On this page7
Build a live cross-channel marketing dashboard in Claude by connecting your accounts once inside Otto via OAuth, adding Otto's hosted HTTP MCP server at https://otto.islaintel.com/api/mcp, and supplying one organization-scoped otto_mcp__ bearer token in your AI client. Across Otto MCP's 227 tools across 15 integrations, Claude queries Google Analytics 4, Search Console, Meta Ads, Google Ads, LinkedIn Ads, Shopify, Google Sheets, and PageSpeed Insights in one chatβcutting token usage by 80β90% compared to 3,000-row CSV uploads while keeping analytics read-only and website or ad edits behind human approval.
Copy-paste Claude prompts to launch your cross-channel marketing dashboard (start here)
Five copy-paste prompts below give you live visibility across revenue, search rankings, ad spend, and landing-page bottlenecks the moment you connect Claude to Otto MCP.
Prompt 1: The 60-second executive cross-channel scorecard
Query my connected accounts through Otto MCP and generate an Executive Cross-Channel Scorecard for the last 7 days compared to the previous 7 days. Pull:
1. Total sessions, engaged sessions, and key conversion events from GA4.
2. Total clicks, impressions, and average CTR from Google Search Console.
3. Total ad spend, impressions, clicks, and reported conversions across Meta Ads, Google Ads, and LinkedIn Ads.
Calculate blended CPA and blended ROAS, flag any metric that changed by more than 15%, and list the top three drivers behind the changes.
Prompt 2: The revenue and KPI target pacing tracker
Check my live e-commerce performance and budget pacing for this month to date:
1. Pull live Shopify orders, gross sales, average order value (AOV), and top five selling products by revenue.
2. Call spreadsheets.values.batchGet on my connected Google Sheet "Marketing KPI Targets" (Tab: "2027 Goals", Range: "A1:E12") to retrieve monthly target revenue, target blended ROAS, and maximum allowable CAC.
3. Compare actual Shopify sales and paid ad spend against the pacing targets for today's day of the month.
Present a table showing Metric, Actual MTD, Target MTD, Pacing %, and Projected Month-End.
Prompt 3: The organic search and AI visibility panel
Audit my organic search performance and technical hygiene over the last 28 days:
1. Run searchAnalytics.query via Search Console to extract striking-distance queries (ranking between positions 8 and 20) with at least 100 impressions and a CTR below 2.5%.
2. Group these queries by target landing page URL.
3. For the top three striking-distance URLs, run psi_analyze_url to retrieve mobile Core Web Vitals (LCP, INP, CLS) and mobile performance scores.
Output a prioritized table of on-page and technical fixes to lift these pages onto page one.
Prompt 4: The paid media CPA and attribution drift audit
Reconcile my paid ad channel reporting against GA4 over the last 14 days:
1. Pull reported conversions, spend, and cost per conversion directly from Meta Ads, Google Ads, and LinkedIn Ads.
2. Run runReport on GA4 to retrieve sessions and key conversion events attributed to paid search, paid social, and cross-network campaigns.
3. Compare in-platform conversion numbers against GA4 first-touch and session-based conversions.
Identify campaigns suffering from attribution drift, double-counting, or rising CPA, and highlight any ad set with zero conversions over $200 spent.
Prompt 5: The action stage prompt
Based on the landing-page drop-offs and poor-performing ads identified in the dashboard above:
1. Propose exact headline and copy revisions for the underperforming landing page to match the highest-intent Search Console query.
2. Draft the update directly to my connected website as a WordPress draft (or GitHub pull request preview).
3. If any Meta or LinkedIn ad creative has a CPA more than 30% above our target, draft a paused replacement creative with new body copy and staged headlines.
Do not publish anything live or alter active campaign budgets; wait for my explicit confirmation on all staged drafts.Static CSV Artifact dashboards vs. single-channel local MCP servers vs. Claude + Otto MCP
Teams building dashboards in Claude often upload static CSV exports or maintain separate local connectors. Uploading a raw 3,000-row Ads Manager CSV can consume 40,000 to 80,000 tokens in a single Claude turn, whereas querying aggregated metrics through Otto MCP cuts token usage by 80β90%. Google's official Google Analytics MCP server is strictly for read-only GA4 report queries and cannot modify Google Analytics settings, so it stays isolated from your ads, CMS, and Shopify data in one Claude chat.
