Marketing Metrics That Actually Matter: A Simple Small-Business Dashboard by Matan Guy

By Matan Guy · November 12, 2026 · 7 min read

I'm Matan Guy, and for a long time I looked at my marketing numbers the wrong way. I opened Ads Manager, felt good or bad depending on the day, and changed things on instinct. The problem wasn't too little data. It was too many numbers that didn't lead to any decision. Here are the marketing metrics I actually use, how to calculate them, and the one-page weekly dashboard I keep.

Vanity metrics vs decision metrics

A vanity metric is a number that feels good but doesn't tell you what to do next. Likes, followers and video views often fall into this group. They aren't useless, but they don't tell you whether marketing is making the business healthier.

A decision metric is a number that, if it moves, tells you to change something specific. If cost per acquisition rises, you look at the creative, the audience or the page. My test is simple: for every number on my dashboard, I should be able to say "if this goes up or down, I will do X." If I can't, it doesn't belong there.

The core marketing metrics, defined in plain words

Here is the set I come back to. I'll use letters and round, hypothetical numbers only to show the math, not as targets or results.

Cost and attention metrics

CPM (cost per 1,000 impressions) — how much it costs to show your ad a thousand times. Formula: ad spend ÷ impressions × 1,000. It reflects audience and season, not ad quality.
CTR (click-through rate) — the share of people who saw the ad and clicked. Formula: clicks ÷ impressions × 100. A low CTR usually points to the hook or the message.
CPC (cost per click) — what each click costs. Formula: ad spend ÷ clicks. Cheap clicks only matter if those people go on to do something.

Conversion metrics

Conversion rate — the share of visitors who take the action you want, such as buying or leaving details. Formula: conversions ÷ visitors (or clicks) × 100. If CTR is fine but this is weak, the problem is usually the page, the price or the offer.
Cost per acquisition (CPA) / customer acquisition cost (CAC) — what it costs to get one customer. Formula: marketing spend ÷ new customers. If you spend S and get N orders from new customers, your CAC is S ÷ N.
Average order value (AOV) — the average amount per order. Formula: revenue ÷ number of orders.

Profit metrics

ROAS (return on ad spend) — revenue attributed to ads divided by ad spend. If ads drove R in revenue and you spent S, ROAS is R ÷ S.
Contribution margin after ad spend — what's left from sales after the direct costs of each order and the ads. Formula: revenue − product cost − shipping − payment and platform fees − ad spend. This is the number that tells you whether marketing is actually paying for itself.
Customer lifetime value (LTV) — how much gross profit a typical customer brings over their whole relationship with you. A simple version: average order value × gross margin % × average number of orders per customer.
Repeat purchase rate — the share of customers who bought more than once in a given period. Formula: customers with 2+ orders ÷ total customers × 100.

Why ROAS alone can mislead you

ROAS is popular because it's one tidy number. The trouble is what it leaves out.

Imagine two products, both with a ROAS of 3: every 1 unit of ad spend brought back 3 units of revenue. If the first is cheap to make and ship, plenty is left. If the second has thin margins, the same ROAS can lose money on every order. ROAS doesn't know your product cost, shipping, fees or returns.

It also depends on attribution settings, and a campaign that reaches people who were going to buy anyway can show a strong ROAS while adding little new business.

So I treat ROAS as a quick signal and contribution margin after ad spend as the truth. Before running ads, work out your break-even ROAS: if your margin before ads is M% of revenue, break-even ROAS is roughly 1 ÷ M. With a margin of half the revenue, break-even ROAS is about 2. Below that, every sale costs you money.

Building a one-page weekly dashboard

My dashboard is a single sheet: one row per metric, one column per week. This table reminds me why each row exists.

Metric	Where to find it	Question it answers
Ad spend	Ads platform	How much did I invest this week?
CPM	Ads platform	Is reaching my audience getting more expensive?
CTR	Ads platform	Are my ads stopping people?
CPC	Ads platform	Is traffic affordable?
Conversion rate	Store or analytics tool	Does the page and offer persuade visitors?
CAC / CPA	Spend ÷ new customers from store data	What does a new customer cost me?
AOV	Store reports	Are people buying more per order?
ROAS	Ads platform (cross-check with store)	Quick read on ad efficiency
Contribution margin after ad spend	My own spreadsheet	Did marketing actually make money?
Repeat purchase rate	Store customer reports	Are customers coming back?
LTV (monthly update)	Store data + margin	How much can I afford to pay for a customer?

Two habits keep it useful. I use the store's numbers for revenue and orders, because the ads platform only sees what it can attribute. And I write one sentence under each week about what changed, so I can connect causes to numbers later.

Tracking basics you need first

A dashboard is only as good as the data behind it. Three foundations matter most.

UTM parameters. These are tags added to the end of a link (source, medium, campaign, and sometimes content) so your analytics tool knows where a visit came from. Use one lowercase naming style so "Instagram" and "instagram" don't become two sources.
Pixel and Conversions API. The pixel reports events from the browser; the Conversions API sends them from your server or store platform, filling gaps when browser tracking is blocked. I covered the setup in my Facebook and Instagram ads guide.
A consistent attribution window. Pick one window and keep it fixed, or old and new results stop being comparable. Platforms update their options and defaults, so check the current settings in your own account rather than relying on what you read anywhere, including here.

My weekly review routine

Once a week, same day:

Update the numbers from the store and the ads platform into the sheet.
Start at the bottom line. Did contribution margin after ad spend go up or down?
Walk up the funnel to find the cause. Margin down? Check CAC. CAC up? Check conversion rate, then CTR, then CPM. The first metric that moved clearly is usually where the problem is.
Pick one or two actions. A new batch of creatives, a page change, a price or bundle test. My approach to testing creatives is in my ad creative framework.
Write down what you changed so next week's numbers have context.

If you change everything at once, you won't know what worked.

Common mistakes with marketing metrics

Tracking everything. Forty rows get ignored; ten good rows get used.
Trusting the platform's revenue blindly. Cross-check with your store's actual orders.
Ignoring margins. Revenue is not profit. Know your product costs before reading ROAS.
Optimizing only for new customers. Repeat customers often make the economics work, so watch retention too.

If you're still choosing what kind of online business to build, the metrics that matter shift with the model, which I discuss in online marketing business models. More about my background is on the about page.

FAQ

How many marketing metrics should a small business track?

Fewer than you think. Around eight to twelve, reviewed weekly, is enough, as long as each one connects to a real decision.

What's the difference between CPA and CAC?

CPA is usually the cost of any conversion you optimize for, such as a lead or a purchase. CAC is the cost of acquiring a new customer. If many buyers are returning customers, CAC is the more honest number for growth.

How often should I recalculate customer lifetime value?

Monthly or quarterly is usually enough. LTV changes slowly, so weekly updates mostly add noise. Recalculate after a meaningful change to pricing or products.