An executive dashboard gains nothing by showing everything.
It must isolate what deserves a decision.
1. Key figures
| Figure | What you need to understand | Source |
|---|---|---|
| 2 or 14 months | In standard GA4, user-level data retention can be set to 2 or 14 months. An executive view must therefore make other arrangements for long-term comparisons where necessary. | Google Analytics Help - Data retention |
| 16 months | Google recommends using the Last 16 months in Search Console to contextualise a traffic drop and identify seasonality. | Google Search Central - Debugging Search traffic drops |
| 3 months | By default, the Search Console Performance report displays click and impression data for the previous three months. Leaders need to know whether the dashboard presents a short- or long-term trend. | Search Console Help - Performance report |
| 5 UTM parameters | Google documents utm_source, utm_medium, utm_campaign, utm_term and utm_content to identify campaigns. Without a naming convention, the overall view mixes sources. | Google Analytics Help - URL builders |
| 100 characters | GA4 event-parameter values are generally limited to 100 characters. Labels that are too long or unstable weaken custom dimensions. | Google Analytics Help - Automatically collected events |
| 13 months | The CNIL limits the lifetime of certain consent-exempt audience-measurement trackers to 13 months. KPI interpretation must account for these constraints. | CNIL - Audience measurement tools |
| Raw events | GA4's BigQuery export makes it possible to query raw events in SQL and combine them with other sources when standard reporting is no longer sufficient. | Google Analytics Developers - BigQuery export |
2. Introduction
An executive dashboard often begins with a good intention: making activity visible. Then the charts accumulate. Traffic, sessions, impressions, CTR, conversions, cost per click, bounce rate, forms, revenue, pipeline, leads, followers, open rate, page views, scroll depth, speed and consent. Everything appears. But when a decision is required, the screen falls silent.
The verdict is simple: a view that prompts no decision provides comfort, not control.
We advocate a stricter approach. Leaders do not need to see every available figure. They need to see the signals that explain whether the business is gaining visibility, attracting the right prospects, converting better, selling more effectively, protecting its margin and detecting risks. Everything else belongs in operational diagnosis.
A good management view is less like a museum of charts and more like a navigation table: a few well-calibrated instruments, read at the right time.
3. Symptoms: when the dashboard creates noise
A noisy dashboard is easy to recognise. It shows curves without thresholds. The KPIs change at every meeting. GA4 data does not match the CRM. Paid campaigns are judged on cost per lead, but never on sales quality. SEO is reduced to traffic, without queries or landing pages. Satisfaction metrics are not connected to retention. There are so many alerts that they no longer trigger any response.
Everything is surfaced. Nothing is prioritised.
This is not a tooling problem. Looker Studio, GA4, BigQuery, Sheets, CRM and Search Console can all support clear management. The noise comes from a design flaw: data is added because it exists, not because it informs a decision. Analytical curiosity is confused with executive accountability.
The consequence is subtle but real: teams spend time commenting on fluctuations instead of correcting what matters.
4. Stakeholders: leadership, marketing, sales, GA4, CRM, Search Console and Looker Studio
For leadership, useful reporting must connect several stakeholders. Otherwise, it becomes a mirror of a single tool.
| Stakeholder | Role in the dashboard | Question to resolve |
|---|---|---|
| Leadership | Decides trade-offs, budgets, priorities and risks. | What deserves a decision this month? |
| Marketing | Owns acquisition, content, campaigns and conversion. | Which channels create valuable demand? |
| Sales | Qualifies leads and tracks pipeline, opportunities and revenue. | Which demand actually turns into revenue? |
| SEO | Tracks visibility, queries, landing pages and internal linking. | Which pages support organic growth? |
| GA4 | Measures events, conversions, journeys and sources. | Which behaviours precede an important action? |
| Search Console | Shows clicks, impressions, CTR and position. | Is organic demand genuinely growing? |
| CRM | Confirms qualification, sales cycle, value and status. | Are the generated leads worth the effort? |
| Looker Studio | Combines and visualises sources. | Does the decision become clear? |
| BigQuery | Acts as an analytics warehouse when raw data is required. | Do we need to go beyond standard reporting? |
Interpretation becomes rigorous when these stakeholders accept one rule: a KPI without an owner is merely decoration.
5. Definition: what makes an executive dashboard useful
A useful executive dashboard is a concise, reliable and decision-oriented view that brings together the indicators required to manage acquisition, conversion, sales quality, revenue, satisfaction, risk and action priorities over a defined period.
