One dashboard for every channel: social, SEO, paid and media

Most marketing teams already have plenty of dashboards. The paid team lives in ad managers, the social lead screenshots native insights every Monday, SEO has a Search Console export, and someone in RevOps owns the pipeline report. Each view is accurate on its own terms. The trouble starts in the weekly meeting, when nobody can say whether the LinkedIn push or the webinar actually moved demand, because the numbers on the table were never designed to sit next to each other.

This post explains why that happens, what a single marketing overview should contain, and how to build one people open without being reminded. If you want the hands-on version, follow the unified marketing dashboard setup guide; for the full measurement framework, read the whitepaper One View of Marketing.

Why do fragmented channel reports hide what works?

Every platform reports on itself, with its own definitions, windows and blind spots. Three examples from the platforms’ own documentation show how far apart the counting rules are:

• Meta moved Facebook and Instagram to a single Views metric: the number of times a reel or video was played, or a photo or text post was on screen. Repeat views by the same person count, so one person seeing a photo three times in a day is three views.²Source 2 · Meta for Creators, 2024Introducing Views to simplify content metricscreators.facebook.com
• LinkedIn’s organization share statistics are organic only (sponsored activity is excluded), are available for a rolling 12-month window, and its engagement figure counts clicks alongside likes, comments and shares over impressions.³Source 3 · LinkedIn Marketing API, Microsoft LearnOrganization Share Statisticslearn.microsoft.com
• Google Search Console assigns every link inside an AI Overview the same position, and counts a click on any external link in it as a click.⁴Source 4 · Search Console HelpWhat are impressions, position, and clicks?support.google.com Its chart and table totals can differ because some views aggregate by property and others by page.⁵Source 5 · Search Console HelpPerformance report (Search results): Overview and basic setupsupport.google.com

None of these rules is wrong. They are simply different. Put "impressions" from four platforms in one column and add them up, and you get a number that describes nothing. Look at each platform in its own tab, and you lose the one thing a marketing leader needs: a sense of where the next dollar or hour should go.

What should one marketing dashboard show?

A good overview answers three questions in under a minute: Are we on track? Which channels are driving it? What changed and why? That means it is built around outcomes, not around data sources. The diagram below shows the shape we use: many sources feed a normalization layer, and one dashboard reads from shared definitions.

A shared vocabulary of six concepts

Instead of forcing every platform into one metric, map each one onto a short list of concepts and keep the original metric name visible. This is what makes a cross-channel comparison honest.

The Visits row is where comparisons become fair, because one analytics tool classifies all channels with one rule set. GA4’s default channel group, for example, treats a session as Organic Social when the source matches its list of social sites or the medium is a social value such as social or sm.¹Source 1 · Google Analytics Help[GA4] Default channel groupsupport.google.com It also now has an AI Assistant channel for visits from assistants such as ChatGPT, Gemini or Copilot, while Google’s AI Overviews and AI Mode stay in Organic Search.¹Source 1 · Google Analytics Help[GA4] Default channel groupsupport.google.com Consistent UTM parameters are what make this row trustworthy; our UTM taxonomy guide covers the naming rules.

How to lay out the overview so people actually open it

Dashboards that get used follow a reading order. Ours has five bands, top to bottom:

1. 1Outcome strip. Four to six numbers against target: pipeline created, signups or revenue, blended efficiency (for example MER), and total spend. Each tile shows period-over-period change and the source.
2. 2Channel matrix. One row per channel, columns for the six concepts. This is the only place cross-channel comparison happens, and it compares like with like.
3. 3Trends. Weekly visits and conversions by channel group, so seasonality and launches are visible.
4. 4Diagnostics. Channel-native detail on demand: top posts, top queries, campaign ROAS, email cohorts, coverage list.
5. 5Annotations. Launches, budget changes, algorithm updates, tracking changes. Without them, every spike becomes a debate.

Notice what the illustrative chart makes easy: the week-29 social spike is obvious, and an annotation ("founder post went wide") turns it from noise into a lesson. The small AI Assistant line is also visible, which matters precisely because it is small and growing.

How do you compare channels that count differently?

Three rules keep a single overview honest:

• Compare rates within a channel, outcomes across channels. An engagement rate of 4% on LinkedIn and 4% on Instagram are not the same thing, but a qualified signup is a qualified signup wherever it came from.
• Pick one referee for visits and conversions. Use your analytics tool or your CRM as the single source for anything downstream of the click, and show platform-reported conversions only as a diagnostic.
• Separate correlation from incrementality. A dashboard shows what happened. Whether a channel caused it is a question for experiments and models such as incrementality tests or media mix modeling. Google’s open-source MMM, Meridian, is one example of the modeling route.⁶Source 6 · Google for DevelopersMeridiandevelopers.google.com

For paid channels specifically, the blended ROAS post covers why platform ROAS and finance-grade efficiency diverge, and which three numbers a CFO will accept.

Where do AI search and earned media fit?

These two are usually missing from marketing overviews because their data is partial. That is a reason to show them carefully, not to leave them out. For earned media, track mentions, outlet tier and share of voice from a media monitoring feed, then link coverage to referral sessions where a link exists. For AI search, combine what is measurable (AI Assistant sessions in analytics, AI Overview clicks inside Search Console’s Web totals, and citation counts where an engine reports them) with a fixed panel of prompts you check on a schedule. Our post on generative engine optimization explains exactly what can and cannot be measured today.

Make the dashboard a product, not a project

Unified dashboards decay when definitions drift, connectors break quietly, or the person who built it moves on. The fix is ordinary product discipline:

How to build it in Kimo

In Kimo, the overview is the Command center, built on the same models as every other dashboard. Connect the sources, let Kimo map them to the shared concepts, and start from the Marketing command center template.

1. 1
   Step 1:

   Connect sources

   Authorize social pages, Search Console, GA4, ad accounts, your email tool, media monitoring and CRM from Connectors.

2. 2
   Step 2:

   Review the mapping

   Kimo proposes how each native metric maps to the six concepts. Accept, rename or exclude, and the definition is stored once in the model.

3. 3
   Step 3:

   Publish the overview

   Deploy the template, set targets for the outcome strip, and add the first annotations.

4. 4
   Step 4:

   Ask follow-ups in plain English

   Use Ask Kimo for questions like "Which channel drove the most qualified signups last month, and what changed?" Every answer shows the measures and filters it used.

Teams like this DTC brand start with the overview and then add channel-specific views, such as the Social media overview and the SEO funnel. See the Kimo Marketing product page for the full picture.