Every analytics product wants to sell you a wall of charts.
You do not need a wall of charts. You need to know, in about eleven seconds, whether the business is fine or whether something is wrong. That's it. That's the entire job of a dashboard for a company your size.
A wall of charts fails at that job. Twenty numbers means no number is important. You look at it twice, feel vaguely informed, and never open it again. I have watched owners pay for beautiful dashboards they haven't logged into since the demo.
Here's what actually works: six numbers, one screen, delivered to you rather than waiting for you to go find it.
The six numbers
1. Leads this week, versus last week and versus this week last year.
Not "sessions." Not "impressions." Humans who raised their hand and asked you for something. This is the top of everything, and it's the number that moves first when something breaks. When leads drop, you have between two and eight weeks before it hits your bank account, which is exactly enough time to fix it if you notice.
The year-over-year comparison matters because most small businesses are seasonal and you'll drive yourself insane comparing July to March.
2. Leads contacted within one hour.
Expressed as a percentage. This is the single best operational number in a small service business, because it is entirely within your control and it directly determines conversion.
If this number is 40%, you don't have a marketing problem. You have a phone problem, and no amount of ad spend will fix it. Fixing it costs almost nothing.
3. Quotes sent, and quotes accepted.
Two numbers, side by side, and the ratio between them.
If quotes sent is falling, your top of funnel or your speed is the issue. If quotes sent is fine but acceptance is falling, something changed — your pricing, your competition, or the quality of the leads coming in. These are completely different problems with completely different fixes, and one number can't tell them apart.
4. Revenue booked this week, and revenue collected this week.
Booked is what you sold. Collected is what actually hit the account. Small businesses die in the gap between those two, and if you only look at one of them you will be surprised at exactly the wrong moment.
5. Jobs completed but not yet invoiced.
This is the number nobody puts on a dashboard and everybody should. It is free money sitting in a drawer. Work you already did, at cost you already absorbed, that you have not asked to be paid for.
In most businesses I look at, this number is bigger than the owner expects, and the moment it's visible it goes to nearly zero, because nobody wants to look at it every day.
6. One leading indicator specific to you.
This depends on your business. Pick the thing that predicts revenue 30 to 60 days out:
- Booked jobs on the calendar for next month
- Estimates outstanding, unanswered
- Subscription or contract renewals coming up
- Active customers who haven't purchased in 90 days
- Pipeline value in the "sent quote" stage
You know what predicts your revenue. Put it on the board.
What does not belong on it
Website traffic. Interesting, occasionally. Actionable, almost never. Traffic that doesn't convert is a vanity number, and traffic that does convert already shows up in number one.
Social media followers. If your follower count changing would not change a single decision you make this week, it doesn't belong on a dashboard. Keep it on the platform where you can look at it when you're curious.
Email open rates. Directionally useful for a specific campaign, useless as a health metric, and increasingly unreliable to measure at all.
Anything you cannot act on. This is the test. For every number on the dashboard, ask: if this went red tomorrow, do I know what I'd do? If the answer is no, it's decoration.
The delivery matters more than the design
Here's the part that determines whether this works.
The dashboard has to come to you. Not live at a URL you're supposed to remember to visit. In your text messages, or your email, on a schedule, whether you asked for it or not.
Every owner I've built a login-required dashboard for stopped checking it inside six weeks. Every owner I've sent a Monday morning summary to still reads it a year later. The difference isn't the data. It's the friction. A dashboard behind a login is a thing you have to decide to look at, and on a busy Monday you will not decide to look at it.
So: Monday morning, 7am, six numbers in a text or a plain email. Each with an arrow. Up, down, flat. Nothing else.
That's the whole product.
Add the thresholds and it starts working for you
Once the six numbers exist and arrive on schedule, the last upgrade is to make the system tell you when something is wrong instead of waiting for you to notice.
Pick a floor for each number. When it breaches, you get a message. Something like:
- Leads this week are down more than 30% from the four-week average — alert
- Response rate under one hour drops below 70% — alert
- Uninvoiced completed jobs exceed some dollar figure — alert
- Zero leads in 48 hours during business hours — alert immediately, because this usually means your form is broken
That last one is worth the entire exercise by itself. Broken forms and broken phone numbers are silent. There is no error message. The only way you find out is that things get quiet, and by the time "quiet" registers as a pattern rather than a slow week, you've lost a month.
Build it cheap first
Do not go buy a business intelligence platform. Here's the sequence:
- Week one: Write the six numbers on a whiteboard and fill them in by hand every Monday. Yes, by hand. This forces you to define them precisely and to find out which ones you can't even get.
- Week two: Notice which numbers were painful to gather. That pain is telling you where your data doesn't exist. Fix the data capture, not the dashboard.
- Week three: Automate whichever numbers come from systems you already have. Most CRMs, phone systems, and invoicing tools will export or send this.
- Week four: Get it into a Monday morning message.
The whiteboard version teaches you more than the software version. Skipping straight to software means you automate the collection of numbers you haven't thought hard about.
The point of a dashboard
It is not to know your business better. You already know your business. You've been in it every day for years.
It is to know sooner. To find out on the 8th that leads are down instead of finding out on the 30th when the deposits don't come. Every problem in a small business is cheap to fix early and expensive to fix late, and the only thing standing between the two is whether you saw it.
Six numbers. One screen. Delivered, not visited.
If you want it built and landing in your phone every Monday without you touching it, that's the kind of thing I set up. Get in touch and tell me what you'd want the sixth number to be.