Dashboards (Looker, etc.)

Most marketing dashboards are graveyards. Forty tiles, three filters, and nobody opens them after week two. This lesson teaches you how to build a dashboard that an exec actually checks on Monday morning, and that a campaign manager actually uses to kill a bad ad on Tuesday afternoon.

Quick Summary

• A good marketing dashboard answers 5-7 recurring questions with live data, nothing more.
• 72% of marketers export dashboard data to Excel when the dashboard fails them. Your goal is to make that unnecessary.
• Build two separate dashboards: one for executives (revenue, ROI, pacing) and one for operators (channel details, creatives, conversion rates).
• Every chart must map to a pre-written decision. If you cannot complete "if this number does X, we will do Y," cut the chart.
• Trust beats sophistication. Stale data and undefined metrics kill dashboards faster than bad design.

What a Marketing Dashboard Actually Is

A marketing dashboard is a single page (or a small set of linked pages) that answers a small set of recurring questions, in priority order, using live data.

Think of it like a car's instrument cluster. Your speedometer, fuel gauge, and engine warning light are always visible. Your full service history is in the glove box. The dashboard shows what you need to drive, not everything the car knows.

In tools like Looker Studio (formerly Google Data Studio), Tableau, or Power BI, you connect data sources (like Google Analytics 4, Meta Ads, your CRM) and build visual summaries on top. The tool does the math. You make the decisions.

Why This Skill Matters (With Real Numbers)

The marketing analytics market hit $7.12 billion in 2025 and is forecast to reach $14.55 billion by 2031, growing at 12.65% per year. AI-driven analytics adoption jumped from 31% of organizations in 2024 to 56% in 2025, and is projected to hit 78% by 2028.

But here is the trust gap: 87% of marketers say data-driven decisions are critical. Only 32% actually trust their data. The result? Most dashboards become "definition fights" in meetings instead of decision tools.

Key stats that explain the opportunity:

• Companies using effective dashboards make decisions 5x faster than competitors
• Effective dashboards cut reporting time by 80%
• 72% of users regularly export data to Excel when their dashboard fails to answer their question
• 40% of users rate their dashboards 3 out of 5 or lower
• Most marketing dashboards fail within 90 days: not because of bad data, but because nobody opens them

ASUS unified marketing data across all global regions using a centralized dashboard setup and now saves 90 hours per week on manual reporting, work that used to take days now takes minutes. Admiral Media moved from weekly reporting cycles to real-time reporting and saves 35-40 hours per week via automation.

The Two Types of Dashboards You Need

Most marketing teams make one mistake: they try to serve every audience with one dashboard. It does not work.

Dashboard 1: The Executive Scorecard

Audience: CMO, VP Marketing, CEO Refresh: Daily Rule: One screen, no scrolling, no filters needed

What goes on it:

• Blended customer acquisition cost (CAC = total marketing spend divided by new customers)
• Return on ad spend (ROAS = revenue from ads divided by ad spend)
• Pipeline created this month vs. target
• Budget spent vs. budget allocated
• One trendline showing week-over-week momentum

Dashboard 2: The Operator Deep Dive

Audience: Campaign managers, media buyers, growth analysts Refresh: Hourly for paid channels, daily for organic Rule: Filterable by channel, date range, audience segment, and creative

What goes on it:

• CAC and ROAS broken down by channel (Google, Meta, LinkedIn, etc.)
• Conversion rate at each funnel stage
• Top 10 and bottom 10 performing ad creatives
• Spend pacing by campaign
• Cost per lead (CPL) vs. target

The Step-by-Step Playbook

Step 1: Write the Decision First

Before you open Looker Studio, write this sentence for every chart you plan to build:

"If [metric] goes [above/below] [threshold], we will [specific action]."

Example: "If CAC from Meta exceeds $95, we will pause broad audience campaigns and shift budget to retargeting."

If you cannot finish the sentence, the chart does not belong on the dashboard. This rule alone eliminates 60% of the clutter in most dashboards.

Step 2: Pick Your 5-7 Core Metrics

For most marketing teams, these are the canonical metrics:

MQL = Marketing Qualified Lead (a lead that meets your scoring criteria). SQL = Sales Qualified Lead (a lead that sales has accepted).

Step 3: Connect Your Data Sources

In Looker Studio, click "Add data" and connect:

• Google Analytics 4 (for website traffic and conversions)
• Google Ads (for paid search performance)
• Meta Ads via a free connector like Supermetrics or Porter Metrics
• Your CRM (HubSpot, Salesforce) for pipeline and revenue data

Step 4: Design for 5 Seconds

Research shows dashboards must communicate their main message in under 5 seconds or users start ignoring them. Use this layout hierarchy:

1. Top row: 3-4 scorecards (large numbers) showing your headline KPIs
2. Second row: Trend lines showing those numbers over time
3. Third row: Breakdown charts (by channel, by audience, by campaign)
4. Bottom: Tables for sorting and drilling down

Limit your color palette to 2-3 colors. Use green for "on track," red for "off track," and one accent color for highlighted numbers. Test with 8% of your users in mind, that is the share of people with some form of color blindness.

Step 5: Add Context to Every Number

A number without context is meaningless. Instead of showing "CAC: $67," show:

CAC: $67 | Up 12% vs. last month | 85% to target ($79)

That single line tells three stories: the absolute value, the trend, and the performance vs. goal. Add a tooltip explaining how the metric is calculated (for example, "CAC = Total marketing spend / New customers acquired, last 30 days").

Choosing the Right Tool

For 90% of marketers, Looker Studio is the right answer. It is free, connects to all major ad platforms and analytics tools, and produces professional outputs. Start there before paying for anything.

Common Mistakes That Kill Dashboards

• Tile sprawl: More than 12 charts on one page means nobody scans it. Move secondary metrics to a linked detail page.
• No definitions: If three people define "conversion" differently, every meeting becomes an argument. Put the formula in a tooltip on every KPI.
• Vanity metrics at the top: Impressions and page views belong below the fold. CAC, ROAS, and pipeline belong at the top.
• No comparison anchor: A number alone says nothing. Always show vs. previous period, vs. target, or vs. last year.
• Building without user input: The B2B SaaS company that bumped dashboard adoption from 23% to 87% in two weeks did one thing: they interviewed five stakeholders before building a single chart.
• The "set and forget" trap: Dashboards that are not reviewed quarterly become the "dashboard graveyard", reports nobody opens. Schedule a 30-minute review every quarter to remove unused charts and add new decision questions.

The One-Line Takeaway

A dashboard is not a data display, it is a decision machine, and every chart that does not trigger a specific action is dead weight slowing your team down.