Marketing dashboard: from click to revenue

A marketing dashboard is one self-refreshing screen holding the five to eight figures you steer your marketing on: spend per channel, cost per lead, conversion at each step and the revenue that eventually comes out of it. The difference with the reporting inside your ad platform is that this screen also shows what a channel returned, not only what it cost.

Marketing is the most heavily measured department most companies have. Every click has a source, every campaign a cost line, every newsletter an open rate. Google Ads, GA4, LinkedIn, Meta and your email platform each hand you their own reporting without being asked, and every one of them is more detailed than anything finance had twenty years ago.

And yet marketing is also the department where the budget conversation is still most often settled on conviction. That is not because too little gets measured. It is because each of those systems measures exactly one stretch of the route, and stops at the moment it gets interesting: at the lead. What that lead turned out to be worth sits in a different system, and that system does not talk to the rest.

A marketing dashboard is where the whole route can be followed in one go. Below: which KPIs belong on it, why your channels together report more conversions than you gained customers, what such a screen looks like, in which order you build it, and when you are better off waiting a while.

What is a marketing dashboard?

It is one screen that keeps itself up to date, showing the figures you use to decide where the next slice of budget goes. Not a collection of charts about marketing, but the five to eight numbers a decision hangs on in the marketing meeting.

The difference with the reports you already have is not in the figures, it is in who puts them side by side. Google Ads reports on Google Ads. Meta reports on Meta. Both accurate, both complete, and both exclusively about themselves. As long as nobody lays those reports next to each other, you are comparing channels on numbers that do not share a definition.

In most organisations that merging is still done by hand. A few exports each month, a spreadsheet, half a day of work. The result is not wrong, it is just already a month old by the time it is finished. A dashboard does the same work again every night without anybody sitting down for it.

Who looks at it also decides what belongs on it. In a mid-sized company that is usually one marketer or a small team, with a director behind them who mainly wants to know whether the money is doing anything. That is a different starting point from the examples you find online, where an agency tracks twelve clients side by side and the reporting itself is the product.

Which KPIs belong on a marketing dashboard?

Split them across three steps: what you put in, what happens along the way, and what comes out. A dashboard with only the first two shows how hard people worked. Only the third shows whether that work returned anything.

• Spend per channel. The amount and its share of the total. This is the only figure on the screen you can change yourself tomorrow morning.
• Cost per lead. Spend divided by the leads from that channel. Useful for comparing channels, risky to steer on, and the next section explains why.
• Conversion per step. Visitor to lead, lead to conversation, conversation to customer. This is where you see it stall, and it is rarely at the ad.
• Number of leads and where they came from. The inflow. Without this figure a drying-up market looks like a quiet summer for two months.
• Revenue per channel. What was actually invoiced to customers who arrived through that channel. This figure does not come out of your ad platform.
• Cost per customer. Spend divided by customers, not by leads. For most companies the gap between those two numbers is wider than the gap between their channels.
• Customer lifetime value. What a customer returns on average for as long as they stay, not just on their first order. Without it, a channel delivering expensive but loyal customers looks worse than one delivering cheap customers who order once and disappear.
• POAS rather than ROAS. ROAS divides revenue by ad spend. POAS does the same with margin, and that is almost always the more useful number. A ROAS of four sounds excellent, right up until you notice it is on products at twenty percent margin and every euro of ad budget leaves you eighty cents.

Eight is the ceiling, not the target. Start with the four your team already talks about and add the rest only when somebody asks. What separates a KPI from an ordinary metric is the target beside it: with no target, you are reading a position rather than a signal.

The list of what does not belong on it is almost longer. Followers, impressions, reach, open rates and click-through rates: all measurable, all useful as an explanation after the fact, and none of them a lever you can pull. You will find them on nearly every example dashboard published online, which is exactly why those screens are so often impressive and so rarely used. They can sit in a second layer for anyone who wants to dig. On the opening screen they take the place of a figure a decision actually hangs on.

The figure marketing rarely has in house

Every KPI in the list above except the last four already sits in a system you could open today. Those last four are the ones about money coming in, and they live somewhere else: in your CRM, your accounting system or your ERP. That is where the real work of a marketing dashboard is, and it is exactly why most marketing reporting stops where it starts to matter.

What that costs you in insight is easy to work out. Take two channels, each with a budget of € 4,000 a month.

Channel A produces 80 leads, so € 50 per lead. Ten percent of those become customers: 8 customers, or € 500 per customer. The average order is € 3,000 at a 25 percent gross margin, so € 750 of margin per customer. After the acquisition cost you keep € 250 per customer, and € 2,000 across the channel.

