Key takeaways
- Measure what leads to money: leads, cost per lead, conversion rate, cost per customer and revenue. Treat followers, impressions and clicks as supporting information.
- Set up tracking before spending: analytics with conversion events, Google Search Console, call tracking and a simple record of where each customer came from.
- ROI, ROAS and CAC answer different questions. Use them together, and be clear about which costs and revenues you include.
- Reports should show denominators, time periods and uncertainty. A percentage without the number behind it is not a result.
- Attribution is imperfect. Combine analytics with a simple question to customers: “How did you hear about us?”
Why measurement decides whether marketing is an investment or an expense
Marketing without measurement is a guess with a budget. You may be doing well and not know it, or wasting money and not see it. Good measurement answers four practical questions: what is working, what is not, where should the next dollar go and are we getting closer to our goals?
It does not need to be complicated. A small business can steer well with a handful of numbers reviewed monthly. The difficulty is not in the maths; it is in choosing the right numbers, collecting them reliably and interpreting them honestly. This guide explains the metrics that matter, how to set up tracking, how to calculate return on investment, and how to report without fooling yourself or anyone else. It builds on our small business strategy guide and the funnel and lead generation guide.
Three kinds of metrics
It helps to separate metrics by how close they are to revenue.
| Type | Examples | Use |
|---|---|---|
| Activity metrics | Posts published, emails sent, ads launched, pages created | Check that work is being done; never judge success by them alone. |
| Attention metrics | Impressions, reach, followers, visits, rankings | Early signals of visibility; useful for diagnosing problems at the top of the funnel. |
| Outcome metrics | Leads, calls, bookings, sales, revenue, repeat purchases | The measures that decide whether marketing is paying for itself. |
When a report is full of activity and attention numbers and light on outcomes, ask why. Sometimes outcomes are hard to measure, but often they simply were not set up.
The core metrics most businesses need
Leads
A lead is a person who has taken a step that indicates interest, such as submitting a form, calling, messaging or booking. Count leads by source (organic search, paid search, social, email, referral, direct) and by type. Record quality too: how many were genuine prospects, how many were spam or wrong fit.
Conversion rate
Conversion rate is the share of people who take a desired step. It is useful at several points: visitors to leads, leads to quotes, quotes to customers. Always show the numerator and denominator, for example “18 leads from 640 visitors (2.8 percent).” Small denominators make rates unstable, so avoid over-interpreting a rate based on a handful of events.
Cost per lead (CPL)
CPL is the marketing cost divided by the number of leads in the same period. It is most useful for comparing channels and tracking whether a campaign is getting more or less efficient. A low CPL is not good if the leads never buy, so pair it with cost per customer.
Customer acquisition cost (CAC)
CAC is the total sales and marketing cost to win a new customer, divided by the number of new customers in the period. Decide what to include and be consistent: advertising spend, agency fees, software, and possibly staff time. A narrow CAC that counts only ad spend will look better than a fully loaded one, so state the definition.
Customer lifetime value (CLV)
CLV is the gross profit you expect from a customer over the whole relationship. A simple estimate is average order or job profit, multiplied by how many times a customer buys, over the typical lifespan. Compare CLV to CAC. If you spend $300 to win a customer who brings $900 in profit, the economics are healthy. If CAC exceeds CLV, growth loses money.
Return on ad spend (ROAS) and return on investment (ROI)
- ROAS is revenue attributed to advertising divided by ad spend. A ROAS of 4 means $4 of revenue for each $1 of ads.
- ROI is net profit from marketing divided by marketing cost: (revenue minus all costs) divided by cost. It accounts for margin and for costs beyond ads.
ROAS ignores profit margin, so a high ROAS on a low-margin product can still lose money. ROI is closer to the truth but needs reliable cost and margin data. Use both and state clearly what is included.
