network actually is, why two companies take almost all of the growth, the difference between on-site, off-site and in-store inventory, how the measurement is genuinely better and quietly self-marked, and how to decide whether an RMN belongs in your plan.
13 min read · September 21, 2026

ChatGPT Ads Self-Service Opens in Israel, Saudi Arabia, Turkey and the UAE — and Turkey Has No Currency Row
Between 2026-09-17 and 2026-09-18 OpenAI's Ads Manager Availability article went from 52 self-service countries to 56, adding Israel, Saudi Arabia, Turkey and the United Arab Emirates. The Minimum Campaign Spend table gained ILS 60 and SAR 50 the same day, and quietly cut the AED minimum from 70 to 65. Turkey got no lira row at all. The full before-and-after, what the missing TRY row means for a Turkish advertiser, and the home-country targeting restriction that applies to every new account in all four markets. Verified 2026-09-18.
13 min read · September 18, 2026

ChatGPT Ads Context Hints: How to Write Them, and Where OpenAI's Own Docs Disagree
OpenAI published a dedicated help-centre article on context hints on 2026-09-16 — the ad-group-level field that tells ChatGPT when your ads are relevant. It says write natural phrases, not keyword lists, and that hints are not targeting. But OpenAI's own Advertiser API reference still describes context_hints as 'free-form audience or placement hints' and every copy-pasteable example in it is a bare keyword list. The exact wording from both surfaces, the 2,000-hint ceiling, the replace-not-append trap, and the undocumented exclusion_hints field.
14 min read · September 17, 2026