# Adrails > Adrails is ad management with AI agents for Meta Ads and Google Ads. AI agents that build campaigns, run tests, monitor spend and explain performance across your ad accounts. Agents read the ad accounts and the business data connected to a workspace, build campaigns, run split tests, explain performance and prepare the next changes; automations run reports, alerts and pauses on a schedule or a threshold. Built by TBSS Labs, LLC. ## Key facts - Definition: Adrails is ad management with AI agents for Meta Ads and Google Ads. AI agents that build campaigns, run tests, monitor spend and explain performance across your ad accounts. - Company: Adrails is built and operated by TBSS Labs, LLC, founded by Thomas and run from Dubai, United Arab Emirates. Contact: thomas@adrails.ai. Public MCP repository: https://github.com/devthomy/adrails-mcp. Not affiliated with Meta Platforms, Inc. or Google. - Ad platforms: Meta Ads and Google Ads. - Business sources read next to ad spend: Shopify, Stripe, Klaviyo, Triple Whale, PostHog, RevenueCat and Google Drive. - MCP server: https://adrails.ai/api/mcp, Streamable HTTP with OAuth 2.1, used from Claude, ChatGPT, Claude Code, Cursor and Codex. Tools reference: https://adrails.ai/docs/mcp/tools. Machine-readable server card: https://adrails.ai/.well-known/mcp/server-card.json. - Control model: an agent prepares a change and an owner or admin applies it in Adrails after the platform is read again; a budget move is capped at 30% per change; what Adrails creates on a platform starts paused; an undo exists while the platform allows one; a proposal expires after 30 minutes. - Split tests: one variable per test, two to four variants, a verdict at 95% confidence and never before every variant has 10 conversions. - Plans: Starter $149 a month or $1,490 a year, 300 action credits a month; Pro $299 a month or $2,990 a year, 750 action credits a month; Business $599 a month or $5,990 a year, 2,000 action credits a month. Unlimited ad accounts and team members on every paid plan, every connector included, and no percentage of advertising spend. Billable reads and completed advertising changes share the workspace action credits, including calls from MCP clients. One request can use several credits; setup and discovery calls are free. ## Product - [Features](https://adrails.ai/features): What Adrails does at each step, from campaign builds to reporting. - [Give your next ad task to an agent.](https://adrails.ai/features/ai-ads-agent): Build, test and improve ad campaigns with specialist AI agents that read your ad accounts and business data before they act. - [Make the weekly review repeatable.](https://adrails.ai/features/ad-automation): Run ad workflows on a schedule or a threshold: weekly reviews, stop-loss alerts, creative fatigue checks and automatic pauses, across accounts. - [Turn a brief into a campaign you can inspect.](https://adrails.ai/features/campaign-builder): Turn a brief into editable campaigns, ad sets or ad groups and ads, with budgets, audiences, keywords and creative attached, ready to publish. - [One account today. More without another account fee.](https://adrails.ai/features/multi-account-ad-management): Group ad accounts by client, store or market and run campaigns, reports and automations across them without switching workspaces. - [Know what changed before deciding what to change.](https://adrails.ai/features/ad-reporting): Combine ad performance across platforms with commerce and attribution data, then ask an analysis agent what changed and why. - [Spot the creatives that need a closer look.](https://adrails.ai/features/creative-fatigue-monitoring): Track delivery pressure and performance change on every ad, then prepare a creative response instead of reacting to one metric alone. - [Meta Ads MCP server](https://adrails.ai/features/meta-ads-mcp): Connect Claude, ChatGPT or Claude Code to your Meta ad accounts through the Adrails MCP server. Read results, prepare launches and changes from the chat. - [Google Ads MCP server](https://adrails.ai/features/google-ads-mcp): Connect Claude, ChatGPT or Claude Code to Google Ads through the Adrails MCP server. Research keywords, build and edit campaigns from the chat. - [Compare variants before calling a winner.](https://adrails.ai/features/split-testing): Split testing for Meta and Google ads: change one variable, launch from one review step and read a winner at 95% confidence from your own conversions. - [Keep the campaign structure that works.](https://adrails.ai/features/campaign-templates): Campaign templates for Meta Ads and Google Ads: save the settings of a