There is a strange problem in digital marketing today.
We have more data than we have ever had before, yet an awful lot of business owners still have no idea whether their marketing is actually working.
Open Google Analytics and you can find sessions, users, events, engagement rates, traffic sources, landing pages, key events, revenue, acquisition reports, attribution paths, and dozens of other measurements.
Open an advertising platform and another avalanche arrives.
Impressions. Reach. Frequency. CPM. CPC. CTR. Conversions. Cost per conversion. Video views. Engagements. ROAS.
Then somebody sends you a monthly marketing report with a collection of colorful graphs, arrows pointing upward, and a sentence announcing that your campaign generated 384,000 impressions.
Great.
Did we make any money?
That is the question I wish more businesses would ask.
Analytics are incredibly powerful, but numbers without context are just numbers. The purpose of digital marketing analytics isn’t to create an impressive-looking dashboard. It is to help you understand what people are doing, why they are doing it, where your marketing is succeeding, where money is being wasted, and what you should do differently next.
For the average small-business owner, you really don’t need to become an expert in every metric available.
You need to understand the handful of numbers that tell the story of your customer journey.
And once you understand how those numbers fit together, digital analytics become considerably less intimidating.
Start With the Question Your Analytics Are Supposed to Answer
Before talking about click-through rates, conversion rates or acquisition costs, we need to establish something much more important:
What are you actually trying to accomplish?
It sounds ridiculously obvious, but this is where a surprising amount of marketing goes wrong.
Imagine two businesses.
One is a local roofing contractor.
The other is an online retailer.
The roofing contractor probably cares about qualified phone calls, estimate requests, booked inspections and ultimately signed contracts.
The online retailer cares about product views, add-to-cart activity, checkout starts, purchases, average order value and repeat customers.
They can advertise on exactly the same platform and still need completely different definitions of success.
This is why I don’t believe there is one universal digital marketing metric that matters most.
There is a hierarchy.
At the top are business outcomes.
Underneath those are conversions.
Under conversions are intent signals.
Under intent signals are engagement metrics.
And underneath engagement are visibility metrics.
Think of it like a funnel:
Visibility → Attention → Engagement → Intent → Conversion → Revenue → Retention
Every layer matters, but the closer a metric gets to actual business results, the more weight I generally give it.
That distinction alone can prevent business owners from wasting enormous amounts of time obsessing over numbers that look impressive but don’t necessarily move the business forward.
Level One: Revenue and Customer Acquisition
If you want to know which analytics deserve the most attention, start at the bottom of the funnel rather than the top.
How many customers did marketing generate?
How much revenue did those customers produce?
How much did it cost to acquire them?
Those questions matter far more than how many people happened to see an advertisement.
Customer Acquisition Cost
Customer acquisition cost, usually called CAC, is one of the most useful numbers a business can understand.
The basic calculation is straightforward:
Customer Acquisition Cost = Acquisition Spending ÷ New Customers Acquired
Suppose you spend $5,000 on marketing during a month and acquire 25 new customers.
Your acquisition cost is:
$5,000 ÷ 25 = $200 per customer.
Is $200 good?
That depends entirely on what a customer is worth.
If the average customer generates $150 in total profit, you have a serious problem.
If the average customer generates $4,000 in profit over the relationship, spending $200 to acquire that customer might be fantastic.
That is why benchmarks need context.
HubSpot’s updated research, for example, shows substantial differences in lead and acquisition costs between industries. Its 2026 analysis cites an average B2B cost per lead of roughly $84 across channels, while certain financial-services and legal leads can exceed $650. (HubSpot Blog)
That doesn’t mean $84 is automatically “good” or $650 is automatically “bad.”
A $650 lead that becomes a $30,000 client can be enormously valuable.
A $20 lead that never buys anything is worth very little.
Cheap marketing isn’t necessarily efficient marketing. Profitable marketing is efficient marketing.
Return on Ad Spend
For businesses running paid advertising, ROAS — Return on Ad Spend — is another important metric.
The basic calculation is:
Revenue Attributed to Advertising ÷ Advertising Spend
Spend $2,000 and generate $8,000 in attributable revenue, and your ROAS is 4:1.