The following table contrasts these approaches against Otto MCP's 227 tools across 15 integrations.
| Feature | Static CSV Uploads in Claude Artifacts | DIY Single-Channel Local MCP Servers | Claude + Otto MCP Server |
|---|---|---|---|
| Setup workflow | Repeated manual exports each reporting cycle (illustrative comparison) | Per-device local MCP install; Google's GA4 MCP server handles read-only report queries with local credentials | One hosted HTTP endpoint (https://otto.islaintel.com/api/mcp), OAuth once inside Otto, and one otto_mcp__ bearer token in Claude |
| Token consumption per query | High (40,000 to 80,000 tokens per 3,000-row CSV upload) | Low to moderate (queries single local tools individually) | 80β90% lower token usage via server-side aggregation |
| Cross-channel synthesis | Manual (requires uploading 4β6 separate spreadsheets) | None (isolated local tools cannot cross-reference ads, CMS, and GA4) | Automatic across GA4, Search Console, paid ads, Shopify, and Sheets |
| Data freshness | Stale the moment the CSV file is exported | Live on the machine where local credentials reside | Live in Claude Web, Desktop, Claude Code, Cursor, ChatGPT, OpenClaw, and Hermes via remote HTTP MCP |
| Secret and credential management | Manual token and CSV handling on personal machines | Local environment files and Application Default Credentials per device | Centralized OAuth inside Otto; raw API secrets never leave the server |
| Action capability | Read-only; cannot push changes or fixes | Read-only; cannot update sites or stage ads | Draft-first WordPress updates, GitHub pull requests, and paused ad staging |
Connect Claude to Otto MCP to query your live marketing channels in one conversation.
Otto MCP
Ask your AI assistant what changed in your ads
Connect Claude or ChatGPT to Otto, then ask about Meta Ads, Google Ads, GA4, and Search Console in the same chat. The logins stay in Otto.
How to connect your cross-channel marketing dashboard in Claude
Setting up your dashboard takes four steps after you connect accounts in Otto and add the hosted MCP endpoint. Unlike local MCP servers that require installation on each device, remote servers are available from any internet-connected MCP clientβthe pattern Otto's HTTP endpoint follows at https://otto.islaintel.com/api/mcp. You connect accounts once inside Otto via OAuth; the AI client holds only your organization-scoped otto_mcp__ bearer token while OAuth secrets never leave Otto's server.
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Claude Client β
β (Claude Web / Desktop / Code / Cursor) β
ββββββββββββββββββββββββββββββββ¬βββββββββββββββββββββββββββββββ
β HTTPS + Bearer otto_mcp__
βΌ
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Otto Hosted Remote MCP Server β
β https://otto.islaintel.com/api/mcp β
β (227 Tools, OAuth Handling) β
βββββββββ¬βββββββββββββββ¬βββββββββββββββ¬βββββββββββββββ¬βββββββββ
β β β β
βΌ βΌ βΌ βΌ
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Analytics ββ Paid Media ββ E-Commerce ββ Execution β
β GA4 Data API ββ Meta Ads ββ Shopify ββ WordPress β
β GSC Query ββ Google Ads ββ Sheets API ββ GitHub PR β
β PageSpeed ββ LinkedIn ββ Target Rows ββ Vercel β
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
- Connect your accounts once inside Otto. Log into your Otto dashboard and authorize Google Analytics 4, Google Search Console, Meta Ads, Google Ads, LinkedIn Ads, X Ads, ChatGPT Ads, Shopify, Google Sheets, WordPress, GitHub, Vercel, Google Business Profile, Facebook Pages, Instagram, and YouTube. OAuth secrets remain securely encrypted on Otto's servers; you never paste raw private API keys into your Claude chat prompts.