campaign, ad set or ad once, apply them in any account of the workspace, publish paused. - [Know where to keep, raise, cut or investigate.](https://adrails.ai/features/budget-recommendations): Ad budget recommendations from your last two complete weeks of figures: keep, raise, cut or investigate each budget, then apply a change capped at 30%. - [Send a report your client can read anywhere.](https://adrails.ai/features/client-reporting): Client reporting as a PDF built from Analysis: download it for the period and accounts you choose, or let an automation email it every week. - [Send conversion events from your site to your ads.](https://adrails.ai/features/conversion-tracking): Server-side tracking for ads: your server signs each sale, Adrails records it once and sends it to your ad accounts within a value limit and a daily limit. - [See how each creative performs across its ads.](https://adrails.ai/features/creative-analytics): Creative analytics for paid media: each image, video and text listed once with its spend, results, CPA, ROAS and signal across every ad that runs it. - [Build your keyword shortlist from Google’s data.](https://adrails.ai/features/keyword-research): Keyword research tool reading Keyword Planner through your Google Ads account: monthly searches and top-of-page bids by country, then a Search campaign. - [Pricing](https://adrails.ai/pricing): Plans sized by monthly action credits, with unlimited connected ad accounts. ## Platforms - [Platforms](https://adrails.ai/platforms): The ad platforms Adrails builds, tests and reports on. - [Meta Ads management](https://adrails.ai/platforms/meta-ads): Build Meta campaigns, run split tests, watch spend and report across ad accounts with AI agents and automations that run on a schedule. - [Google Ads management](https://adrails.ai/platforms/google-ads): Build Search, Performance Max, Display, Demand Gen and Shopping campaigns, test ads and read results beside your other accounts with AI agents. ## Who it is for - [Use cases](https://adrails.ai/use-cases): Workflows by the team or the business that runs the ads. - [AI workflows for paid media buyers](https://adrails.ai/use-cases/media-buyers): Build campaigns, investigate performance, run split tests and automate the checks you repeat every day across your ad accounts with AI agents. - [Run client ad accounts from one agent workspace](https://adrails.ai/use-cases/agencies): Standardize campaign creation, cross-account reporting and client reports for a paid media agency without flattening every client into one workflow. - [Run your ads with your store data as an ecommerce founder](https://adrails.ai/use-cases/ecommerce-founders): Connect your ad accounts to your store data, understand what changed, launch campaigns and let automations watch spend while you run the business. - [Give your growth team one paid media operating workflow](https://adrails.ai/use-cases/in-house-growth-teams): Connect analysis, campaign preparation, split tests and reporting so growth, creative and leadership work from the same paid media context. - [Turn ad performance into clearer creative briefs](https://adrails.ai/use-cases/creative-strategists): Connect creative analysis to live ad account context, identify useful patterns and turn findings into briefs the paid media team can act on. - [Paid ads for SaaS and B2B: run acquisition on CAC payback](https://adrails.ai/use-cases/saas): SaaS advertising measured on trials, demos and paying customers: read CAC payback and LTV to CAC next to ad spend, and automate on the result. - [Lead generation ads: optimize to qualified leads, not form fills](https://adrails.ai/use-cases/lead-generation): Run lead generation ads on qualified leads: send the leads sales accepts back to Meta and Google Ads, cap cost per lead by automation and share one report. - [App advertising measured on trials and renewals, not installs](https://adrails.ai/use-cases/mobile-apps): App advertising for subscription apps: RevenueCat trials, paying users and renewals read next to Meta and Google spend, with cost per paying user by campaign. - [Local business advertising: leads per location, not clicks](https://adrails.ai/use-cases/local-services): Local business advertising read as leads and cost per lead per location, with alerts per account, templates across locations and a PDF report per business. - [Dropshipping ads: run Meta campaigns on margin, not ROAS alone](https://adrails.ai/use-cases/dropshipping): Dropshipping ads run on margin: break-even ROAS per product, test budgets and kill rules, Shopify orders next to Meta spend, and stop-loss automations. - [Real