For every advertising dollar spent, four dollars in revenue came back.
But there is an important warning here.
Revenue isn’t profit.
If you sell $100 products with an $80 cost of goods, a seemingly impressive ROAS can hide weak economics.
This is why sophisticated measurement eventually needs to move beyond revenue toward contribution margin and profit.
Small-business owners don’t necessarily need to build complicated financial models on day one. They should, however, understand that “$4 in sales for every $1 in ads” does not automatically mean “$3 in profit.”
Level Two: Conversions — Where Marketing Turns Into Action
Once we move one level up the funnel, we arrive at conversions.
This is where digital marketing becomes particularly interesting because a conversion represents somebody doing something you wanted them to do.
That could mean:
- Buying a product
- Requesting an estimate
- Completing a lead form
- Scheduling an appointment
- Calling your business
- Registering for an event
- Signing up for a trial
- Joining an email list
Google has actually changed some of its terminology around this. In Google Analytics, important business actions are now called key events, while “conversion” is increasingly used for important actions used to measure and optimize advertising campaigns. (Google Help)
Google recommends events such as generate_lead, purchase, sign_up, begin_checkout, add_to_cart, and other actions depending on the business model. (Google Help)
Don’t get too caught up in terminology.
The practical question remains:
Did people do what we wanted them to do?
Conversion Rate
Conversion rate tells you what percentage of people completed the desired action.
Suppose 1,000 people visit your landing page and 50 request an estimate.
Your conversion rate is 5%.
Now this becomes incredibly useful when you compare it against other metrics.
Imagine traffic increases dramatically, but conversions stay flat.
That’s telling you something.
Perhaps you’re attracting the wrong audience.
Perhaps the advertising promise doesn’t match the landing page.
Perhaps the offer isn’t compelling.
Perhaps the website creates too much friction.
Perhaps visitors don’t trust you enough yet.
This is where analytics stop being a scoreboard and start becoming a diagnostic tool.
Level Three: Cost Per Lead
For service businesses especially, Cost Per Lead, or CPL, is one of the first numbers I want to know.
The formula is simple:
Marketing Spend ÷ Leads Generated
You spend $1,500.
You generate 30 leads.
Your CPL is $50.
Again, though, this number cannot live in isolation.
Imagine Campaign A generates 100 leads at $20 each.
Campaign B generates 40 leads at $40 each.
At first glance, Campaign A appears substantially better.
Then you look deeper.
Only five Campaign A leads become customers.
Twenty Campaign B leads become customers.
Suddenly that cheap $20 lead isn’t looking nearly as attractive.
Campaign A:
$2,000 spend → 100 leads → 5 customers.
Actual acquisition cost: $400 per customer.
Campaign B:
$1,600 spend → 40 leads → 20 customers.
Actual acquisition cost: $80 per customer.
This is one of the biggest lessons in digital marketing analytics:
Never confuse lead quantity with lead quality.
The cheapest lead is not necessarily the best lead.
The best lead is the one most likely to become a profitable customer.
That is also why connecting marketing analytics to a CRM, sales process or even a well-maintained lead-tracking system can completely change your understanding of marketing performance.
Google’s current recommended event framework reflects this full-funnel approach. For lead-generation businesses, Google recommends distinguishing among generated leads, qualified leads, working leads, converted leads and leads that ultimately don’t convert. (Google Help)
That is far more valuable than simply announcing:
“We generated 72 leads.”
My next question is:
What happened to them?
Level Four: Click-Through Rate — Is Your Message Working?
Now we start moving higher in the funnel.
Click-through rate, or CTR, tells you what percentage of people who saw something actually clicked it.
If an advertisement receives 10,000 impressions and 300 clicks, the CTR is 3%.
CTR can tell you a great deal about the relationship between your audience and your message.
A weak CTR can indicate several possibilities:
Your creative isn’t grabbing attention.
Your headline isn’t compelling.
Your offer isn’t relevant.
Your targeting is too broad.
Your call to action isn’t clear.
Or the audience simply isn’t interested.
A strong CTR generally tells us something different:
The message created enough curiosity, relevance or interest for somebody to take the next step.