This definition matters. It describes a function, not a tool. The dashboard may live in Looker Studio, Sheets, Power BI, Notion or a CRM. Its quality depends less on presentation than on its ability to answer three questions: what is happening, why, and what are we doing about it?
It is not merely reporting. It is a decision-making forum.
6. Why this is becoming a priority in 2026
Businesses have more data than before, but do not necessarily make better trade-offs. GA4 has shifted measurement towards events; Search Console provides detailed views by query, page, country and device; CRMs track sales qualification; Looker Studio combines different sources; BigQuery makes it possible to query raw events when standard reports become insufficient.
The difficulty comes from this abundance. The more sources multiply, the more likely leaders are to receive a dashboard that adds incompatible metrics together: GA4 sessions and Search Console clicks, CRM leads and GA4 conversions, advertising spend and unattributed revenue, customer satisfaction and email open rates.
Consent introduces another limitation. The CNIL regulates trackers, and Google consent mode adapts tags to user choices. A fall in sessions may therefore be caused by a channel, a tag, a CMP, seasonality or a genuine change in demand. A serious management view must display measurement limitations, not conceal them.
In 2026, the challenge is no longer visualisation. It is making trade-offs with humility and precision.
7. KPI, metric and signal: three levels that must not be confused
A KPI is a performance indicator tied to a strategic objective. A metric describes a useful measure that is not necessarily decisive. A signal warns of a phenomenon that warrants investigation.
This distinction removes a great deal of noise. Signed revenue may be a KPI. The number of leads may be a metric. A sudden increase in 404 pages is a signal. All three are useful, but they should not occupy the same place on an executive dashboard.
Where an immature view stacks everything at the same level, a controlled dashboard establishes a hierarchy: outcome, probable cause, action. As in modern painting, strength sometimes comes from what is removed from the frame.
Control begins with hierarchy.
8. Recommended method: 8 building blocks for an executive dashboard
The method below is not a proprietary Logiks method. It draws on good practice in marketing management, analytics, CRM, SEO, consent and data governance.
This type of reporting is built like an architecture: foundations, columns, circulation and light. Not like a wall of posters.
8.1. Define recurring decisions
Start with the decisions leadership must make: increase a budget, stop a channel, strengthen an offer, improve a page, accelerate sales follow-up, correct a tracking problem, prioritise a segment, recruit or revisit positioning.
Each decision needs one primary indicator, one contextual indicator and a threshold. Without a threshold, the dashboard narrates. With a threshold, it alerts.
The decision comes before the visualisation.
8.2. Separate outcomes, acquisition and quality
At this level, outcomes, demand and quality must be distinguished. Signed revenue does not tell the whole story if the pipeline is emptying. Traffic is worthless if leads are poor. Acquisition cost becomes misleading if the sales conversion rate falls.
We recommend three simple columns: business outcomes, demand generation and demand quality. This structure avoids confusing volume with value.
No volume without quality. No quality without sufficient volume.
8.3. Choose a small number of primary KPIs
An effective executive view contains few primary indicators. For an SME or B2B website, KPIs may include revenue or pipeline, qualified leads, cost per qualified enquiry, form conversion rate, SEO's share of enquiries, sales conversion rate, lead response time, satisfaction or retention.
The exact list depends on the business model. A SaaS company will examine activation, churn, MRR, pipeline and trial-to-paid conversion. A service business will examine qualified leads, appointments, closing rate, average value and margin.
Simplicity is not a shortcoming. It is a deliberate position.
8.4. Keep diagnostic metrics at the second level
Diagnostic metrics remain necessary, but they must not overwhelm the first page. Sessions, page views, CTR, impressions, scroll depth, speed, engagement rate, CTA clicks, 404 pages or cookie refusals help explain a problem. They are not always executive KPIs.
Place them in a diagnostic tab that can be opened when the main signal deteriorates. The leader sees the alert; the operational team investigates the cause.
The executive view opens the door. It does not replace the workshop.
8.5. Connect GA4, Search Console and CRM
GA4 measures behaviour on the website. Search Console reveals organic demand before the click. The CRM confirms commercial value. Separating these views produces half-truths.
To be useful, reporting must connect at least three levels: acquisition, conversion and qualification. A page may attract many SEO clicks but few leads. A campaign may generate few forms but excellent opportunities. A channel may look expensive yet produce the best customers.