estate advertising read in appointments, not clicks](https://adrails.ai/use-cases/real-estate): Real estate advertising measured in booked valuations, viewings and mandates: leads from your server, a cost per lead line per office, a report per client. - [Law firm advertising: case inquiries measured to the signed case](https://adrails.ai/use-cases/law-firms): Law firm advertising read in consultations booked and cases signed: intake events sent back to Meta and Google Ads, a cost per lead line per practice area. - [Dental advertising measured on booked patients](https://adrails.ai/use-cases/dental-practices): Dental advertising run on booked appointments and seated patients: conversions sent back within Meta and Google health data rules, a CPL line per practice. - [Home services advertising measured in booked jobs](https://adrails.ai/use-cases/home-services): Home services marketing measured in booked jobs: calls and forms recorded from your dispatch system, sent back as conversions with value, templates per area. - [Insurance advertising: quote requests to policies, by product line](https://adrails.ai/use-cases/insurance): Insurance advertising measured on bound policies: premiums sent back as conversions, CPL and spend thresholds per product line, one PDF report per agency. - [Restaurant advertising: reservations and orders, not reach](https://adrails.ai/use-cases/restaurants): Restaurant advertising read as reservations and orders per location: conversions from your server, one template per opening, spend alerts, a report per owner. - [Gym advertising: trial sign-ups and memberships per club, not clicks](https://adrails.ai/use-cases/gyms): Gym advertising read as trials booked and memberships started per club, with Stripe revenue next to spend, a cost per trial line and a report per franchisee. ## AI clients and business sources - [Integrations](https://adrails.ai/integration): The AI clients that work with Adrails, and the sources it reads. - [Meta Ads and Google Ads in Claude](https://adrails.ai/integration/claude): Add the Adrails MCP server to Claude on claude.ai or the desktop app. Read Meta Ads and Google Ads results, prepare launches and changes from the chat. - [Meta Ads and Google Ads in ChatGPT](https://adrails.ai/integration/chatgpt): Add the Adrails MCP server to ChatGPT with Developer mode. Read Meta Ads and Google Ads results, prepare launches and changes from the chat, with OAuth. - [Meta Ads and Google Ads in Claude Code](https://adrails.ai/integration/claude-code): Add the Adrails MCP server to Claude Code with one command. Read Meta Ads and Google Ads results and prepare launches and changes from the terminal. - [Meta Ads and Google Ads in Cursor](https://adrails.ai/integration/cursor): Add the Adrails MCP server to Cursor through mcp.json. Read Meta Ads and Google Ads results and prepare launches and changes from the editor, with OAuth. - [Connect Meta Ads accounts to Adrails](https://adrails.ai/integration/meta-ads): Connect Meta ad accounts: which profiles and accounts enter the workspace, what Adrails reads, what it can write and how to remove access. - [Connect Shopify orders, products and store performance to Adrails](https://adrails.ai/integration/shopify): Use Shopify products, orders, customers and store analytics as business context for paid media analysis in Adrails. - [Connect Klaviyo lifecycle performance to paid media analysis](https://adrails.ai/integration/klaviyo): Bring Klaviyo email, SMS, flow and attributed revenue metrics into the same analysis workspace as advertising spend. - [Use Triple Whale attribution and profit context in Adrails](https://adrails.ai/integration/triple-whale): Connect Triple Whale summary metrics to analyze paid media with blended revenue, attribution and profit context. - [Connect Stripe payments and subscriptions to Adrails](https://adrails.ai/integration/stripe): Bring Stripe charges, refunds and subscription context into paid media analysis with restricted read access. - [Connect PostHog product events to paid acquisition analysis](https://adrails.ai/integration/posthog): Use daily PostHog product event counts as downstream context for paid media analysis in Adrails. ## Connect - [Connect](https://adrails.ai/connect): Each source Adrails reads, connected to one AI client, and the vendors' own MCP servers compared. - [Shopify MCP: Shopify's own servers, and store revenue next to ad spend](https://adrails.ai/connect/shopify-mcp): Shopify MCP explained: Shopify's connector for Claude and ChatGPT, its Dev and Storefront servers, and how Adrails puts store revenue next to ad spend. - [Stripe MCP: Stripe's own server, and subscriptions next to ad spend](https://adrails.ai/connect/stripe-mcp): Stripe MCP explained: the server at mcp.stripe.com, its sign-in, tools and clients, and how Adrails puts payments, MRR and subscriptions next to ad spend. - [PostHog MCP: PostHog's server, and product events next to ad spend](https://adrails.ai/connect/posthog-mcp): PostHog MCP explained: what PostHog's own server at mcp.posthog.com reads and writes, and how Adrails sets daily product events next to ad spend by campaign. - [Google Drive MCP: Google's server, and Drive files as ad creatives](https://adrails.ai/connect/google-drive-mcp): Google Drive MCP explained: what Google's server at drivemcp.googleapis.com reads and writes, and how Adrails turns shared Drive folders into paused ads. - [RevenueCat MCP: RevenueCat's server, and app revenue next to ad spend](https://adrails.ai/connect/revenuecat-mcp): RevenueCat MCP: what RevenueCat's server at mcp.revenuecat.ai reads and writes, and how Adrails sets MRR and subscriptions next to ad spend by campaign. - [Klaviyo MCP: Klaviyo's own server, and email revenue next to ad spend](https://adrails.ai/connect/klaviyo-mcp): Klaviyo MCP explained: the server at mcp.klaviyo.com, its tools, sign-in, roles and read-only mode, and how Adrails puts email revenue next to ad spend. - [Triple Whale MCP: the official server, and net profit next to ad spend](https://adrails.ai/connect/triple-whale-mcp): Triple Whale MCP explained: what Triple Whale's read-only server at mcp.triplewhale.com answers through Moby, and how Adrails sets net profit next to ad spend. - Connect Meta Ads to [Claude](https://adrails.ai/connect/meta-ads-to-claude), [ChatGPT](https://adrails.ai/connect/meta-ads-to-chatgpt), [Claude Code](https://adrails.ai/connect/meta-ads-to-claude-code), [Cursor](https://adrails.ai/connect/meta-ads-to-cursor): Add the Adrails MCP server to Claude and read Meta Ads spend, results and structure, then prepare launches and changes you apply in Adrails. - Connect Google Ads to [Claude](https://adrails.ai/connect/google-ads-to-claude), [ChatGPT](https://adrails.ai/connect/google-ads-to-chatgpt), [Claude Code](https://adrails.ai/connect/google-ads-to-claude-code), [Cursor](https://adrails.ai/connect/google-ads-to-cursor): Add the Adrails MCP server to Claude and read Google Ads spend, structure and keywords, then prepare launches and changes you apply in Adrails. - Connect Shopify to [Claude](https://adrails.ai/connect/shopify-to-claude), [ChatGPT](https://adrails.ai/connect/shopify-to-chatgpt), [Claude Code](https://adrails.ai/connect/shopify-to-claude-code), [Cursor](https://adrails.ai/connect/shopify-to-cursor): Add the Adrails MCP server to Claude and ask for Shopify orders, revenue and products next to your ad spend. Read only, OAuth sign-in, no API key. - Connect Stripe to [Claude](https://adrails.ai/connect/stripe-to-claude), [ChatGPT](https://adrails.ai/connect/stripe-to-chatgpt), [Claude Code](https://adrails.ai/connect/stripe-to-claude-code): Add the Adrails MCP server to Claude and ask for Stripe subscriptions, MRR and payments next to your ad spend. Read only, OAuth sign-in, no API key. - Connect Klaviyo to [Claude](https://adrails.ai/connect/klaviyo-to-claude), [ChatGPT](https://adrails.ai/connect/klaviyo-to-chatgpt), [Claude Code](https://adrails.ai/connect/klaviyo-to-claude-code): Add the Adrails MCP server to Claude and ask for Klaviyo email and SMS results next to your ad spend. Read only, OAuth sign-in, no API key. ## Blog - [Blog](https://adrails.ai/blog): Paid media guides with sourced figures, newest first. - [Google Ads scripts vs MCP and AI agents: what to use](https://adrails.ai/blog/google-ads-scripts-vs-mcp): Compare Google Ads scripts, MCP tools and AI agents for reporting, budget monitoring and campaign changes, with a worked workflow and review checklist. - [Paid media client reporting template for Meta and Google Ads](https://adrails.ai/blog/client-reporting-template): Download a reusable paid media client report template with metric definitions, a worked example, action owners and a checklist for Meta and Google Ads. - [Adrails vs Pipeboard vs Adspirer: the MCP servers compared](https://adrails.ai/blog/adrails-vs-pipeboard-vs-adspirer): Compare Adrails, Pipeboard and Adspirer MCP servers by ad platforms, account access, campaign changes and review controls, using their own documentation. - [ChatGPT for Meta Ads: what works through MCP](https://adrails.ai/blog/chatgpt-for-meta-ads): Use ChatGPT for Meta Ads through MCP: check plan access, read campaign results, prepare changes and understand what still needs