- Copy your organization-level MCP token. Navigate to the MCP settings panel in Otto to copy your hosted endpoint URL (
https://otto.islaintel.com/api/mcp) and your personal bearer token prefixed withotto_mcp__. - Add Otto as a remote connector in Claude. Follow the How to Connect Claude to Otto MCP setup guide to install the connector. In Claude Desktop, Claude Code, or Cursor, add the remote endpoint to your configuration: ``
json { "mcpServers": { "otto": { "url": "https://otto.islaintel.com/api/mcp", "headers": { "Authorization": "Bearer otto_mcp_your_token_here" } } } }`` - Run your first cross-channel query. In your Claude chat, ask: "Check my connected Otto accounts and confirm which tools are active." Claude will confirm access across your connected analytics, paid media, Shopify store, and CMS environments.
Under the hood, Otto handles the complex parameters required by provider APIs. When Claude requests analytics data, Otto calls the Google Analytics Data API v1 runReport method using your property ID, specified date ranges, dimensions, and metrics. For organic search performance, Otto invokes Google Search Console searchAnalytics.query API with the requested date filters and row groupings. For target tracking, Otto calls the Google Sheets API v4 spreadsheets.values.batchGet method to retrieve cell ranges. By aggregating these metrics before returning clean JSON tables to Claude, Otto cuts token consumption by 80β90% compared to dumping raw CSV spreadsheets into your context window.
Dashboard views 1β4: GA4 funnels, Search Console queries, paid media ROAS, and Shopify + Google Sheets KPIs
Dashboard views 1 through 4 focus on traffic acquisition, conversion funnel health, paid media attribution, and e-commerce cash flow.
View 1: GA4 traffic, landing pages, and funnel drop-offs
Evaluating engagement across marketing channels requires structured reporting on landing pages, active sessions, and key conversion events. The official Google Analytics Data API v1 runReport method generates filtered, paginated dimension and metric tables for any connected GA4 property. Otto queries these tables directly so you can spot drop-offs without exporting manual reports from the GA4 interface.
Run a GA4 funnel analysis for the last 30 days. Pull landingPagePlusQueryString, sessionDefaultChannelGroup, sessions, engagedSessions, bounceRate, and keyEvents. Filter for landing pages with more than 100 sessions, calculate session-to-key-event conversion rates per page, and highlight any landing page where engagement rate dropped below 40%.
This view surfaces pages driving traffic that fail to generate actions. You can connect your property through Connect Google Analytics 4 (GA4) to Otto and view automated traffic trends with Otto Website Analytics and weekly briefings.
View 2: Search Console striking-distance queries and CTR audit
Keyword rankings are valuable when you identify queries where small improvements yield high traffic gains. According to the Google Search Console searchAnalytics.query documentation, the API retrieves search traffic rows filtered by date, page, device, and country, grouped by dimensions you define.
Query Search Console over the past 28 days using searchAnalytics.query. Return the top 25 queries ranking between position 7.0 and 19.0, sorted by total impressions descending. Display query, clicks, impressions, CTR, and average position. Highlight queries with over 200 impressions where CTR is below 2.0%, and identify the landing page URL serving each query.
Marketers use this view to discover quick-win keywords that can be lifted to page one with title tag optimizations and internal links. Read Connect Google Search Console to Otto to connect your domain, and review Otto AI SEO Agent for automated keyword tracking.
View 3: Paid media spend, CPA, and attribution reconciliation
Paid advertising reports often suffer from platform self-attribution. Ad platforms report conversions that do not match back-end analytics. Guillermo Marti, Head of Growth & Marketing at IslaIntel, structures cross-channel ad reviews by contrasting platform-reported conversions from Meta Ads, read-only Google Ads, and LinkedIn Ads against GA4's unified conversion events.
Compare paid media spend and performance across Meta Ads, Google Ads, and LinkedIn Ads for the last 14 days against GA4 key events. For each platform, display spend, clicks, platform-reported conversions, and platform CPA. Then pull GA4 conversions attributed to each channel. Calculate the discrepancy percentage between platform conversions and GA4 conversions, and identify any ad campaign where blended acquisition cost exceeds $65.