Channel B produces 40 leads, so € 100 per lead. Twice as expensive, and on any marketing dashboard that stops at the lead, that is the worse channel. Except 25 percent of those leads become customers: 10 customers, or € 400 per customer. The average order there is € 5,000, so at the same margin that is € 1,250. Net € 850 per customer, and € 8,500 across the channel.

Four times the result from the same budget, out of the channel that lost on cost per lead. Steer on that one number and you move budget in precisely the wrong direction, with figures to back the decision up.

Which makes the point sitting underneath this whole article. The problem is almost never that too little is measured. Marketing measures more than any other department, and it is still marketing where the budget conversation gets settled on experience and persuasion. One more figure does not fix that. The missing piece is the link between the click and the invoice, and that link is a connection, not another chart.

In practice that means one thing: the source of your customer has to travel with them. Record where every lead came from, carry that field through into the order, and make sure it reaches your accounting system. Then revenue per channel is a sum rather than a reconstruction. That margin per customer comes from the same corner as the figures on a financial dashboard, and the step from lead to signed deal is worked out on the sales dashboard.

Why your channels together report more conversions than you have customers

Google Ads reports 40 conversions. Meta claims 25. GA4 lands on 48. Your CRM holds 31 new leads, and accounting eventually recognises 9 of them as customers. Five systems, five answers to what feels like the same question.

Sound familiar?

The awkward part is that none of those five figures is wrong. They simply are not counting the same thing. Every ad platform claims the conversion it had a touchpoint in, inside its own lookback window. Someone who saw a LinkedIn ad, clicked a Google search ad a week later and only then filled in the form gets claimed by both. They are not lying, they are each reporting neatly on their own share.

Google describes that crediting as a choice rather than a measurement: in Analytics, the attribution model you pick decides which touchpoint gets the credit, and there are several to pick from. Change the model and the outcome changes without a single extra customer appearing.

Add those platform reports together and you have won more customers than you sent out quotes. That makes for a pleasant month for everyone except the bookkeeper.

The fix is not a better model but a clear agreement: one system counts the customer, the rest supply context. In nearly every mid-sized company that system is the CRM, because that is where the customer who actually signed is recorded. The ad platforms stay on the dashboard for what they do know well, namely what was spent and how many people clicked. What counts as a conversion gets settled once and then applies to every channel. That is the same agreement that precedes any form of data-driven working: decide what a figure means first, display it second.

Online marketing dashboard: which sources do you connect?

An online marketing dashboard needs four to six sources in practice, and each supplies a different part of the same story:

• The ad platforms. Google Ads, Meta, LinkedIn, sometimes Microsoft Advertising. Spend, impressions and clicks come from here. These are the most reliable figures on the whole screen, because they are invoices.
• Google Analytics 4. Behaviour on your site: which pages get visited, where people drop out, which conversions happen on the site itself.
• Search Console. The organic side. Impressions and positions in search results, the one channel with no media budget behind it.
• Your email platform. Sends, clicks and unsubscribes. Most interesting combined with the CRM, because then you see which recipients eventually became customers.
• The CRM. Leads, their source and their status. This is where the customer gets counted, which makes it the hinge of the entire dashboard.
• Your accounting system or ERP. Invoiced revenue and margin. Without this source the dashboard stops at cost.

A ready-made connection exists for almost all of those systems. Microsoft maintains an overview of the data sources Power BI Desktop supports, and Google Analytics, Salesforce and the well-known accounting packages are on it. Anything that is not almost always offers an API or a periodic export.

One thing is worth knowing up front, because it comes as a surprise later otherwise. Since websites started running consent banners, your analytics no longer measures everybody. For visitors who decline, Google Analytics models the behaviour based on comparable visitors who did consent. That is a defensible solution, but it produces an estimated figure that sits on your dashboard just as confidently as an invoiced amount.

That need not be a problem as long as everyone knows. Keep modelled figures out of the same tile as figures from your accounting system, and use them for trends rather than for settling accounts. For the question of whether a channel is growing, an estimate is fine. For the question of how much revenue that channel produced, you reach for the CRM and the invoice.

What does a good marketing dashboard look like?

Three layers stacked, and the order is not arbitrary. It runs from the decision to the explanation, which is exactly the direction somebody reads a screen in when they have ten seconds for it.

• The top says where you stand. Three or four tiles with the outcomes against their target: revenue from marketing, cost per customer, POAS, new customers. Every number with its target beside it and the gap spelled out. Anyone who stops reading here has to know whether something is wrong.
• The middle says where it came from. Per channel, what it cost and what it returned, on the same row. This is the layer where the budget decision gets made, and therefore the layer where the return column cannot be missing.
• The bottom holds the movement. The trend over the last twelve months and the drill-through to the detail. Nobody starts here, and everybody who wants to work something out ends here.