review in Adrails. - [How to run Google Ads from Claude](https://adrails.ai/blog/run-google-ads-from-claude): Use Claude for Google Ads: connect the account, ask for results, keyword research and new campaigns, and see what Claude can change and what it only reads. - [Claude connectors for marketing teams](https://adrails.ai/blog/claude-connectors-for-marketing): The Claude connectors a marketing team should add for ad accounts, revenue, email, CRM, SEO and creative, what each reads or changes, and what to check first. - [How much do Facebook ads cost in 2026?](https://adrails.ai/blog/facebook-ads-cost): How much Facebook ads cost in 2026: CPM, CPC, CTR and cost per lead benchmarks from LocaliQ, Databox, Gupta Media and Lebesgue, and the budget a test needs. - [How much do Google Ads cost in 2026?](https://adrails.ai/blog/google-ads-cost): How much Google Ads cost in 2026: average cost per click, CTR, conversion rate and cost per lead by industry from LocaliQ, and how to set a first budget. - [Facebook Ads vs Google Ads: which to run first](https://adrails.ai/blog/facebook-ads-vs-google-ads): Facebook Ads vs Google Ads compared on intent, targeting, formats, 2026 costs, time to signal and measurement, with a rule for which to run first by business. - [Meta Andromeda: what changed in ad retrieval, and what advertisers do about it](https://adrails.ai/blog/meta-andromeda): Meta Andromeda explained from Meta's own posts: what the ad retrieval engine changed, what Meta recommends on creative and structure, and what to check weekly. - [Facebook lead ads: instant forms vs landing pages](https://adrails.ai/blog/facebook-lead-ads): How Facebook lead ads work, when Meta instant forms beat a landing page, why cost per lead misleads, and how to send qualified leads back to Meta. - [Google Ads bidding strategies: which one, when, and when to switch](https://adrails.ai/blog/google-ads-bidding-strategies): Google Ads bidding strategies compared: what each optimizes, the conversion data Google asks for, Target CPA vs Target ROAS, and a protocol for switching. - [CAC payback and LTV:CAC for SaaS paid acquisition](https://adrails.ai/blog/saas-cac-payback): The CAC payback period formula, the LTV:CAC ratio, SaaS CAC benchmarks with their sources, and how to run paid acquisition on payback from Stripe revenue. - [App install campaigns on Meta and Google: CPI and what comes after](https://adrails.ai/blog/app-install-campaigns): How app install campaigns work on Meta and Google Ads, CPI benchmarks with sources, measurement after ATT, and why app advertising is judged on paying users. - [Marketing mix modeling for small ad teams](https://adrails.ai/blog/marketing-mix-modeling): Marketing mix modeling for small ad teams: what MMM answers next to attribution and incrementality testing, and when Meridian or Robyn is worth the effort. - [Dropshipping ads, margin first](https://adrails.ai/blog/dropshipping-ads): How to run dropshipping ads on Meta without losing money: break-even ROAS from real costs, a Facebook ads budget by test stage, and the kill rule. - [Real estate Facebook ads under the housing special ad category](https://adrails.ai/blog/facebook-ads-for-real-estate): Real estate Facebook ads under Meta's housing special ad category: the targeting you lose, what works for listings and valuations, and the 2026 cost per lead. - [PPC for lawyers: Google Ads for law firms when a lead costs $132](https://adrails.ai/blog/ppc-for-lawyers): PPC for lawyers: what a legal click costs, match types and negatives for legal intent, Search against Local Services Ads, and the budget 20 inquiries need. - [Google Ads for dentists: cost per new patient, not cost per click](https://adrails.ai/blog/google-ads-for-dentists): Google Ads for dentists: what a dental click and lead cost in LocaliQ's 2026 benchmarks, radius targeting, health policy, call assets and Local Services Ads. - [Local Services Ads vs Google Ads: pay per lead or per click](https://adrails.ai/blog/local-services-ads-vs-google-ads): Local Services Ads vs Google Ads for home services: how Google prices a lead against a click, Google Verified, ranking, control and which to run first. - [Insurance leads from your own ads](https://adrails.ai/blog/insurance-leads-from-ads): Generate insurance leads with your own Meta and Google ads instead of buying shared leads: 2026 costs, the rules by product line and cost per bound policy. - [Restaurant marketing with paid ads: a weekly plan](https://adrails.ai/blog/restaurant-marketing-with-ads): Restaurant marketing with paid ads: a four-week plan with the objective, radius, dayparting, offer and