Why ten KPIs and not fifty?

It is tempting to put everything you can measure on a dashboard. Everyone knows the result: a screen full of charts nobody looks at anymore, because it is unclear which number matters.

A KPI is not just any figure. It is a number where you know what to do when it moves the wrong way. If you cannot answer that question, it is information, not a KPI. That information can live in a detail report, but not on the overview you open on Monday morning.

Ten is not a magic number. It is roughly what a managing director or team lead can take in at a glance. The list below is a starting point: not every business needs all ten, and sometimes you swap one for a metric that matters more in your sector.

Finance: which four numbers should you always see?

Finance is usually where a dashboard starts, because the data is already structured in your accounting system. Together, these four give an honest picture of how the business is doing.

  1. Revenue versus last year and budget. Not just this month's revenue, but compared with the same period last year and with what you budgeted. A standalone revenue figure says little; the deviation says everything. Source: your accounting package.
  2. Gross margin (in euros and percent). Revenue minus direct costs (purchasing, materials, contracted hours), divided by revenue. Growing revenue with a shrinking margin means you are working harder for the same money. Source: accounting, possibly combined with purchasing data.
  3. Cash position and cash forecast. Today's bank balance, plus what you expect to come in and go out over the coming weeks. Profitable companies get into trouble too when the money lands at the wrong moment. Source: bank and accounting.
  4. Outstanding receivables and days sales outstanding (DSO). How much customers still owe you, and how many days it takes on average for an invoice to be paid. A common formula: outstanding receivables divided by revenue over the period, multiplied by the number of days in that period. A rising DSO points to a problem in invoicing or collection. Source: accounting.

If you use Moneybird, Odoo or another common package, an integration with Power BI is often the quickest route to these four numbers. Read more about the approach in Connect Power BI to your accounting.

Sales: how do you know enough work is coming in?

Finance looks back. Sales tells you what is going to happen in the coming months. Three numbers are enough to see whether the pipeline is healthy.

  1. Weighted pipeline value. The value of all open quotes and opportunities, multiplied by the likelihood they will close. A pipeline of two hundred thousand euros where half has been sitting still for months is worth less than it looks. Source: your CRM.
  2. Quote-to-order conversion. The share of quotes you win, ideally per month and per salesperson or product group. A falling conversion rate often says more about price or positioning than falling revenue does. Source: CRM or quoting module.
  3. Customer concentration. The share of your largest customers in total revenue, for example the top 5 or top 10. If a large part of your revenue depends on a few customers, that is a risk you will not see in the revenue chart. Source: accounting or CRM.

These numbers usually come from a CRM such as HubSpot or Teamleader. Keep in mind: they are only as good as the discipline with which opportunities are tracked. If deals are only entered once they are signed, you are not measuring a pipeline but history.

Operations: which numbers fit your business?

Operations is where businesses differ most. An installation company steers on different things than an online shop or a consultancy. These three apply broadly; pick the variant that fits you.

  1. On-time delivery. The percentage of orders or projects delivered on the promised date. It is one of the few numbers that directly reflects how your customer experiences you. Source: ERP, order administration or project software.
  2. Productivity: billable ratio or revenue per FTE. In a service business, that is the share of hours worked that you can invoice. In trade or manufacturing, revenue or gross margin per FTE is often more useful. Source: time tracking and HR or payroll system, for example through an integration with Nmbrs.
  3. Days of inventory or lead time. If you hold stock, you want to know how many days of sales are sitting on the shelf: too much is idle cash, too little costs revenue. No stock? Measure the lead time from request to delivery instead. Source: ERP, point-of-sale system or project software.

Torn between two operational metrics? Pick the one your team can actually influence. A KPI nobody can change is soon ignored.

What does this look like in practice?

A fictional example to make it concrete. Imagine a technical wholesaler with 25 employees, an accounting package, a CRM and an ERP for stock and orders. On Monday morning the managing director opens one page with three blocks.

  • Finance at the top: revenue this month against last year and budget, gross margin in percent, bank balance with a six-week forecast, and DSO with the five oldest open invoices below it.
  • Sales in the middle: weighted pipeline by expected closing month, quote conversion over the last three months and the share of the top 10 customers.
  • Operations at the bottom: on-time delivery per week, revenue per FTE and days of inventory per product group.

Each number is coloured against a target the team agreed on itself. Clicking a number opens a detail report. But the overview itself stays one page, and that is what the weekly meeting is about.

How do you start with a KPI dashboard?

You do not need to build all ten at once. An approach that works well in practice:

  1. Pick three to five KPIs to start with. Begin with the numbers that currently take you the most time to pull together by hand. Usually those are the finance figures.
  2. Define each KPI. What counts as revenue: invoiced or delivered? Including or excluding credit notes? Which costs are direct? Write it down before you build anything. Otherwise you end up arguing about the number instead of the cause.
  3. Decide the source and the owner. Which system does the number come from, and who is responsible for the quality of the data there? Check our integrations page to see whether there is already an existing integration for your package.
  4. Build a first version with real data. A rough page with correct numbers beats a polished design with sample data. That way, feedback is about content.
  5. Agree on a target per KPI. A number only means something once you know what good looks like. A DSO of 45 days may be fine in one sector and a problem in another.
  6. Use the dashboard in your regular meeting. A dashboard that merely exists changes nothing. One that is on the table every week does. Only add new KPIs once the first ones are in use.

Do your numbers still come from separate Excel sheets? Then also read Excel or Power BI? When it is time to switch.

What mistakes do you see most often?

  • Wanting too much at once. A dashboard with thirty numbers is no longer opened after a month. Start small and expand once it is used.
  • No fixed definitions. If finance and sales use different revenue figures, the dashboard loses its credibility within a week.
  • Only looking back. A dashboard with only finance numbers tells you what happened. Add at least one forward-looking number, such as pipeline or cash forecast.
  • Ignoring messy source data. Duplicate customers, missing fields or misposted costs show up on a dashboard as wrong numbers. Sometimes the underlying data needs sorting out first.

Help setting up your KPI dashboard

The list above is a starting point, not a rulebook. Which KPIs make the difference for you depends on your sector, your systems and the decisions you take every week. Want to talk through which numbers belong on your dashboard and where they come from? See what we do in dashboarding and how our approach works.