Business dashboard: a live example and how to build one
Ask yourself: how much did your business earn this month? If the answer starts with "let me ask the accountant" or "hold on, the reps will send a spreadsheet" — you don't have analytics. You have people who assemble numbers by hand on request, and those numbers come out a little different every time.
Building business dashboards is exactly what I do to fight that. There are no screenshots in this article — instead, there are live blocks from a demo copy of my working dashboard: the company and the numbers are made up, the mechanics are real. You can poke every block right here in the text, and the full version opens separately.
The eight numbers a day starts with
The top row of the dashboard answers "how are we doing?" in ten seconds. How many leads came in and how many are real, how much we spent on ads, how much we sold — and separately, how much money actually landed in the account. Here's that block, live:
What to look at here. "Sales" and "Cash-in" are two different numbers, and that's not a mistake: sales per the CRM almost never match the money that reaches the account. The gap between them shows where the sales are "on paper." And every number has a plan and a run-rate: plan-vs-actual says "you're at 71%," the run-rate says "you're pacing to 105%, you'll make it." Without the run-rate, any mid-month point looks like a failure, and the founder rides the team for nothing.
Trends and ad-account events on one chart
Next comes a day-by-day chart: leads, qualified leads, invoices, sales, spend, visits. The trick isn't the lines — it's the diamonds above them: those are events from the ad accounts. Raised a budget, swapped a creative, paused a campaign — every action is marked on the timeline. When leads dip, you don't have to rack your brain over "wait, what did we change back then?" — hover over a diamond and you see that budgets were being tweaked three days before the drop.
The funnel: where the money leaks out
Leads → qualified → invoices → sales, with the conversion rate at each step. The red "already lost" badges are a separate piece of honesty: how many of the qualified leads are already gone while you're busy celebrating the number at the top.
Next to the funnel on the full dashboard are the loss reasons: "too expensive," "couldn't reach them," "bought elsewhere." Not the reps' gut feelings, but counted categories out of the CRM. If "couldn't reach them" suddenly doubles, the problem isn't the ads, it's response speed — and you see it a week before revenue drops.
Reps: plan-vs-actual by person
A card for each: leads, qualified leads, invoices, sales, plan, and average deal size. The "qualified → invoice" conversion instantly shows who closes customers and who just collects leads. I've watched this one card replace an hour-long standup — the numbers are on screen, and arguing "I had a lot on my plate" becomes pointless.
The trends table: from the month down to a single ad
The deepest block. Weeks or months across the top, metrics down the left. Now click the "Leads" row: it expands into channels, then campaigns, ad groups, and individual ads. The same move works for qualified leads, invoices, sales, and spend. This answers questions on the level of "which ad brought in the million-dollar sale" — with no exports and no pivot tables.
Why the numbers lie: what to fix before the dashboard
Now the unpleasant part. A dashboard is a storefront, and a storefront shows what's on the shelves. If half the deals in the CRM have no stages, reps enter customers however they feel like it, and UTM tags aren't set or get lost along the way — the dashboard will neatly display garbage.
That's why every project of mine starts not with charts but with an audit: how leads land in the CRM, where tags go missing, who moves deals through the stages and when. This is usually where the real reason behind "we have no analytics" turns up — the system simply isn't set up for analytics to live in it. This gets fixed in a week or two: end-to-end source tagging, required fields, lead routing. There's a detailed breakdown of that part in the piece on cross-channel analytics without Roistat.
