Believe it or not, I recently worked with a company that had 17,000 (!!) dashboards. Insane!

They’d ended up there through a string of mergers and acquisitions—and never cleaned up..

Unsurprisingly, 99% of those dashboards added zero value to the business.

Also, a few days ago, someone complained to me on LinkedIn that he inherited 768 dashboards from his predecessor - with less than 2% actually being used.

If you’ve followed me for a while, you know I have a love-hate relationship with dashboards.
I think they can turn into a full-blown epidemic if you don’t stop it early.

So, if you're tired of drowning in dashboard requests and want to focus on work that's actually fun, this newsletter episode is for you.

So, what's the ideal number of dashboards?

I often provocatively say that the ideal number of dashboards is zero.

That might be a bit extreme—but honestly, I believe the real number isn’t far off.

Here’s how I think about finding the ideal number for any business.

I think that, in an ideal world, there are three types of dashboards:

1. Outcome Dashboards
2. Health Dashboards
3. Input Dashboards

Outcome Dashboards

Those are dashboards that focus on outcome metrics that you want to actively improve right now.

Outcome metrics are those metrics that are not directly influencable - such as customer acquisition costs or conversion rates.

I will show you some examples from my own company - Data Action Mentor - further below to make this clearer.

Health Dashboards

Those are dashboards that focus on outcome metrics that you want to observe, but they are not part of your targets, goals or key results.

Think of blood pressure. This is a metric that you want to observe but as long as it stays within a healthy range you won't actively work on improving it.

Metrics like revenue and profit typically belong in Health Dashboards.

I don’t include them in Outcome Dashboards because they make terrible key results: you’ll always want to see them go up, no matter what.

Input Dashboards

Those are dashboards that focus on actionable, directly influencable input metrics that have a mathematical or directional relationship with outcome metrics.

The important words here are "actionable" and "influencable". Outcomes are not actionable, because you can change them only indirectly.

Inputs can be directly influenced with the goal to create Outcomes.

A good example for input metrics would be the number of sales calls per sales agent which should (theoretically) have a positive relationship with the outcome metric "Sales".

A framework of dashboards

I think the right answer to the question "How many dashboards" does a company need is:

1 Company-wide Outcome Dashboard + 1 Company-wide Health Dashboard + 1 Outcome Dashboard * relevant business areas (such as domains, functions, countries, shops) + 1 Input Dashboard * relevant business areas

Company Wide Dashboards

The first dashboard I typically build is a company-wide outcome dashboard.

The company-wide Health Dashboard can (and should) be combined into here because metrics can switch between being focus metrics and health metrics pretty quickly.

One quarter, you’re just monitoring the conversion rate. The next quarter, it drops—so it moves into the Outcomes Dashboard.

I’ll show you how I visualize that with an example further below.

Lower-Level Outcome Dashboards

Next, I always try to map the contribution of:

• business units, domains or functions (such as marketing, sales, finance, data, IT etc)
• countries
• shops
• business lines (such as B2B vs B2C)

to the overall company-wide outcomes.

For example, we could have the following setup in a SaaS business, operating in Germany and the US.

• Total MRR (Monthly Recurring Revenue) would be inside the company-wide health dashboard.
• New MRR would be inside the outcome dashboard for marketing, if Marketing is a centralized function
• MRR Germany and MRR US would be inside the two country-level outcome dashboards
• New MRR Germany and New MRR US would appear on a third-level Outcome Dashboard if marketing is decentralized by geography.

Second-Level Input Dashboards

Lastly, it makes sense to have actionable Input Dashboards on the business unit-, country-, shop- or business-line-level

It’s important that every input metric is clearly linked to an outcome metric from one of the Outcome Dashboards.

This is a GREAT use case for KPI trees!

How many levels?

Now, it is possible to sub-divide each relevant business area further down.

For example, the marketing department might be split into Search Advertising, Display Advertising, Email Marketing and SEA.

I always aim to model the organizational structure as a high-level KPI tree that shows how business units, countries, shops, and business lines are connected. Then I build a more detailed KPI tree to map the metrics within each unit.

Then any of these units can have one Outcome Dashboard and one Input Dashboard.

The challenge is maintaining discipline and building systems that keep this dashboard framework intact—even when mergers, acquisitions, or restructurings happen.

The key is to focus on establishing this framework—and protecting it.

In my experience, simply having the framework in place and communicating it regularly within the team and across the organization gets you halfway there.

Dashboards at Data Action Mentor

To show that I practice what I preach, I’ll share the two dashboards I use for my own company: Data Action Mentor.

Data Action Mentor sells digital products such as the Masterclass - "From Dashboard Factory to Strategic Partner."​ 

My company-wide Outcome Dashboard looks like this:

The Dashboard is subdivided into the customer journey steps:

• Acquisition
• Activation
• Revenue
• Retention
• Referral

I find that this split works well for most businesses.

My company-wide Input Dashboard looks like this:

This is currently still work in progress - I only have inputs for the Acquisition part, as this is my major focus right now.

The dashboards are all built in Google Sheets, connected to a BigQuery Data Warehouse.

As you can see, the outcome metrics in the "Acquisition" area focus on converting social impressions into website visitors.

My actionable input levers are 1) the number of posts and 2) the percentage of posts that include a direct link to my website.

I use the column "Prio" to differentiate between Outcome Metrics and Health Metrics.

1️⃣: This is the one metric that matters (OMTM): the actionable input metric that I currently focus on with full attention. In my case, this is the percentage of LinkedIn posts with a direct link to my website in order to drive up the Click-Through Rate. You will notice that most of my LinkedIn posts will have some form of call-to-action to a (usually) free offer (such as a signup to my Newsletter).

⭐️: Currently not pictured is my North Star Metric: This outcome metric is the main indicator for the value generated in the marketplace by my offer. In other words: This metric needs to show that I built something that people love. For Airbnb this metric is "nights booked", for Whatsapp this metric is "messages sent". The metric that I chose here is the percentage of people who tell me that my Masterclass - "From Dashboard Factory to Strategic Partner"​ exceeded their expectations (which is currently at an unbelievable 100%).

✅: These are outcome metrics that I currently actively try to improve (part of my quarterly goals).

🩺: These are health metrics that I observe but currently not actively try to improve.

My Marketing Outcome Dashboard looks like this:

This is the Outcome Dashboard on the second-level for my marketing department, where I look at how well each LinkedIn post is driving website traffic and newsletter signups.

Bottom Line

In my experience, the number of dashboards a company needs depends on just one factor: how many relevant business areas (such as domains, functions, countries, shops, or business lines) exist.

If any relevant business area has more than 3 dashboards (1 Outcome, 1 Health, 1 Input) then something is wrong.

So, while the ideal number of dashboards may be higher than 15 for complex businesses, the number should definitely be below 17,000.