budget per location, from Meta's and Google's own rules. - [Gym ads on Meta: trial offers, lead forms and the cost per new member](https://adrails.ai/blog/gym-ads-on-meta): Gym Facebook ads: the trial offer, instant forms against a booking page, radius, creative fatigue, Meta's health data rules and the cost per new member. - [ChatGPT as an ads manager: what it does alone, with connectors, and where it stops](https://adrails.ai/blog/chatgpt-ads-manager): What a ChatGPT ads manager does: exports and copy alone, what ChatGPT connectors and Developer mode add, and how a change reaches your ad accounts. - [Meta Ads MCP: Meta's official connector and Adrails compared](https://adrails.ai/blog/meta-ads-mcp-official-connector-vs-adrails): What Meta's MCP server at mcp.facebook.com/ads does with its 29 tools, where it stops, and what the Adrails MCP server adds on Meta Ads and Google Ads. - [Google Ads MCP server: official vs hosted, reads and actions](https://adrails.ai/blog/google-ads-mcp-server-official-vs-hosted): Google's open source Google Ads MCP server is read-only and runs where you deploy it. What it does, what it needs, and what a hosted server like Adrails adds. - [Google Ads AI: what Google's own AI does, and what it leaves to you](https://adrails.ai/blog/google-ads-ai): Google Ads AI explained: Performance Max, AI Max for Search, Smart Bidding and text customization, what each decides, and where an AI agent fits. - [MER: the marketing efficiency ratio, explained](https://adrails.ai/blog/marketing-efficiency-ratio): MER in marketing is revenue divided by all marketing cost. How it differs from ROAS and blended ROAS, and how to set a MER target from your margin. - [Target CPA: calculate it, then set it in Google Ads and Meta](https://adrails.ai/blog/target-cpa): How to calculate a target CPA from margin and customer value, then how Target CPA bidding in Google Ads and the cost per result goal in Meta use it. - [Claude for marketing: what it can do for paid media](https://adrails.ai/blog/claude-for-marketing): How to use Claude for marketing and paid media: what it does alone, what needs a connector for live ad data, prompts to copy, and how to act on live ad data. - [Build a Google Ads Search campaign from your Shopify store with Claude Code](https://adrails.ai/blog/claude-code-google-ads-search-from-shopify): Read your Shopify catalogue, check Keyword Planner volumes and build a paused Google Ads Search campaign from Claude Code, with the prompts, step by step. - [A weekly paid media review in Claude that pauses what crosses your threshold](https://adrails.ai/blog/weekly-paid-media-review-claude): Run a weekly paid media review in Claude with exact date windows, pause what crosses your threshold once you confirm, then make the check hourly in Adrails. - [Launch a split test from one sentence in ChatGPT](https://adrails.ai/blog/split-test-from-one-sentence-chatgpt): Ask ChatGPT for a split test in one sentence: Adrails prepares the variants, you confirm, they are created paused, and a winner is named at 95% confidence. - [Pause the ads of out-of-stock products, from the chat](https://adrails.ai/blog/pause-ads-of-out-of-stock-products): Read Shopify inventory, find the ads that still sell each sold-out product and pause them once you confirm, from Claude or ChatGPT through Adrails. - [CBO vs ABO: choosing where the Meta Ads budget sits](https://adrails.ai/blog/cbo-vs-abo): ABO vs CBO explained: when ad set budgets beat Advantage+ campaign budget in Meta Ads, with a decision table for testing, scaling and markets. - [What is a good ROAS?](https://adrails.ai/blog/what-is-a-good-roas): ROAS meaning, the ROAS formula and what a good ROAS is for your business, derived from contribution margin and break-even ROAS, with a table by margin. - [Google Ads negative keywords](https://adrails.ai/blog/google-ads-negative-keywords): How negative keywords work in Google Ads, a starter negative keyword list, match types, and where to add them: campaign, account, lists and Performance Max. - [How to connect Meta Ads and Google Ads to Claude and ChatGPT](https://adrails.ai/blog/connect-meta-and-google-ads-to-claude-chatgpt): Connect Claude, Claude Code, ChatGPT or Codex to your ad accounts through the Adrails MCP server, step by step, then work on them from the chat. - [How to manage multiple Meta ad accounts without losing control](https://adrails.ai/blog/manage-multiple-meta-ad-accounts): A practical system for account grouping, naming, reporting, permissions, review and recurring Meta Ads operations across clients or markets. - [How to automate Meta Ads in stages, from alerts