How to build a dashboard: three paths
An honest map of the options — from experience, with the money and the limits.
| Path | Who it fits | Money | Limits |
|---|---|---|---|
| Excel / Google Sheets | A start, up to 3-5 sources, numbers you can enter by hand or with simple imports | $0 + your time | Manual updates die within a month; auto-collecting from the CRM and ad accounts is no longer a spreadsheet but an integration |
| Off-the-shelf services and BI (Roistat, Power BI, DataLens) | Standard setups, with someone to configure and maintain them | from ~$70-150/mo for a service, or from ~$1,800 for a BI rollout | You pay for the whole machine even if you need a third of it; non-standard logic (custom stages, cash-in, per-rep plan-vs-actual) runs into the platform's limits |
| An n8n pipeline + your own screen | You need logic tailored to the business and full control over the data | a one-off build + a server at a few dollars/mo | You need someone to design and build it; in return, no subscription per report |
Everything you touched above is the third path. Data from the CRM, ad accounts, and web analytics is collected on a schedule into a store by n8n, and the screen is a plain web page: it opens on your phone and is password-protected. No "per user per month" subscription — the server costs about as much as two cups of coffee.
Automated reporting: a dashboard that comes to you
The next level after "open it and look" is when you don't need to look at all. In the morning a summary lands in Telegram: yesterday's leads, spend, sales, deviations from plan. If CPL blows past a threshold, a separate alert fires right away — not in a report a month later. It's the same n8n pipeline: data from the store goes not only to the screen but to the messenger too.
And on top of that runs an AI analyst: you ask a question in plain language — "why did qualified leads dip in June?" — and get an answer calculated from your own data. How agents like that are built and where they pay off is covered in the piece on AI agents for business.
What it costs and how long it takes
Rough market benchmarks: spreadsheets are free, but you pay in time; cross-channel analytics services run from ~$70-150 a month forever; BI studios ask from $1,800-3,600 for a rollout. An n8n pipeline like the one in the demo above takes about two weeks to build: first order in the data and tags, then the store and the screen, then tuning against live numbers. The only recurring cost left is the server.
I build dashboards like these to order: the makeup depends on your CRM, your ad sources, and what's already in your data, so I quote the price after a short look at your setup, not off a rate card. Message me on Telegram or take a look at the service page — the format and stages are there. The real thing to weigh when choosing a path isn't the price of the tool, it's the price of blindness: what a month of ad budget costs when nobody knows for sure whether it's working.
Frequently asked questions
What is a business dashboard in plain terms?
A screen where a business's key numbers are collected in one place and refresh automatically: revenue, leads, spend, the funnel, plan-vs-actual. You open it and in a minute you understand what's happening, without asking staff to pull reports.
How is a dashboard different from a report?
A report is assembled by a person on a deadline, and the numbers are out of date the moment it's sent. A dashboard refreshes itself from the original sources: the CRM, ad accounts, counters, the bank. A report is "what happened"; a dashboard is "what's happening now."
Which metrics belong on a founder's dashboard?
At a minimum: revenue and cash-in, the number and cost of leads, qualified leads, the funnel through to the sale, spend by channel, plan-vs-actual with a run-rate. For the sales team, a breakdown by rep.
Can you build a dashboard in Excel or Google Sheets?
For a start, yes: up to ~5 sources and as long as you're willing to update the numbers by hand. Beyond that you hit auto-collection of data from the CRM and ad accounts — that's where an integration begins, and the spreadsheet stops being a spreadsheet.
What is cross-channel (end-to-end) analytics?
When the customer's path is visible all the way through: from an ad click to money in the till. It ties ad spend to sales through the CRM and answers which channel brings profit and which just burns budget.
Why don't the numbers in the CRM and the ad account match?
Lost UTM tags, leads that bypass the CRM, duplicates, different attribution models and time zones. It's cured by order in tagging and lead routing — without that, the dashboard shows a pretty lie.
How often should a dashboard refresh?
For decisions, 2-3 times a day is enough — in the demo above the store refreshes at 09:00, 15:00, and 21:00. Spend and leads can be pulled more often. A once-a-week manual update is a report, not a dashboard.
Want the same screen for your business?
Tell me what CRM you use and where your ads come from — I'll tell you what can be built, on what timeline, and what I'll need from you. Let's talk about the task, not a slide deck.
Message me on Telegram