Performance is rarely understood through a single tool.
8.6. Display measurement limitations
Credibility requires limitations to be displayed: period covered, data source, refresh delay, consent, tracking changes, missing data, breaks in a time series, Search Console anomalies and CRM changes.
Google Search Central recommends contextualising traffic drops with 16 months of data where relevant. The advice applies more broadly: a KPI without historical context can prompt an overhasty reaction.
Methodological caution does not weaken control. It makes it more robust.
8.7. Establish a review ritual
Without a ritual, the best dashboard becomes wallpaper. Define who reads it, when, with which question, what decision is expected and what record is retained. Weekly for acquisition, monthly for leadership and quarterly for strategy: cadence should follow the decision cycle.
The ritual must produce an action: maintain, correct, test, stop, invest or wait. Otherwise, the meeting becomes a guided tour.
The tool is alive only if it changes the work.
8.8. Plan for the model to evolve
A useful management tool evolves. The KPIs for a redesign are not those for an acquisition phase. Launch indicators are not those of a mature website. SaaS metrics are not those of a consultancy.
Plan a quarterly review: KPIs to retain, metrics to move, sources to improve, alerts to remove, new thresholds and owners to confirm.
Stability does not mean immobility.
9. Logiks recommendations: choose measurable simplicity
We recommend starting with an executive page, not a complete report. That page must keep one promise: in five minutes, you should know whether activity is improving, stable, deteriorating or requires investigation.
First recommendation: limit the initial view to no more than ten indicators. Beyond that, you create a trading floor without a trader.
Second recommendation: replace flattering averages with useful ratios. Cost per lead is not enough; examine the cost of a qualified enquiry. Traffic is not enough; examine the pages that generate enquiries. Conversion rate is not enough; examine the value produced.
Third recommendation: add a “decision” column. For each KPI, state the possible action if the threshold is crossed. This column turns the dashboard into a management tool.
Fourth recommendation: document the sources. A figure from the CRM is not the same type of evidence as a GA4 event or a Search Console click. The view must say so.
Finally, we recommend displaying less certainty and more judgement. Partial data remains useful when its limitation is clear.
10. Decision framework: which KPIs to keep, monitor or remove
| Indicator | Recommended level | Why |
|---|---|---|
| Signed revenue | Executive KPI | It measures the actual outcome. |
| Qualified sales pipeline | Executive KPI | It anticipates future growth. |
| Qualified leads | Marketing and sales KPI | It connects acquisition with value. |
| Cost per qualified lead | Acquisition KPI | It avoids optimising for high volumes of low-quality leads. |
| Form conversion rate | Diagnostic metric | Useful for pages and UX, but insufficient on its own. |
| Search Console clicks | SEO metric | Read alongside impressions, queries and pages. |
| GA4 sessions | Context signal | Too broad to guide decisions on its own. |
| Page views | Secondary signal | Useful for content, rarely strategic on its own. |
| Engagement rate | Diagnostic | Interpret according to intent and page type. |
| Social followers | Often remove | Decorative if not connected to demand, traffic or revenue. |
| Lead response time | Operational KPI | It directly affects sales conversion. |
| Consent refusals | Data-quality signal | It helps interpret measurement fluctuations. |
The framework does not prohibit secondary metrics. It gives them their place. The executive view must remain a decision room, not an attic.
11. Common mistakes: eight dashboards that create the illusion of control
First mistake: confusing aesthetics with usefulness. A beautiful chart without a decision remains decorative.
Second mistake: displaying the same KPIs to every team. Leadership, SEO, acquisition and sales do not manage at the same level.
Third weakness: forgetting sales quality. Marketing can generate large numbers of useless leads.
Fourth pitfall: mixing raw and interpreted data. A validated CRM figure and a web event do not carry the same evidential weight.
Fifth risk: failing to display the period. A seven-day trend does not tell the same story as a sixteen-month change.
Sixth source of confusion: stacking sources without a data dictionary. The metrics become impossible to defend.
Seventh mistake: hiding consent limitations. Missing data eventually gets interpreted as truth.
Final issue: never removing an indicator. A dashboard that is never pruned ends up looking like a library that is never weeded.
12. 30 / 60 / 90-day action plan
12.1. Within 30 days
- list recurring executive decisions;
- inventory existing dashboards;
- identify the KPIs that are actually used;
- remove metrics without an owner;
- connect GA4, Search Console and CRM;
- document sources and periods;
- create an initial executive page.