This review eliminates double-counting across networks. To inspect ad efficiency and diagnose audience fatigue, review Otto AI Paid Ads Manager, explore GA4 and paid media landing page dashboards, and reference 10 Claude workflows for Meta Ads in 2027.
View 4: Live Shopify orders, top products, and Google Sheets KPI target pacing
Revenue metrics need to be compared against operational targets. IslaIntel Chief Operating Officer Enrique VΓ‘squez tracks performance by reconciling live Shopify orders and inventory with spreadsheet budgets. Under the hood, the Google Sheets API v4 spreadsheets.values.batchGet method pulls one or more cell ranges, allowing Claude to evaluate real-time numbers against target projections.
Query my connected Shopify store for orders placed over the last 7 days, including gross sales, refunds, net sales, and units sold for the top five SKUs. Then call spreadsheets.values.batchGet on my connected Google Sheet "Executive Operations" to retrieve monthly sales targets from tab "Q4 Targets", range "B2:F15". Reconcile actual Shopify net sales against target pacing for the current date, and flag whether revenue is pacing ahead or behind budget.
To link your store and data sheets, see Connect Shopify to Otto and Connect Google Sheets to Otto.
Dashboard views 5β8: Core Web Vitals, organic socials & GBP, staged landing-page fixes, and the Monday executive digest
Views five through eight expand beyond reporting into technical diagnostics, local visibility, staged website edits, and automated executive governance.
View 5: Mobile Core Web Vitals and Lighthouse diagnostics
A fast landing page is essential for conversion rates and ad Quality Scores. Otto MCP includes a built-in psi_analyze_url tool that runs Google Lighthouse audits and checks Core Web Vitals directly on any URL without requiring a separate Google API key. This works alongside Otto's automated image optimization, which compresses newly uploaded images and reprocesses existing files during routine maintenance passes.
For my top three paid landing pages and top three organic landing pages identified in Views 1 and 2, run psi_analyze_url for mobile devices. Return Performance, Accessibility, Best Practices, and SEO scores, along with Largest Contentful Paint (LCP), Interaction to Next Paint (INP), and Cumulative Layout Shift (CLS). Identify specific render-blocking scripts, uncompressed images, or layout shifts damaging conversion rates.
View 6: Organic social and local search pulse
Brand awareness relies on consistent local and social distribution. Otto MCP queries organic engagement across Instagram, Facebook Pages, LinkedIn Pages, read-only YouTube channel metrics, and read-only Google Business Profile locations in one consolidated view.
Generate a 14-day Organic Social and Local Pulse. Pull reach, impressions, and engagement rates across my connected Instagram, Facebook Page, and LinkedIn company page. Pull view counts and subscriber metrics from my connected YouTube channel, and pull search impressions, phone call clicks, and driving direction requests from my Google Business Profile. Outline the top two performing organic posts and local search visibility trends.
View 7: Turning dashboard bottlenecks into staged website and ad fixes
Most analytics dashboards stop at identification. An Otto-connected Claude dashboard lets you diagnose an issue and immediately stage the fix. Otto is not a closed website builder; he works directly on your existing site. Every website update stages as a draft: WordPress edits default to draft status across 29 specialized tools (15 read tools and 14 write tools), while changes to custom repositories open as a GitHub pull request with an isolated Vercel deployment preview. On LinkedIn Ads, campaign updates stage in DRAFT status, with $0 budgets automatically rejected.
As Enrique Gomez, Owner of Enzo Maverick Pilates and Experiences Sensory, notes: "Otto has been pivotal for us. The ease of making website changes on the fly lets us sharpen our brand and revise as fluidly as we need." Approved client names and logos on ottowebmaster.com include Enzo Maverick Pilates, Sensory Pilates & Wellness, Wyn Wellness, Complex Physical Therapy & Performance, Urban Boxing, and Tapper Training. Website and ad edits stay draft-first behind human approval gates rather than going live from analytics alone.
