The CEO proudly shows me his dashboard: 47 indicators on a single screen. "We measure everything," he tells me. I ask him which ones he reviews every week. He goes silent.
I've seen this scene dozens of times. And the problem is never lack of data. The problem is that measuring everything is the same as measuring nothing.
The mistake almost every SMB makes
When a company decides to "professionalize its metrics," the first impulse is to cram everything measurable into a giant spreadsheet or colorful dashboard. Sales by day, by week, by salesperson, by product, by region. Expenses broken down into 15 categories. Detailed inventory. Hours worked. Support tickets. Everything.
The result: nobody looks at anything. Or worse, everyone looks at different things and meetings become arguments about which number is correct.
The trap is confusing available data with useful indicators. They're very different things.
What is a KPI, really?
KPI stands for Key Performance Indicator. The key word is key. Not "all indicators," but the few that actually tell you if the business is doing well or badly.
A good KPI has three characteristics:
- It's connected to a business objective. You don't measure for the sake of measuring. You measure because you need to know if you're progressing toward something specific.
- It's actionable. If the number drops, you know what levers to pull to bring it back up. If you can't do anything about it, it's not a useful KPI.
- Anyone can understand it. If you need 10 minutes to explain how it's calculated, it's too complex.
How many KPIs should you have?
Between 5 and 7 for general management. Period.
I know that sounds low. But think about it this way: if you have 7 indicators and one starts failing, you have the time and focus to react. If you have 47, none of them gets the attention it deserves.
This doesn't mean the company only measures 7 things. Each department can have its own operational metrics. But the management dashboard—the one reviewed in the weekly or monthly committee meeting—shouldn't have more than that.
How to choose the right indicators
Here's a process that has worked for me with clients in different industries:
1. Start with objectives, not data
The question isn't "what can I measure?" but "what do I need to achieve this year?" If the objective is to grow sales by 20%, you need sales indicators. If it's to improve profitability, you need margin and cost indicators. It sounds obvious, but the number of companies that measure things unrelated to their priorities is impressive.
2. Think about leading vs lagging indicators
Lagging indicators tell you what happened: monthly sales, net profit, employee turnover. They're important, but by the time you see them, it's too late to react.
Leading indicators predict what will happen: number of quotes sent, sales pipeline, customer NPS. They give you time to correct before the lagging indicator tanks.
A good set of KPIs has both. For example: if you sell B2B services, you might have "Sales closed this month" (lagging) and "Meetings with prospects this week" (leading). If meetings drop, sales will drop in 30-60 days. That's when you act.
3. The 8am meeting test
Imagine you arrive on Monday at 8am and need to review how the business is doing before an important meeting. What 5 numbers would you look at? Those are your candidate KPIs.
If the answer is "it depends" or "I'd need to check several things," you still don't have your key indicators clear.
4. Validate that they're actionable
For each candidate KPI, ask yourself: if this number drops 20%, what would I do? If you don't have a clear answer, that indicator is informative but not a management KPI.
For example: "Percentage of online vs in-store sales" is interesting, but if you're not going to change your channel strategy based on how it moves, it doesn't need to be on your main dashboard.
A concrete example: retail with 3 stores
I worked with a clothing store chain with 3 locations. They had a spreadsheet with 30+ metrics that nobody looked at. After working together, their management dashboard ended up like this:
Main KPIs (weekly review):
- 1. Weekly net sales vs target — the king number
- 2. Average ticket — if it drops, something's wrong with the mix or salespeople
- 3. Conversion (visitors vs buyers) — key leading indicator
- 4. Gross margin % — to avoid selling a lot and earning little
- 5. Days of inventory — to avoid running out of stock or being overstocked
- 6. Customer NPS — early warning for service problems
Six indicators. They fit on a phone screen. The manager reviews them every Monday in 5 minutes and knows exactly where to focus attention that week.
What they don't tell you about KPIs
There are some uncomfortable truths that nobody mentions in management books:
Your KPIs will change. What you measure when the company bills $5M isn't the same as when it bills $50M. What matters during aggressive growth isn't the same as during consolidation. Review your indicators at least once a year.
Sometimes you'll measure the wrong things. And that's okay. It's part of the process. Better to start with something imperfect and adjust than to spend six months designing the perfect set of KPIs.
Numbers without context are misleading. A 15% growth sounds good until you find out the market grew 30%. Always have a benchmark: previous month, previous year, budget, or market.
How to implement this in your company
If you've read this far and want to put this into practice, I suggest:
- Gather your key team (department managers, partners, whoever makes decisions) and do the "Monday at 8am" exercise. What do they absolutely need to know?
- Don't start from scratch. You probably already have metrics somewhere. Review which ones you actually look at and which ones you ignore.
- Define targets for each KPI. A number without a target is just a curiosity. How much is "good" and how much is "bad"?
- Choose a simple format. A shared spreadsheet, a Power BI dashboard, a whiteboard in the office. Whatever works, as long as everyone sees the same thing.
- Review them on a fixed schedule. Weekly or monthly, but always. A KPI that nobody reviews is useless.
The summary
Choosing KPIs isn't a technical data exercise. It's a business prioritization exercise. What really matters? What tells me if I'm doing well or badly? What can I act on?
If you take away just one thing from this article: less is more. Better 6 indicators that everyone understands and reviews than 60 that nobody looks at.
Need help defining your company's KPIs or building a dashboard people actually use? Let's talk.