But — and this is important — CTR is not the finish line.
Suppose your advertisement has an incredible click-through rate.
Everybody clicks.
Nobody buys.
Congratulations. You created an interesting advertisement.
You did not necessarily create effective marketing.
That’s why I view CTR as a diagnostic metric, not an ultimate business outcome.
It tells me whether the front end of the marketing journey is doing its job.
Then I look downstream.
Level Five: Cost Per Click
Cost per click, or CPC, tells you how much you’re paying for each click.
Spend $500 and receive 250 clicks?
Your average CPC is $2.
This metric becomes useful when you combine it with CTR and conversion rate.
Imagine your CPC suddenly increases.
Why?
Perhaps competition increased.
Perhaps targeting changed.
Perhaps your creative became less effective.
Perhaps the platform’s auction environment changed.
Perhaps you’re pursuing a more valuable audience.
Now imagine CPC rises 30%, but your conversion rate doubles.
That campaign may actually have improved.
This is why evaluating individual metrics without context can lead businesses to make terrible decisions.
You don’t necessarily want the cheapest traffic.
You want economically productive traffic.
I would rather pay $5 for a click from somebody seriously looking to hire me than 25 cents for a click from somebody who will never become a customer.
Level Six: Website Engagement
This is where Google Analytics becomes particularly useful.
Once someone reaches your website, what happens?
Do they actually engage?
Do they immediately leave?
Do they read?
Do they explore?
Do they visit service pages?
Do they begin taking actions associated with becoming a customer?
Google Analytics 4 defines an engaged session as one that lasts longer than 10 seconds, contains a key event, or includes at least two page or screen views. Engagement rate is the percentage of sessions meeting those conditions, while bounce rate represents the opposite. (Google Help)
That distinction matters because website traffic by itself tells you very little.
Imagine your website receives:
Month One: 2,000 visitors
Month Two: 4,000 visitors
Traffic doubled.
Fantastic, right?
Maybe.
What if engagement collapses?
What if conversions remain exactly the same?
What if nearly all the new visitors arrive from irrelevant searches?
Your traffic graph looks beautiful while your business gained almost nothing.
Now consider the opposite.
Traffic declines from 2,000 visitors to 1,500.
But conversions increase from 40 to 75.
Which month would you rather have?
I’ll take the second one.
Every time.
Engagement Time: Are People Actually Consuming the Content?
Average engagement time can help determine whether people are actually spending meaningful time with your website.
Google describes average engagement time per session as the average amount of time your site was actually in focus in the user’s browser or your app was active in the foreground. (Google Help)
That makes it more useful than simply assuming that an open browser tab means somebody is paying attention.
But context still matters enormously.
If you publish a 3,000-word educational article and average engagement time is eight seconds, something probably isn’t connecting.
If you have a page whose entire purpose is giving someone your phone number and visitors find it in five seconds and call you, a short engagement time isn’t a failure.
Metrics only become meaningful when compared against the purpose of the page.
Landing Pages: Where Are People Entering Your Business?
One of my favorite areas to examine is landing-page performance.
A landing page is essentially the first page someone encounters when entering your website during a session.
Look at:
Traffic.
Engagement.
Conversions or key events.
Conversion rate.
Traffic source.
And, where possible, revenue.
You may discover something fascinating.
Perhaps one blog article you wrote eight months ago generates a steady stream of Google traffic and leads.
Perhaps your beautifully designed homepage gets enormous traffic but converts poorly.
Perhaps one service page converts at three times the rate of another.
Now you have actionable information.
Don’t just say:
“Page A gets more traffic.”
Ask:
Why does Page A perform differently?
Is the headline clearer?
Is the offer stronger?
Is the search intent different?
Is the call to action easier to find?
Does the page load faster?
Does it establish trust more effectively?
Analytics show you where to investigate.
They don’t always tell you the answer automatically.
Traffic Sources: Where Are Your Customers Coming From?
Another incredibly important question is:
Where did these people come from?
Google Analytics acquisition reporting can break traffic down across sources and channels, allowing businesses to compare sessions, engagement, key events and other outcomes. (Google Help)
You might see traffic from:
Organic search.