Take the underperforming landing page flagged in View 1 with a high bounce rate. Draft an updated hero section, clearer value propositions, and revised CTA copy directly to WordPress as a DRAFT page (or open a pull request on GitHub with a Vercel preview link). Ensure all primary keywords from View 2 are integrated naturally into H2 tags. Do not publish live; stage the draft and output the preview URL for my review.
Explore Otto AI Website Manager to see how staged updates and code reviews work across WordPress and modern git repositories.
View 8: The Monday executive digest and change-log audit
Rather than rebuilding reports manually every week, teams can use Claude to compile an executive audit. Otto tracks deploys and site modifications via otto_get_change_log, pairs with automated weekly SEO reports, and integrates with WhatsApp for one-tap approvals on pending edits.
Run otto_get_change_log for the past 7 days and synthesize a concise Monday Executive Digest. Include:
1. Executive summary of weekly revenue, total ad spend, blended ROAS, and organic search impressions.
2. Summary of completed website and ad updates from the change log.
3. List of pending drafts or pull requests awaiting human approval.
4. Top three priorities for this week based on conversion leaks identified across GA4, paid ads, and Search Console.
Format this as a clean briefing ready to share with stakeholders.
Review our guide on replacing the Monday spreadsheet report with an automated weekly digest to streamline leadership reporting.
Read-only analytics vs. approval-gated writes: how Otto's permission matrix protects your accounts
A live marketing dashboard must maintain clear operational boundaries. In Otto MCP, reporting integrations are strictly read-only, matching Google's security architecture where the official Google Analytics Model Context Protocol server operates exclusively in read-only mode to prevent external configuration changes. Otto enforces this same model across reporting platforms, and places write operations behind approval gates.
| Integration | Read Permissions | Write Permissions | Human Approval Safeguard |
|---|---|---|---|
| Google Analytics 4 | Full read access via runReport | Strictly read-only | Cannot modify property settings, conversions, or user permissions |
| Google Search Console | Query performance, clicks, impressions, CTR, positions | Strictly read-only | Cannot modify sitemaps, indexing settings, or domain verification |
| Google Ads | Full read access to campaigns, ad groups, spend, impressions | Strictly read-only | Cannot alter bids, change campaign status, or spend ad budget |
| Google Business Profile | Locations, reviews, impressions, direction calls | Strictly read-only | Cannot alter business hours, address, or publish local posts |
| YouTube | Read video views, impressions, subscriber analytics | Strictly read-only | Cannot upload videos, delete videos, or modify channel settings |
| Google Sheets | Read cell values and target tables via batchGet | Strictly read-only in MCP catalog | Cannot overwrite spreadsheet formulas, cells, or sheets |
| Shopify | Read live orders, line items, customers, products | Strictly read-only in MCP catalog | Cannot alter product pricing, adjust inventory, or modify orders |
| PageSpeed Insights | Full Lighthouse performance diagnostics via psi_analyze_url | Strictly read-only | Read-only tool using Otto platform keys |
| WordPress | Read posts, pages, tags, categories, media (15 tools) | 14 write tools across content creation | New content defaults to draft status; human must publish |
| GitHub / Vercel | Read repository structure, commits, pull requests | Create branch and open pull request | Every site change opens as a PR with preview; human merges |
| LinkedIn Ads | Read campaign spend, impressions, CTR, leads | Create and stage draft campaigns and ad sets | Campaigns start as DRAFT; $0 budgets rejected; activating requires confirm=true + Reviewer/Admin |
| Meta Ads | Read campaigns, ad sets, ad performance, ad library | Create draft ads, update ad status | Requires Master plan plus Reviewer/Admin approval |
| Facebook Pages & Instagram | Read page impressions, post reach, engagement | Publish live feed posts and media updates | Unlike WordPress drafts, publishes go live immediately; requires Reviewer or Admin permissions |
Checkout pages, contact forms, and legal policy pages require explicit human approval in Otto core before changes go live. This operational matrix ensures Claude functions as an analytical assistant and drafting copilot rather than spending budget or updating production code without your sign-off.