A worked example
Consider a home services company that spends $2,500 in a month on paid search and $1,500 on an agency to manage it and improve the site. Over the month, the website receives 3,200 visits from all sources, with 1,100 from paid search.
| Measure | Figure | How it is calculated |
|---|---|---|
| Marketing cost (paid search programme) | $4,000 | $2,500 ads + $1,500 management |
| Leads from paid search | 50 | Calls and forms attributed to paid search |
| Conversion rate (visits to leads) | 4.5% | 50 leads ÷ 1,100 paid visits |
| Cost per lead | $80 | $4,000 ÷ 50 leads |
| New customers | 12 | Leads that became paying jobs |
| Lead-to-customer rate | 24% | 12 ÷ 50 |
| Cost per customer (CAC) | $333 | $4,000 ÷ 12 customers |
| Revenue from those customers | $30,000 | 12 jobs, average value $2,500 |
| ROAS (on ad spend only) | 12.0 | $30,000 ÷ $2,500 |
| Gross profit at 40% margin | $12,000 | $30,000 × 0.40 |
| ROI on the full programme | 200% | ($12,000 − $4,000) ÷ $4,000 |
These numbers are invented to show the method, not to suggest benchmarks. Notice how the story changes depending on the measure: a ROAS of 12 looks spectacular, while ROI at 200 percent, after margin and management costs, is still very good but more realistic. Notice also how many decisions sit behind each number, such as which leads count as attributed to paid search.
Setting up tracking properly
Reliable numbers need a small amount of technical set-up. Doing it before you spend avoids the common situation of having months of spend and no data.
Website analytics (GA4 or an alternative)
- Install the tracking code through a tag manager or directly, and verify that it is firing on every page.
- Define conversion events: form submissions, click-to-call, click-to-message, bookings, purchases. Mark only the meaningful ones as conversions.
- Use consistent campaign tags (UTM parameters) on links from emails, social posts and ads so that sources are identifiable.
- Exclude internal traffic from your own team where possible.
- Respect privacy and consent. Use consent management where required, and describe tracking in your privacy policy.
Google Search Console
Search Console is the owner-side view of how Google sees your site. It shows impressions, clicks and average position for queries and pages, indexing status and technical issues. Verify your site, submit your sitemap and review the performance report monthly. Remember that positions are averages and vary by device and location.
Google Business Profile insights and Bing Webmaster Tools
For local businesses, profile insights show calls, direction requests and website clicks. Bing Webmaster Tools provides similar owner-side data for Bing. Use them as additional views, noting that definitions differ between tools.
Call tracking
Many leads come by phone. A tracking number on your website, different from the one on your listings, can count calls from the site and, with the right tools, from specific campaigns. Keep your main number consistent on listings and profiles, and record that you use tracking numbers so that records stay accurate.
CRM or lead log
Record every lead with its date, source, stage and outcome. Without this, you cannot calculate lead-to-customer rates or CAC. A spreadsheet is a good start. Whatever you use, record revenue against the lead so that ROI can be calculated.
Ask the customer
Analytics cannot see everything. Referrals, word of mouth, offline exposure and cross-device journeys are often invisible. Add a simple question to your intake form and your first call: “How did you hear about us?” Compare the answers with analytics. Differences are information, not errors.
Attribution: giving credit fairly, and knowing its limits
Attribution is the practice of deciding which channel deserves credit for a conversion. A customer might see an ad, read a guide a week later, search for your name and then call. Which touch gets the credit?
- Last-click gives all credit to the final interaction. It favours bottom-of-funnel channels such as branded search.
- First-click gives all credit to the first interaction, which favours discovery channels.
- Data-driven or multi-touch models spread credit using statistical methods. They require enough data and are only as good as the data that feeds them.
No model is perfect. Privacy rules, browser restrictions and cross-device behaviour mean some conversions are never linked to the channel that helped. The practical approach is to use more than one view: platform data for in-platform optimization, analytics for site-wide comparison, and the customer’s own answer as a sanity check. Treat the differences between them as a range, and avoid claiming more precision than you have.