Paid search.
Organic social.
Paid social.
Email.
Direct traffic.
Referral websites.
Other campaigns.
This is where channel-level performance starts becoming extremely valuable.
Imagine Facebook sends 5,000 visitors and Google Search sends 1,000.
At first glance, Facebook looks like the winner.
But then:
Facebook produces 25 customers.
Google produces 75.
Now the story changes completely.
Facebook may still have value. Perhaps it builds awareness. Perhaps people first discover you there and later search your company on Google.
But now we know something important about purchase intent.
Different channels frequently play different roles in the customer journey.
Attribution: Marketing Is Rarely a Straight Line
This brings us to one of the trickiest subjects in analytics: attribution.
Customers rarely behave like this:
See advertisement.
Click advertisement.
Buy immediately.
Real life is messier.
Someone sees your Facebook post.
Three days later, they see an Instagram advertisement.
A week later, they search your business on Google.
They read two reviews.
They visit your website.
They leave.
Two days later, they receive an email.
Then they come back directly and make a purchase.
Which channel gets credit?
Facebook?
Instagram?
Google?
Email?
Direct?
The answer depends partly on your attribution model.
Google Analytics includes attribution reporting specifically to help businesses understand the different touchpoints customers encounter on their path toward important actions and how different attribution models distribute credit. (Google Help)
This matters because last-click thinking can undervalue the marketing that introduced or nurtured the customer.
For small businesses, you don’t need to become an attribution scientist.
You simply need to remember:
The final click isn’t necessarily the entire story.
Impressions and Reach: Useful, but Easy to Overvalue
Let’s talk about two of digital marketing’s favorite big numbers.
Impressions generally tell you how many times content or advertising was displayed.
Reach generally tells you how many people were exposed to it.
These are useful measurements.
They tell you whether you’re getting visibility.
They help assess awareness campaigns.
They can show whether distribution is expanding.
But impressions are among the easiest numbers to turn into vanity metrics.
“Your campaign generated 600,000 impressions!”
Wonderful.
What happened afterward?
Did anyone click?
Did anyone visit?
Did anyone remember the business?
Did anyone become a lead?
Did anyone buy?
There is absolutely nothing wrong with awareness.
Every customer has to become aware of you before becoming a customer.
The problem comes when businesses mistake visibility for results.
Impressions tell you your message was distributed.
They don’t automatically tell you the message worked.
Social Media Engagement: Useful Signals, Not Revenue
Likes, comments, shares and followers are another category business owners frequently overvalue.
I don’t believe these numbers are meaningless.
Far from it.
A share can dramatically expand distribution.
A comment can demonstrate genuine interest.
Growing followers can increase your owned audience.
Engagement can tell you what topics resonate.
Social analytics are particularly valuable for understanding audience response.
But there is an enormous difference between:
“This post received 500 likes.”
and:
“This post drove 175 website visits, generated 22 email subscribers and produced six qualified leads.”
The second statement connects attention to business outcomes.
That’s where I want to get.
Email Analytics: Don’t Stop at Open Rates
For email marketing, business owners frequently focus heavily on open rates.
They are useful, but I care considerably more about what happens after the email is opened.
Look at:
Open rate.
Click rate.
Click-to-open behavior.
Unsubscribe rate.
Conversions.
Revenue.
Replies.
List growth.
Suppose Email A gets a 45% open rate but almost nobody clicks.
Email B gets a 32% open rate but generates ten purchases.
Which email succeeded?
The answer depends on the objective, but if the objective was sales, Email B clearly produced more of the desired business outcome.
Again, we’re moving from attention toward action.
Ecommerce Businesses Should Watch the Entire Shopping Funnel
For ecommerce businesses, analytics become even more powerful because so much of the purchasing journey can be measured.
Google recommends ecommerce events including product views, add-to-cart activity, cart views, checkout starts, payment-information submissions, purchases, refunds and other shopping behaviors. (Google Help)
That means you can identify where customers disappear.
Imagine:
10,000 people view products.
2,000 add something to their carts.
1,200 begin checkout.
300 purchase.
Something is happening between checkout and purchase.
Now you investigate.
Are shipping costs surprising customers?
