agentleFS
ghazanfar111212ghazanfar111212/claude-skills-performance-marketing/skills/looker-studio-expert/SKILL.md

looker-studio-expert

Skill0 starsChanged 38 days agoMITView on GitHub

Expert skill for building client-ready marketing dashboards in Looker Studio (formerly Google Data Studio). Use whenever the user mentions Looker Studio, Data Studio, marketing dashboards, client reporting, blended data sources, GA4 reporting, multi-currency reporting, cross-platform marketing reports, or "build a dashboard for [client]." Also trigger for "connect [data source] to Looker Studio," "calculated fields in Data Studio," "Supermetrics alternative," or reporting template design. Do NOT use for Tableau, Power BI, or custom-coded dashboards. Covers data source connections, blending, calculated fields, filter design, multi-currency handling, dashboard UX patterns, and client-ready report templates.

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 16 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈16

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When no connector exists (or you don't want to pay):

    line 42
  2. Distinctive:

    Use platform's native export or API

    line 43
  3. Distinctive:

    Blend: When reporting on the same entity across sources (campaign → revenue)

    line 53
  4. Distinctive:

    Separate: When each source answers a different question

    line 54
  5. Distinctive:

    Use VLOOKUP-style calculated fields

    line 161
  6. Distinctive:

    Use color consistently (brand or semantic)

    line 190
  7. Distinctive:

    Avoid default rainbow colors - looks amateur

    line 196
  8. Distinctive:

    Always include a date range control at the top

    line 217
  9. Distinctive:

    Date range: Default for every dashboard

    line 223
  10. Distinctive:

    Use extract mode where available

    line 246
  11. Distinctive:

    Before sharing with a client:

    line 250
  12. Distinctive:

    Use n8n or Make to pull key metrics → send summary

    line 271
  13. Distinctive:

    Not handling currency conversions (mixing $5 AUD with $5 USD)

    line 281
  14. Distinctive:

    No labels on pie chart slices

    line 285
  15. Distinctive:

    Not explaining what each metric means (client won't ask, just mistrust)

    line 289
  16. Distinctive:

    Document expected variance ranges per metric

    line 303

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Looker Studio Expert

When to Use This Skill

  • Building a new client dashboard from scratch
  • Blending data across ad platforms, analytics, and CRM
  • Creating multi-currency reports (AUD, USD, GBP, PKR, AED)
  • Designing reusable templates across clients
  • Troubleshooting data discrepancies between platforms and dashboards
  • Reducing manual reporting time with automation

Data Source Fundamentals

Native Connectors (Free, Built-in)

  • Google Ads
  • Google Analytics 4
  • Google Sheets
  • BigQuery
  • YouTube Analytics
  • Search Console

Partner Connectors (Usually Paid)

  • Supermetrics: Widest connector coverage, monthly per destination
  • Windsor.ai: Good pricing, destination-first model
  • Funnel.io: Enterprise-grade, strong for ecommerce
  • Dataslayer: Budget alternative
  • Catchr: Newer, growing connector library

Most-Used Non-Native Connectors for Agencies

  • Meta Ads (Facebook/Instagram)
  • TikTok Ads
  • LinkedIn Ads
  • Shopify
  • HubSpot / GHL (via Sheets usually)
  • Bing Ads (Microsoft Advertising)

The Sheets Fallback Pattern

When no connector exists (or you don't want to pay):

  1. Use platform's native export or API
  2. Push to Google Sheets via Zapier/n8n/Make
  3. Connect Sheets → Looker Studio
  4. Schedule refresh

This pattern is especially useful for: CRM data, offline conversions, custom KPIs, aggregated data.

Blending Data

When to Blend vs. Separate Sources

  • Blend: When reporting on the same entity across sources (campaign → revenue)
  • Separate: When each source answers a different question

Blend Mechanics

  • Specify join keys (date is most common)
  • Choose join type: Left, Right, Inner, Full Outer, Cross
  • Left join is safest default (primary source drives rows)

Common Blend Recipes

  • Paid performance + revenue: Google Ads + GA4 (by date + campaign)
  • Multi-platform spend + leads: Meta Ads + Google Ads + CRM (by date + source)
  • Cost + conversions: Ad platform + offline sales data (by date + UTM)

Blend Gotchas

  • Blended fields can't be used in some chart types
  • Can't create calculated fields on blended sources directly (do in source first)
  • Date dimension must match granularity across sources (daily vs weekly)

Calculated Fields

Syntax Essentials

CASE
  WHEN REGEXP_MATCH(Source, "^google") THEN "Google"
  WHEN REGEXP_MATCH(Source, "^facebook") THEN "Meta"
  ELSE "Other"
END

Commonly Used Calculated Fields

CPA (Cost Per Acquisition)

Cost / Conversions

ROAS (Return on Ad Spend)

Revenue / Cost

CTR (Click-Through Rate)

Clicks / Impressions

Conversion Rate

Conversions / Clicks

Multi-Currency Normalization to USD

CASE
  WHEN Currency = "AUD" THEN Spend * 0.65
  WHEN Currency = "GBP" THEN Spend * 1.26
  WHEN Currency = "PKR" THEN Spend * 0.0036
  WHEN Currency = "AED" THEN Spend * 0.27
  ELSE Spend
END

(Rates are illustrative - pull live rates via Sheets formula and reference)

Clean Campaign Names (Extract Prefix)

REGEXP_EXTRACT(Campaign, "^([A-Z]+)-")

Date Grouping (Week of Year)

WEEK(Date)

Conditional Formatting Field

CASE
  WHEN ROAS >= 4 THEN "🟢 On Target"
  WHEN ROAS >= 2 THEN "🟡 Needs Attention"
  ELSE "🔴 Below Target"
END