Metrics by channel
| Channel | Primary outcome metrics | Supporting metrics |
|---|---|---|
| SEO and local search | Organic leads, calls and direction requests, revenue from organic | Impressions, clicks and positions for target queries, indexed pages, reviews |
| Paid search | Leads, CPL, cost per customer, ROAS or ROI | Impression share, click-through rate, search terms, quality of leads |
| Social advertising | Leads, CPL, cost per customer | Reach, frequency, click-through rate, creative performance |
| Organic social | Enquiries from social, referral traffic | Engagement, saves, shares, follower quality |
| Email and SMS | Revenue per send, repeat purchases, replies | Delivery, open and click rates, unsubscribe and complaint rates |
| Content and video | Leads influenced, assisted conversions | Engaged visits, scroll depth, watch time, returning visitors |
| Website | Conversion rate by page and device | Speed and stability, error rates, form abandonment |
Measuring visibility in AI answers
A newer question is how often AI assistants and AI search features mention or cite your business. Measuring this is harder than measuring clicks, because answers vary by question, location, device and moment, and some of the process is hidden. Responsible measurement records the conditions, keeps the raw answer and sources, repeats questions, separates mention from recommendation and citation, and shows how many attempts sit behind any rate. Visits from AI tools should be measured separately from citations, using your own analytics. Our AI visibility audit applies this discipline across six AI surfaces, and the GEO guide explains the fundamentals.
How to build a report people will use
A good report is short, honest and tied to decisions. Use this structure each month.
- Headline outcomes: leads, customers and revenue compared with last month and with the same month last year where relevant.
- Efficiency: cost per lead and cost per customer by channel, with the cost definition stated.
- What changed and why: two or three observations with evidence, such as “conversion rate rose after the form was shortened.”
- What we will do next: specific actions, owners and dates.
- Caveats: data gaps, tracking issues, small samples and anything that could mislead.
Always include time periods, denominators and sources. Avoid cherry-picked date ranges and unexplained percentage changes from tiny bases. If a number cannot be measured reliably, say so, instead of filling the gap with an estimate that looks exact.
Using benchmarks carefully
Industry averages for conversion rate or cost per click are tempting, but they vary widely by sector, location, offer and season, and the sources behind them are often unclear. Use benchmarks as a rough sense check, not a target. The most useful comparison is your own performance over time, measured the same way. If you must use external numbers, note the source, date and definition.
Choosing metrics for your type of goal
Different goals call for different primary metrics. Start with the goal and pick the one number that proves it.
| Goal | Primary metric | Guardrail metric |
|---|---|---|
| More enquiries | Qualified leads per month | Cost per lead and lead quality |
| More online sales | Revenue and orders from marketing sources | Profit margin, return rate and ROAS or ROI |
| Higher customer value | Average order value or job value | Conversion rate and customer satisfaction |
| More repeat business | Repeat purchase rate and lifetime value | Unsubscribe rate and review scores |
| Stronger local presence | Calls, direction requests and bookings from the profile | Review rating and response rate |
| Greater search visibility | Clicks and impressions for target queries | Conversion rate of organic visitors |
A guardrail metric stops you from improving one number at the expense of another. For example, you can lower cost per lead by attracting poor-fit enquiries, which is why lead quality sits beside it.
A data quality checklist
Before you trust a report, run through a few simple checks. Most disappointing numbers come from tracking problems, not from marketing problems.
- Is tracking firing everywhere? Test key pages, forms, buttons and checkout on desktop and mobile, and confirm that events appear in your analytics.
- Are conversions duplicated? A thank-you page that reloads can count one lead twice. Compare analytics numbers with real enquiries in your inbox or CRM.
- Is spam filtered? Remove bot and spam submissions from lead counts, or your cost per lead will look better than it is.
- Are time periods and time zones consistent? Compare like periods, and be careful with partial months.
- Are definitions stable? Write down what counts as a lead, a qualified lead and a customer, and do not change it without noting the date.
- Is anything missing? Phone calls, in-store sales and offline referrals often sit outside analytics. Add them manually where you can.
Privacy, consent and responsible measurement
Good measurement respects the people being measured. In Canada, businesses must follow federal and provincial privacy rules such as PIPEDA, and Canada’s anti-spam legislation (CASL) requires consent for commercial electronic messages. In practical terms: tell visitors what you track in a clear privacy policy, use consent tools where required, collect only the data you need, protect it, and honour requests to opt out. Do not send marketing emails or texts to people who have not agreed, and make unsubscribing easy.
These rules change and vary by situation, so confirm your obligations with a qualified adviser. The commercial benefit of responsible practice is real as well: trust is part of your brand, and a reputation for respecting customers makes every other part of marketing easier.