Is checkout confusing?
Are payment options limited?
Is the site malfunctioning on mobile?
Is there a trust issue?
Analytics don’t necessarily tell you why people abandoned checkout.
They tell you where the leak is happening.
That tells you where to start looking.
Stop Looking at Analytics Individually
This might be the most important section of this entire article.
Individual metrics can mislead you.
Patterns tell stories.
Here are a few examples.
High Impressions + Low CTR
People are seeing your marketing but not responding.
Investigate:
Creative.
Headline.
Audience targeting.
Offer.
Message.
High CTR + Low Website Engagement
Your advertisement is generating interest, but the experience after the click isn’t meeting expectations.
Investigate:
Message mismatch.
Page speed.
Mobile experience.
Landing-page design.
Content relevance.
High Engagement + Low Conversion Rate
People appear interested but aren’t taking the final step.
Investigate:
Offer strength.
Pricing.
Trust signals.
Calls to action.
Forms.
Checkout friction.
Lots of Leads + Few Customers
You probably have a lead-quality or sales-process problem.
Investigate:
Targeting.
Lead qualification.
Follow-up speed.
Sales scripts.
Offer expectations.
Strong Conversion Rate + High Acquisition Cost
Your funnel may work, but traffic is expensive.
Investigate:
Targeting efficiency.
Bidding.
Channel mix.
Organic acquisition.
Referral opportunities.
Retention.
Low Traffic + Strong Conversion Rate
This can actually be encouraging.
Your website may be doing its job.
You might simply need more qualified traffic.
Traffic Rising + Conversion Rate Falling
Be careful.
You’re growing audience volume, but the incremental audience may be less relevant.
This is a classic example of why growth in one metric doesn’t necessarily mean the business improved.
The Small-Business Analytics Dashboard I Would Actually Use
If I owned a local small business and had only fifteen minutes each week to review marketing performance, I wouldn’t look at 75 metrics.
I would build a simple dashboard around roughly ten questions.
1. How much did we spend?
2. How much revenue did marketing generate?
3. How many leads did we generate?
4. How many of those leads became customers?
5. What was our cost per lead?
6. What was our customer acquisition cost?
7. Which channels generated the most customers — not merely traffic?
8. Which landing pages generated the most conversions?
9. Is our conversion rate improving or declining?
10. Where is the biggest leak in our funnel?
Then I would use supporting metrics — CTR, CPC, engagement rate, traffic, impressions, social engagement and others — to diagnose the answer.
That’s the hierarchy.
Business metrics first. Diagnostic metrics second. Vanity metrics last.
Compare Trends, Not Random Snapshots
Another common mistake is looking at analytics without comparison.
“Website traffic was 4,200 visitors this month.”
Okay.
Compared with what?
Last month?
Last year?
The previous 90-day average?
The same season last year?
Before the campaign started?
Analytics become significantly more useful when you establish context.
I generally like looking at:
Month over month.
Quarter over quarter.
Year over year.
Campaign versus campaign.
Channel versus channel.
Landing page versus landing page.
Device versus device.
New visitors versus returning visitors.
But be careful with seasonality.
A landscaping company in Florida may behave differently from a Christmas retailer.
A tax preparer shouldn’t necessarily compare April with July.
Context matters.
Segment Your Data Before Drawing Conclusions
Averages can hide enormous differences.
Suppose your website conversion rate is 4%.
Sounds useful.
But then you segment the data:
Desktop: 7%.
Mobile: 1.5%.
Now you have discovered something potentially important.
Maybe the mobile site is slow.
Maybe the form is difficult to complete.
Maybe the button is hard to tap.
Maybe mobile traffic simply has different intent.
The point is that the overall average concealed the problem.
You can segment by:
Device.
Location.
Traffic source.
Campaign.
Landing page.
New versus returning visitors.
Audience.
Product.
Service.
Time period.
Analytics become much more powerful when you stop asking:
“What is happening?”
and start asking:
“Where is it happening, and to whom?”
Make Sure You’re Tracking the Right Things
None of this matters if your tracking is broken.
I cannot emphasize this enough.
Before making major decisions from analytics, verify that your important actions are actually being recorded correctly.