Multi-Currency Handling (Your Context)

Strategy Options

Option 1: Single Reporting Currency

  • Pick one currency (usually USD for global, AUD for AU-focused)
  • Convert everything at data source level or in calculated fields
  • Pros: Clean, comparable numbers
  • Cons: FX fluctuations affect historical trends

Option 2: Multi-Currency Toggle

  • User selects view currency via filter
  • Calculated field responds to parameter
  • Pros: Flexibility for stakeholders
  • Cons: More complex to maintain

Option 3: Separate Tabs per Currency

  • One tab per market currency
  • Consolidated tab in single reporting currency
  • Pros: Locals see their numbers, exec sees unified
  • Cons: More pages to maintain

FX Rate Management

  • Store rates in a Google Sheet (updated weekly or monthly)
  • Reference sheet as data source
  • Join to main data by currency code
  • Use VLOOKUP-style calculated fields

Dashboard UX Patterns

The 3-Page Agency Dashboard Structure

Page 1: Executive Summary

  • Big number cards: Spend, Revenue, ROAS, Leads, CPA
  • Trend charts: Spend vs Revenue over time
  • Top-line observations (text callouts)
  • Month-over-month % change on key metrics

Page 2: Channel Performance

  • Breakdown by channel (Google, Meta, etc.)
  • Cost + performance table
  • Best/worst performing campaigns
  • Channel trend comparison

Page 3: Deep Dive / Insights

  • Campaign-level detail
  • Ad-level performance
  • Keyword/audience breakdown
  • Geographic performance

Visual Hierarchy Rules

  • Biggest chart = most important insight
  • Scorecards at top (glanceable KPIs)
  • Left-to-right reading flow (Western audiences)
  • Group related metrics visually
  • Use color consistently (brand or semantic)

Color Palette

  • Brand colors for primary series (agency or client brand)
  • Semantic colors: Green (good), Red (alert), Yellow (warning), Gray (neutral)
  • Sequential palettes for heatmaps
  • Avoid default rainbow colors - looks amateur

Typography

  • Headings: 18-24pt, bold, consistent font
  • Body: 10-12pt
  • Metrics/numbers: Larger than surrounding text
  • Limit to 2 fonts max across entire report

Date Range Patterns

Default Date Range

  • Last 30 days is the safe default for agency reporting
  • Include comparison: "vs. previous period" or "vs. previous year"

Custom Date Ranges for Specific Views

  • MTD (Month to Date) for in-month pacing
  • QTD (Quarter to Date) for executive reviews
  • YTD (Year to Date) for annual planning
  • Custom fiscal year for Australian clients (July-June)

Date Range Control

  • Always include a date range control at the top
  • Enable "Apply to all pages" when it makes sense

Filters & Controls

Control Types

  • Date range: Default for every dashboard
  • Drop-down: For campaign, source, account selection
  • Input box: For search-style filtering
  • Checkbox: For multi-select categories
  • Slider: For metric ranges (e.g., CPA thresholds)

Filter Strategy

  • Page-level filters (apply to one page)
  • Report-level filters (apply to all pages) - use sparingly
  • Chart-level filters (hidden from user) - for hardcoded scope

Performance Optimization

Why Dashboards Get Slow

  • Too many data sources on one page
  • Complex blends recalculating on every load
  • Unfiltered scans of large datasets

Speed Improvements

  • Pre-aggregate in BigQuery or Sheets before connecting
  • Cache data (use data freshness settings appropriately)
  • Reduce date range defaults
  • Limit number of charts per page to 8-12
  • Use extract mode where available

Client-Ready Template Checklist

Before sharing with a client:

  • Client logo in header
  • Clear title and reporting period
  • No placeholder text or test data
  • All filters working and defaulted sensibly
  • Mobile view checked (many execs view on phone)
  • Links to methodology/definitions if complex KPIs
  • Refresh schedule documented
  • Permissions set correctly (view-only for client)
  • PDF export tested (for email reports)
  • Date formats localized (DD/MM/YYYY for AU/UK, MM/DD/YYYY for US)
  • Currency symbols and decimal places consistent

Automation & Delivery

Scheduled Emails

  • Looker Studio supports scheduled PDF email delivery
  • Weekly or monthly cadence typical
  • Send to client + internal CC

Alternative: Slack/Email Automations

  • Use n8n or Make to pull key metrics → send summary
  • Better for daily pulse checks (not full dashboard)

Permissions Management

  • View access: Clients (they don't need edit)
  • Edit access: Internal team
  • Owner: Account manager or team lead (not a single freelancer)

Common Mistakes

  • Using one massive page with 30 charts (overwhelming)
  • Not handling currency conversions (mixing $5 AUD with $5 USD)
  • Inconsistent date ranges across charts on same page
  • Auto-colors that don't match client brand
  • Showing raw platform numbers that don't match GA4 (without explanation)
  • No labels on pie chart slices
  • Using 3D charts or excessive animation
  • Ignoring mobile preview
  • Hardcoding campaign names that break when clients rename them
  • Not explaining what each metric means (client won't ask, just mistrust)

Troubleshooting Data Discrepancies

Why Numbers Don't Match Across Sources

  • Attribution windows: Meta 7-day-click vs GA4 last-click
  • Conversion definitions: What counts as a conversion in each platform
  • Timezone differences: Platform UTC vs. account local time
  • Currency: Platform reports in local currency, dashboard in reporting currency
  • Sampling: GA4 samples high-volume accounts
  • Bot filtering: Different rules per platform
  • Tracking issues: Missing tags, broken UTMs

Reconciliation Approach

  • Document expected variance ranges per metric
  • Flag when variance exceeds threshold
  • Have "source of truth" designated per KPI (e.g., Shopify for revenue, not platform pixels)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.