to actions](https://adrails.ai/blog/automate-meta-ads-in-stages): Automate Meta Ads in stages: alerts first, then automatic pauses within a ceiling, then agents that prepare budget and campaign changes. - [Meta Ads account audit checklist: 20 checks and next actions](https://adrails.ai/blog/meta-ads-account-audit-checklist): Audit your Meta Ads account with 20 checks across tracking, delivery, creative and operations. Learn what to inspect and how to prioritize fixes. - [Meta Ads campaign naming conventions that scale](https://adrails.ai/blog/meta-ads-campaign-naming-conventions): Build a readable naming system for Meta campaigns, ad sets and ads across clients, markets, objectives and creative tests. - [Ad fatigue: how to detect creative fatigue in Meta Ads](https://adrails.ai/blog/detect-meta-ads-creative-fatigue): What ad fatigue and creative fatigue are, the signs in frequency, CTR, CPA and conversion rate, and how to fix a fatigued Facebook ad without guessing. - [Budget pacing for Meta and Google Ads](https://adrails.ai/blog/meta-ads-budget-pacing): Calculate budget pacing for Meta and Google Ads. Compare spend with plan, work out the daily spend left, and check accounts without mixing currencies. - [Meta Ads automated rules vs AI agents](https://adrails.ai/blog/meta-ads-automated-rules-vs-ai-agents): Understand when Meta Ads work belongs in a deterministic rule, a scheduled workflow or an AI agent, and how the three work together. - [Why Meta Ads and Shopify revenue do not match](https://adrails.ai/blog/meta-vs-shopify-revenue): Understand attribution windows, identity, time zones, refunds and conversion definitions behind Meta Ads and Shopify revenue differences. - [How to structure Meta Ads campaigns across multiple markets](https://adrails.ai/blog/meta-ads-campaign-structure-multiple-markets): Structure Meta Ads campaigns across countries and regions while preserving budget control, localization and readable reporting. - [Learning Limited in Facebook and Meta ads: causes and fixes](https://adrails.ai/blog/meta-ads-learning-limited): Why a Facebook ad set shows Learning Limited, how the Meta learning phase works, and fixes that add signal instead of fragmenting the account further. ## Buyer guides - [Buyer guides](https://adrails.ai/best): Paid media software compared by the job each product does. - [Best MCP servers for Meta Ads and Google Ads in 2026](https://adrails.ai/best/mcp-servers-for-meta-and-google-ads): Six MCP servers that connect Claude, ChatGPT and Claude Code to ad accounts, compared by platforms, hosting, writes and price: official and hosted. - [Best AI tools for Meta Ads in 2026](https://adrails.ai/best/ai-meta-ads-tools): Compare AI tools for Meta Ads by the work they perform: campaign creation, optimization, creative analysis, rules and AI agents. - [Best Meta Ads management tools for agencies in 2026](https://adrails.ai/best/meta-ads-management-tools-for-agencies): Compare Meta Ads tools for agencies managing multiple clients, campaign creation, automation rules, split tests and client reporting. - [Best Facebook Ads automation software in 2026](https://adrails.ai/best/facebook-ads-automation-software): Compare Facebook and Meta Ads automation software across native rules, external rule engines, AI optimization suites and AI agents. ## Free tools - [Free tools](https://adrails.ai/tools): Calculators and checkers that run in the browser, with no account connection. - [Free UTM builder](https://adrails.ai/tools/utm-builder): Build a tagged campaign URL for Meta Ads, Google Ads or email. Keep existing URL parameters, encode campaign names and copy the result. No signup needed. - [Google Ads budget calculator](https://adrails.ai/tools/google-ads-budget-calculator): Convert a monthly Google Ads budget into an average daily budget. Estimate clicks, conversions and CPA from your own CPC and conversion rate assumptions. - [Free Meta Ads audit](https://adrails.ai/tools/meta-ads-audit): Run a free Meta Ads audit checklist for tracking, delivery and creative. See example findings from Adrails rules, then connect your account for a review. - [Meta Ads spend pacing calculator](https://adrails.ai/tools/meta-ads-spend-pacing-calculator): Calculate expected spend, pacing percentage and the daily budget required to finish a Meta Ads budget period on target. - [Ad fatigue calculator](https://adrails.ai/tools/creative-fatigue-calculator): Check ad fatigue on Facebook and Instagram: compare frequency, CTR and CPA with an earlier period to spot a creative fatigue pattern. - [Meta Ads campaign naming convention generator](https://adrails.ai/tools/campaign-naming-convention-generator): Generate