Common measurement mistakes
- Counting everything as a conversion. Page views and scrolls are not leads.
- Mixing definitions. Changing how you count from month to month makes comparisons meaningless.
- Ignoring lead quality. Cheap leads that never buy are expensive.
- Trusting one platform’s numbers. Platforms tend to credit themselves. Cross-check.
- Judging too soon. SEO and content take months, and paid campaigns need enough data.
- Treating a zero or missing value as zero. Missing data may be a tracking failure, not a true zero.
- Reporting without action. A metric that does not inform a decision is clutter.
The quarterly review: turning numbers into decisions
Monthly reports keep you informed, but a quarterly review is where strategy changes. Once every three months, set aside an hour with the people who own the budget and ask five questions. Which channel produced the lowest cost per customer, and is it still scalable? Which channel produced the most customers, and what did it cost to get them? Which stage of the funnel is the weakest, and what is the cheapest fix? What did we test, what did we learn and what will we test next? And what will we stop doing, so that the budget moves to what works?
Write the answers down, with the numbers, and agree two or three priorities for the next quarter. Revisit them at the next review. This simple rhythm is what separates businesses that improve steadily from those that repeat the same activity for years without knowing whether it pays.
A simple monthly dashboard
If you want a starting point, track these ten items in a spreadsheet each month.
- Total leads, split by source.
- Qualified leads and the lead-to-customer rate.
- New customers and revenue from marketing sources.
- Marketing spend by channel, including fees.
- Cost per lead and cost per customer by channel.
- Website conversion rate overall and on mobile.
- Search Console clicks and impressions for core queries.
- Google Business Profile calls, directions and website clicks.
- New reviews and average rating.
- Three notes: what we learned, what we changed and what we will change next.
Smartise can set up tracking, build a dashboard and report on it in plain language as part of our digital marketing and SEO work. To discuss measurement for your business, contact us.
Frequently asked questions about marketing metrics and ROI
What are the most important digital marketing metrics?
For most businesses the most important are leads, conversion rate, cost per lead, cost per customer and revenue from marketing sources. Impressions, followers and rankings are useful supporting signals but do not prove that marketing is paying for itself.
What is the difference between ROI and ROAS?
ROAS is revenue attributed to advertising divided by ad spend, and ignores margin and other costs. ROI is net profit from marketing divided by its total cost, so it accounts for margin and fees. Use both and state exactly what each includes.
How do I calculate customer acquisition cost?
Divide total sales and marketing costs for a period by the number of new customers acquired in that period. Decide in advance which costs to include, such as ad spend, agency fees, software and staff time, and keep the definition the same over time.
How do I know which channel brought a customer?
Use analytics with tagged links and conversion events, call tracking for phone leads and a CRM record of every lead’s source, and ask customers how they heard about you. No single method is perfect, so compare them and treat differences as a range.
How often should I review marketing results?
Review paid campaigns weekly at first, then fortnightly or monthly once stable. Review overall results and strategy monthly, and review SEO and content on a quarterly rhythm, because they take longer to show effects.
Why do numbers in Google Analytics and Google Ads differ?
They use different attribution rules, time windows, conversion definitions and data collection methods, and consent settings can affect what each can see. Compare them for direction, not exact equality, and choose one source of truth for each decision.
Can Smartise set up tracking and reporting for me?
Yes. Smartise sets up analytics, Search Console, call tracking and lead logging, builds a simple dashboard and reports monthly in plain language with clear definitions and caveats. Contact us to discuss your needs.
More digital marketing guides
Small Business Digital Marketing Strategy
A practical plan to choose goals, audience, channels and budget, then execute and measure.
Read more about Small Business Digital Marketing Strategy →Local SEO for London, Ontario
How to win local search: profile, citations, reviews, local content and technical SEO.
Read more about Local SEO for London, Ontario →Marketing Funnel & Lead Generation
Build a funnel from awareness to conversion, with landing pages and follow-up that work.
Read more about Marketing Funnel & Lead Generation →Ready to grow with Smartise?
Tell us about your business and goals. We will recommend the smallest set of changes that makes the biggest difference, and tell you plainly what we can and cannot promise.