Start with clarity, not exhaustiveness.
12.2. Within 60 days
- define alert thresholds;
- create diagnostic tabs;
- connect relevant CRM data;
- check UTM conventions;
- incorporate consent limitations;
- train dashboard readers;
- establish the monthly review ritual.
Management begins to produce decisions.
12.3. Within 90 days
- audit decisions made using the dashboard;
- remove unused indicators;
- improve unreliable sources;
- decide whether BigQuery is necessary;
- document breaks in time series;
- create a monthly archive;
- review KPIs against the sales strategy.
At this point, the dashboard no longer merely describes activity. It directs action.
13. FAQ: executive dashboards and marketing KPIs
13.1. How many KPIs should an executive dashboard contain?
There is no universal rule, but the initial view should remain concise. For an SME, five to ten primary KPIs are often enough: revenue, pipeline, qualified leads, cost per qualified enquiry, conversion, valuable SEO traffic, satisfaction or response time.
13.2. What is the difference between a KPI and a metric?
A KPI is directly connected to a strategic objective. A metric helps explain a phenomenon. GA4 sessions may be a metric; qualified leads from a priority channel may become a KPI.
13.3. Should web traffic appear on an executive dashboard?
Yes, but rarely on its own. Traffic must be segmented: organic, paid, direct, strategic pages, queries, conversions and CRM quality. A global volume can reassure without explaining anything.
13.4. Is Looker Studio enough for an executive dashboard?
Often, yes. Looker Studio can connect sources such as Google Analytics, Google Ads and Sheets through connectors. When joins become complex or raw history matters, BigQuery or a data warehouse may become useful.
13.5. How can you avoid a decorative dashboard?
Connect each KPI to a decision, an owner, a source and a threshold. If nobody knows what to do when the figure changes, the indicator should be removed, moved or reformulated.
13.6. Does incomplete consent-dependent data make the dashboard useless?
No. It primarily requires the limitations to be documented. A serious dashboard states what is measured, what depends on consent, what comes from the CRM and what must be interpreted cautiously.
14. Conclusion: the dashboard becomes a decision table
An executive dashboard is not a reporting object. It is a decision table. It brings together a small number of the right indicators. It separates KPIs, metrics and signals. It displays its limitations. It triggers action.
We are not seeking more spectacular dashboards. We are seeking better-equipped decisions: invest, stop, correct, wait or investigate further.
It is no longer merely a dashboard.
The decision table sets its tempo: see, judge, act.
15. Main sources
- Google Analytics Help - Data retention - accessed on 17 June 2026 - https://support.google.com/analytics/answer/7667196
- Google Analytics Help - Automatically collected events - accessed on 17 June 2026 - https://support.google.com/analytics/answer/9234069
- Google Analytics Help - About key events - accessed on 17 June 2026 - https://support.google.com/analytics/answer/9267568
- Google Analytics Help - URL builders: collect campaign data with custom URLs - accessed on 17 June 2026 - https://support.google.com/analytics/answer/10917952
- Google Search Central - Debug Google Search traffic drops - accessed on 17 June 2026 - https://developers.google.com/search/docs/monitor-debug/debugging-search-traffic-drops
- Search Console Help - Performance report - accessed on 17 June 2026 - https://support.google.com/webmasters/answer/7576553
- Google Search Central Blog - Introducing weekly and monthly views in Search Console - published in December 2025, accessed on 17 June 2026 - https://developers.google.com/search/blog/2025/12/weekly-monthly-views-search-console
- Looker Studio / Google Cloud Docs - About data sources - accessed on 17 June 2026 - https://docs.cloud.google.com/data-studio/about-data-sources
- Google Analytics Developers - BigQuery export for Google Analytics - last updated on 20 August 2025, accessed on 17 June 2026 - https://developers.google.com/analytics/bigquery/overview
- Google Tag Platform - Set up consent mode on websites - accessed on 17 June 2026 - https://developers.google.com/tag-platform/security/guides/consent
- CNIL - Cookies: solutions for audience-measurement tools - accessed on 17 June 2026 - https://www.cnil.fr/fr/cookies-solutions-pour-les-outils-de-mesure-daudience
- CNIL - Sheet n°16: Use analytics on your websites and applications - accessed on 17 June 2026 - https://www.cnil.fr/en/sheet-ndeg16-use-analytics-your-websites-and-applications