Google provides Realtime reporting and DebugView specifically to help verify whether events are firing correctly. (Google Help)
Test your own forms.
Test phone-click tracking.
Test purchases.
Test appointment scheduling.
Test thank-you pages.
Test newsletter signups.
Check whether duplicate events are firing.
Check whether internal traffic is distorting the numbers.
Check your UTM campaign tagging.
Make sure your ad platforms and analytics platforms aren’t measuring completely different definitions of success.
Bad data can create false confidence.
And false confidence is sometimes worse than having no data at all.
Don’t Chase Industry Benchmarks Too Aggressively
Business owners love asking questions like:
“What’s a good CTR?”
“What’s a good conversion rate?”
“What’s a good CPC?”
“What’s a good cost per lead?”
Benchmarks can be helpful.
They give you context.
But your most valuable benchmark is often your own historical performance.
If your conversion rate was 2.5% and you improve it to 4%, that’s meaningful.
If your acquisition cost falls from $300 to $190 while customer quality remains stable, that’s meaningful.
If organic search begins generating 40% more qualified leads year over year, that’s meaningful.
You don’t run the average business.
You run your business.
Your geography, industry, margins, customer lifetime value, competition, sales process, pricing and brand all affect what “good” looks like.
Analytics Should Lead to Decisions
This is where I think businesses sometimes miss the entire purpose of measurement.
Analytics aren’t meant to be admired.
They’re supposed to change behavior.
Every meaningful report should eventually lead to one of four conclusions:
Keep doing this.
Do more of this.
Change this.
Stop doing this.
If your reporting never changes your decisions, you probably aren’t using analytics effectively.
Maybe one advertising campaign consistently generates customers below your target acquisition cost.
Scale it carefully.
Maybe another generates cheap clicks but terrible leads.
Fix it or stop it.
Maybe an organic article keeps producing leads six months after publication.
Create more content around that subject.
Maybe mobile visitors convert terribly.
Fix the mobile experience.
Maybe email customers have dramatically higher repeat-purchase rates.
Invest more heavily in building your list.
The numbers should create questions.
The questions should create experiments.
The experiments should create better numbers.
That’s the cycle.
The Goal Isn’t More Data. It’s Better Decisions.
Digital marketing has given small businesses access to measurement capabilities that previous generations of entrepreneurs could barely have imagined.
A local business can now understand where customers came from, what advertisement they clicked, which page they visited, how they interacted with that page, whether they completed an important action and — with the right setup — whether that activity ultimately produced revenue.
That’s extraordinary.
But having access to data isn’t the same as understanding it.
And understanding data isn’t the same as using it wisely.
The business owner who obsessively watches impressions while ignoring customer acquisition costs isn’t really practicing data-driven marketing.
Neither is the marketer celebrating clicks without asking whether those clicks turned into customers.
The goal isn’t to find the biggest number.
The goal is to understand the relationship between the numbers.
Visibility tells you whether people had the opportunity to see you.
CTR tells you whether your message motivated them to investigate.
CPC tells you what that attention cost.
Engagement tells you whether the experience held their interest.
Conversion rate tells you whether interest became action.
Cost per lead tells you what acquiring opportunities costs.
Customer acquisition cost tells you what acquiring actual customers costs.
Revenue tells you what those customers generated.
Retention and lifetime value tell you what those relationships may ultimately be worth.
Put those pieces together and marketing stops looking like a collection of disconnected charts.
It becomes a story.
A customer saw you.
Something caught their attention.
They investigated.
They engaged.
They considered.
They acted.
They bought.
Maybe they came back.
Your analytics simply help you understand where that story is working — and where the story breaks down.
That is ultimately what good digital marketing analytics should do.
They shouldn’t overwhelm a business owner with numbers.
They should create clarity.
Because the most important question in marketing isn’t:
“How many impressions did we get?”
It isn’t:
“How many people visited the website?”
And it definitely isn’t:
“How many likes did that post receive?”
The question is much simpler:
Did our marketing help accomplish the business objective — and what can the data teach us about doing it better next time?
If your analytics can answer that question, you’re measuring what matters.
Add comment
Comments