Otto Webmaster vs. Opteo, Madgicx, Synter, and Ryze AI: pricing, honest limitations, and FAQs
Choosing the right dashboard and marketing tool depends on whether your team needs single-channel ad recommendations or cross-channel execution that extends to your live website.
- Opteo monitors Google Ads accounts and suggests bid, negative keyword, and search query improvements. However, Opteo covers Google Ads exclusively, with no support for Meta Ads, SEO, website code, or Claude MCP connectivity.
- Madgicx provides creative analytics and rule-based bid automation specifically for Meta Ads in e-commerce environments. It does not manage SEO, Google Search Console, or website landing pages.
- Synter provides paid media API connectors across 40+ ad networks for Claude, ChatGPT, and Cursor users, but does not deploy technical SEO fixes, manage WordPress sites, or open GitHub pull requests.
- Ryze AI focuses primarily on paid ads and SEO/GEO workflows, but does not manage live website edits via GitHub pull requests, WordPress pages, or Google Business Profile updates.
- Otto Webmaster unites cross-channel analytics, organic search, paid advertising, and approval-gated website execution across 227 tools and 15 integrations.
To evaluate these options in detail, see our comparison guide on the 7 best Ryze AI alternatives for paid ads and SEO in 2027.
Plans and pricing
Otto MCP uses the same transparent plans as the core Otto platform rather than charging an isolated MCP fee.
- Trial: $1 for 3 days. Includes full website updates, SEO management, and read-only access to connected ads and socials. A homepage redesign built from your existing content and images is included and approved by you before publishing.
- Pro: $199/month (or $99/month billed annually). Includes full website updates, SEO maintenance, and read-only access to paid ads and socials.
- Master: $3,500/month (or $3,000/month billed annually). Adds full approval-gated write and execution access to paid ads and socials, with human sign-off on every live change. Cancel anytime.
Review all plan details on the Otto Trial, Pro, and Master pricing page.
Honest limitations
Otto Webmaster is built for clear, controlled execution, which comes with deliberate boundaries:
- Write access to paid ad accounts and social publishing requires the Master plan; Pro is limited to read-only access for ad and social tools.
- White-label agency portal features are limited.
- Otto operates through the hosted MCP server today; the Otto CLI for terminal environments is coming soon and is not currently shipping.
- Otto supports 15 integrations; unannounced third-party platforms like Wix, Webflow, and TikTok are not part of the live MCP catalog.
- Otto never guarantees specific search rankings, Google Map Pack positions, or ChatGPT citations; he structures clean technical code, schema, and directory listings so search crawlers and AI assistants can accurately index your brand.
Frequently asked questions
Do I need a separate paid Google Analytics or Search Console API key? No. You authenticate your Google accounts via standard OAuth inside Otto once. Otto manages token refreshes server-side, and your Claude connection uses your single otto_mcp__ token.
Does Otto MCP run locally or in the cloud? Otto MCP runs as a remote HTTP server at https://otto.islaintel.com/api/mcp. Because it handles processing in the cloud, it does not require local Python or Node dependencies on your computer.
Can Claude accidentally publish changes or spend my ad budget? No. Reporting tools are strictly read-only. Content edits in WordPress default to draft status, code updates open as GitHub pull requests with Vercel preview links, and LinkedIn ad modifications require explicit confirmation parameters plus Reviewer or Admin permissions.
Which AI assistants are supported? Otto MCP works with Claude Desktop, Claude Web, Claude Code, Cursor, ChatGPT, OpenClaw, and Hermes, while core Otto tasks can also be managed via WhatsApp, web dashboard, and email.
Connect Claude to Otto MCP to launch your live cross-channel marketing dashboard today.
Otto Website Analytics
Monday should say what actually changed.
Otto checks that the numbers are still counting, then sends the week in plain language. Visitors, the pages that moved, and the fix worth doing next.
See how Otto does it