consistent names for Meta campaigns, ad sets and ads using brand, market, objective, audience and creative tokens. - [ROAS calculator](https://adrails.ai/tools/meta-ads-roas-calculator): Calculate ROAS with the ROAS formula, revenue divided by ad spend, plus CPA and revenue per conversion. Works for Google Ads, Meta Ads or any channel. - [Break-even ROAS calculator](https://adrails.ai/tools/break-even-roas-calculator): Calculate break-even ROAS with the formula AOV ÷ contribution per order, or 1 ÷ margin, plus maximum CPA and contribution margin. - [Target CPA calculator for Meta Ads](https://adrails.ai/tools/target-cpa-calculator): Calculate target CPA, target CPC, required traffic and projected ROAS from an ad budget and acquisition goal. - [Meta Ads CPM, CPC, CTR and CPA calculator](https://adrails.ai/tools/meta-ads-metrics-calculator): Calculate CPM, CPC, CTR, CPA and click-to-conversion rate from Meta Ads spend, impressions, clicks and conversions. - [Facebook Ads cost and budget calculator](https://adrails.ai/tools/meta-ads-budget-calculator): Estimate what Facebook and Instagram ads will cost for a revenue goal: total budget, daily spend, orders, clicks and the CPC the plan needs. - [MER and blended ROAS calculator](https://adrails.ai/tools/mer-calculator): Calculate MER, the marketing efficiency ratio, and blended ROAS from total revenue and spend across Meta, Google and every other channel. - [Free Google Ads audit](https://adrails.ai/tools/google-ads-audit): Audit your Google Ads account with a free checklist, no login required. Review tracking, bidding, search terms and Performance Max, then prioritize fixes. ## Documentation - [Documentation](https://adrails.ai/docs): Product guides, integrations, the MCP server and the Tracking API. - [MCP server](https://adrails.ai/docs/mcp): Connection, permissions, tokens, revocation and endpoints. - [Connect an AI client](https://adrails.ai/docs/mcp/connect): Claude, Claude Code, ChatGPT and Codex, step by step. - [MCP tools reference](https://adrails.ai/docs/mcp/tools): Every tool, its inputs and who can call it. - [MCP security](https://adrails.ai/docs/mcp/security): What a connection can reach, and what Adrails records. - [Set up tracking](https://adrails.ai/docs/tracking/setup): The script, the two keys, and how a sale or lead is recorded once. - [Send conversions to the platforms](https://adrails.ai/docs/tracking/forwarding): Meta dataset, Google Ads conversion action, value and daily limits. - [Split tests](https://adrails.ai/docs/guides/split-tests): What can vary, how a test is launched, followed and ended. - [Automations](https://adrails.ai/docs/guides/automations): Scheduled and threshold workflows, templates and credits. - [Client reports](https://adrails.ai/docs/guides/client-reports): PDF reports from Analysis or an automation, and what they hold. - [Plans and billing](https://adrails.ai/docs/guides/plans-and-billing): What each plan includes, Preview, changing and cancelling. ## Company - [Adrails brings your ad operations into one workspace.](https://adrails.ai/about): Adrails is ad management with AI agents for paid media teams and agencies, built and operated by TBSS Labs, LLC. - [Contact Adrails: write to the founder](https://adrails.ai/contact): Contact Adrails at thomas@adrails.ai for product questions, partnerships and press. Find the documentation, security information and MCP repository. - [Security boundaries for connected advertising data](https://adrails.ai/security): How Adrails handles provider access, stored credentials, account writes, roles and the removal of a connection when you disconnect a provider. - [How Adrails researches comparisons, guides and benchmarks](https://adrails.ai/methodology): The editorial standards used for Adrails product comparisons, operational guides, calculations and future aggregate benchmarks. - [Privacy](https://adrails.ai/privacy): Privacy policy. - [Terms](https://adrails.ai/terms): Terms of service. ## Optional - [Comparisons](https://adrails.ai/concurrent): Adrails next to Meta Ads Manager and the tools paid media teams already use, with primary sources and a review date. - [Tool stacks](https://adrails.ai/stack): Where ad spy, research, creative and landing page tools stop and the launch takes over. - [Full text](https://adrails.ai/llms-full.txt): Every public page of this site as Markdown, for a model that needs the content rather than the index. - [Sitemap](https://adrails.ai/sitemap.xml): Every indexable URL with its last modification date. - [RSS](https://adrails.ai/blog/feed.xml): The blog as RSS 2.0.