Four rules keep such a screen readable. Everything fits on one screen without scrolling, because what sits below the fold does not get looked at, not even by people who promise otherwise. Colour means deviation, not decoration: with four tiles coloured because it looks better, nobody can see which tile is the problem any more. A figure always sits in the same place, so that after three weeks you know where to look without reading. And every tile drills through to the layer beneath, because the question of how that happened always comes.

That structure also solves the urge to build one screen everybody gets something out of. What you get is a screen nobody gets anything out of. The marketer lives in the middle layer, the director in the top one, and those are two different reading moments on the same screen rather than two separate screens.

Five steps to a dashboard people actually use

The order below is the reverse of how it usually goes. Almost every project starts by connecting the ad platforms, because that is the technically easiest part and the fastest way to get something to look at.

1. Start from the decision, not from the source. Which choice has to get better? Usually: where the next slice of budget goes. Write that question down before anything gets built, because it governs everything after it.
2. Pick the KPIs and write down a definition for each. What counts as a lead, from which moment somebody is called a customer, and over which period you calculate. One page is enough. Without that page you end up arguing about figures instead of about budget.
3. Connect sources in order of value, not of convenience. So the CRM and the accounting system first, because that is where the answer to your question sits. The ad platforms after. Reverse it and after a week you have a beautiful cost overview and still no idea what it returned.
4. Build one screen and have somebody who did not build it test it. Sit them in front of it with no explanation and ask what they think is going on. If that answer does not match reality, the screen is at fault, not the reader.
5. Agree who owns which figure and when you look. Not the person who maintains the dashboard, who can tell you exactly which table a number comes from but not whether it is the right number. In marketing that line often runs straight through two departments: cost per lead belongs to marketing, conversion from lead to customer belongs to sales. Leave that unsettled and a disappointing month belongs to both, and therefore to nobody.

On the looking: weekly at the channels, monthly at the return. Refreshing once a night covers both. Cost per lead and conversion per step do not move by the hour, and a figure that keeps jumping invites watching rather than steering. Only high daily volume, such as a webshop during a sale week, makes anything faster worth it.

At board level this usually ends up as one or two tiles. The full marketing funnel does not fit on a management dashboard and does not belong there either: what the leadership team needs is what marketing costs and which share of new revenue comes out of it. The rest is the team’s own working screen.

Looker Studio or Power BI?

For a marketing dashboard this is the choice that actually matters, and the answer hangs on one question: does it stay within marketing sources, or does the rest of the business join in?

If you stay inside Google Ads, GA4 and Search Console, Looker Studio is the obvious starting point. It is free, the connections to Google products are ready to go, and you have something working within a day. For a marketer tracking their own campaigns that is often enough, and turning it into a BI project would be silly.

The limit comes into view once three things happen at the same time: sources outside the Google ecosystem, history you want to keep for years, and definitions that have to match across channels. At that point you start doing by hand in Looker Studio what a data model should be doing for you, and that is exactly when Power BI turns out cheaper than it looks. What the licences cost is set out in our article on Power BI licensing; for most mid-sized teams it comes down to a handful of users.

That is also where the difference sits between a template and a custom marketing dashboard. Ready-made templates exist for every ad platform, and for what they do they are fine: showing what that one channel did. But no template knows your definition of a lead, your stages in the CRM or your margin per product group. Those three are what decide whether the screen produces a decision. Custom here does not mean prettier, it means your own figures fit in it.

If you are considering having a marketing dashboard built, this is the question to answer beforehand rather than halfway. Building in the wrong tool does not cost money, it costs the second build.

When a marketing dashboard is not your first move

There are three situations where we would suggest waiting a while.

• Lead source is not being recorded. If a lead arrives without anything capturing where it came from, revenue per channel is a guess with a chart around it. Sort that field out first, in your form and in your CRM.
• There is no fixed moment to discuss it. A screen without a meeting gets looked at for two weeks and then not again. Set the meeting up first, with a manual list if you have to.
• The budget is not fixed. With no agreed amount per period, every spend chart is an observation. There is nothing to steer with, only something to explain afterwards.

In all three cases the answer is not never, but something else first. What you can do in the meantime: pick one channel and follow it through to the invoice by hand for a quarter. If that produces a different conversation about where budget should go, you know the screen will pay for itself. If it produces nothing, you are a few hours down rather than a whole project. Our approach opens with the question of which decision has to get better, not with which sources we could connect.

---

Do you know at the end of the quarter which channel brought you customers, or only which one gave you the most clicks? Want to know which sources have to be joined up for that, and what it would cost you? Get in touch, and we will walk your channels through with you.

Frequently asked questions