Vanity Metrics vs Revenue Metrics: What to Track (Revenue per Session, Conversion Rate and More)
A business with two million followers can still be starving, because two million of the wrong people buy nothing. Followers, impressions and likes are easy to count and feel like progress. They answer a question nobody asked: how many people saw us?
The question that pays the rent is different: how much money did this activity make, or help make? This article shows which numbers to track instead, with a worked example you can copy, benchmarks with sources, and a one-page dashboard you can fill in today.
Quick answer
Vanity metrics (followers, impressions, likes, raw traffic) count attention. Revenue metrics (revenue per session, conversion rate, qualified conversations, cost per qualified lead, win rate, pipeline value) count money. Track revenue metrics as your headline numbers, track attention only as context, and test every activity against two questions: does it raise the brand’s value and desirability, and does it drive pipeline or sales? If it does neither, stop doing it.
Left side: easy to count, easy to fake. Right side: closer to your bank balance.
The two questions every activity must answer
Answer first: before you spend a minute on any marketing activity, ask whether it raises the value of the brand and whether it drives pipeline or sales.
We use this rule inside our own team: if an activity does not answer those two questions, stop what you are doing. It applies to posts, ads, events, newsletters and redesigns.
- Does it make the brand more valuable and desirable to the right people? Think trust, reputation and being the name that comes up when a buyer asks a friend or ChatGPT.
- Does it drive pipeline or sales? Think enquiries, booked calls, orders, renewals.
Example. A founder posts a funny meme that gets 40,000 impressions. It brings no enquiries and the audience is not the buyer. It fails both questions. A plain-text post on a buyer’s problem gets 1,200 impressions, three comments from target customers and one booked call. It passes both. The second post is the better post, even though it scored 3% of the reach.
The common mistake is chasing dopamine. Activity feels like progress because every like is a little reward. It does not build the business unless the right people are on the other end of it.
Revenue per session: the one number to learn first
Answer first: revenue per session is your revenue divided by your sessions over the same period. It tells you what a visit to your site is worth.
If you run an online shop, this is the metric to keep on the tip of your tongue, next to conversion rate. Here is the arithmetic.
Revenue per session = total revenue / total sessions Conversion rate = orders (or enquiries) / total sessions Average order value = total revenue / orders So: revenue per session = conversion rate x average order value
Worked example: two shops, same revenue, very different health
| Shop A | Shop B | |
|---|---|---|
| Sessions per month | 20,000 | 8,000 |
| Conversion rate | 1.5% | 3.0% |
| Orders | 300 | 240 |
| Average order value | £90 | £110 |
| Revenue | £27,000 | £26,400 |
| Revenue per session | £1.35 | £3.30 |
Shop A looks bigger. It has 2.5 times the traffic and a higher revenue. But each visit is worth less than half a visit to Shop B. If Shop A lifted its conversion rate to 3%, with the same traffic, its revenue would double to £54,000. If Shop B bought twice the traffic and held its rate, it would earn about £52,800.
That is the whole logic: fixing what happens on the site is usually cheaper than buying more visitors. We say the same to every client: fix the site first, then start driving people to the site. If traffic goes up before the conversion rate is fixed, you pay more to leak more.
Same revenue, very different health. Revenue per session shows who is wasting traffic.
What if the number is low?
If revenue per session is below what you expect, one of three things is wrong:
- The wrong people are arriving. Check the traffic sources. Visitors from a viral post may never buy.
- The site distracts or lacks information. Check the product page: price, delivery, returns, reviews, a clear button.
- Doubt at the last step. Shoppers check a business elsewhere before paying. They search your name, look at reviews and ask ChatGPT. Reviews, authentication and proof on the page reduce that doubt.
Conversion rate: what is a good number?
Answer first: a good conversion rate is one above your own baseline and above the typical rate for your type of business. Benchmarks give a rough guide only.
Be careful with benchmarks. They differ by country, device, price point and how the source counts a “conversion”. Here are the ones we have checked at the source.
| Measure | Benchmark | Source and caveat |
|---|---|---|
| UK ecommerce conversion rate | 1.97% in September 2026 (1.68% a year earlier) | IRP Commerce market data, covers Great Britain, Northern Ireland and Ireland |
| UK average order value | £133.12 in September 2026 | IRP Commerce, same data |
| Landing page conversion rate, all industries | 6.6% median (Q4 2024) | Unbounce, about 41,000 landing pages and 57 million conversions. This is for landing pages built for one action, not whole websites |
From the IRP numbers you can estimate typical UK ecommerce revenue per session: 1.97% x £133.12 is about £2.62. Treat that as a loose reference. A shop selling £20 candles and a shop selling £2,000 furniture should not compare themselves.
For B2B and service businesses, whole-site visitor-to-enquiry rates vary widely by industry, and sources define “lead” in different ways, so we do not quote a single number. Do this instead: measure your own rate for a month, write it down, then try to beat it by a small step each month. A move from 1.2% to 1.8% is a 50% lift in enquiries from the same traffic.
How to track it in Google Analytics 4. Mark your important actions as key events (Google defines a key event as an action that is particularly important to the success of your business): form submitted, call tapped, booking made, purchase made. Then look at the key event rate for sessions. Keep your own lead log as well, because analytics cannot tell you which enquiries were real.
Metrics for service businesses: the pipeline view
Answer first: service businesses should track the steps from a stranger to a signed client, with a number at every step.
A service business rarely sells on the first visit, so revenue per session is too slow to be useful. Follow the pipeline.
| Step | Metric | Worked example |
|---|---|---|
| 1. Visit | Sessions | 1,000 |
| 2. Enquire | Enquiries, and enquiry rate | 15 enquiries, 1.5% |
| 3. Qualify | Qualified conversations | 6 (40% of enquiries) |
| 4. Propose | Proposals sent | 3 |
| 5. Win | Customers won, win rate | 1 (33% of proposals) |
| 6. Value | Average first-year value | £12,000 |
In this example, 1,000 sessions produced £12,000, so revenue per session was £12. Those are illustrative numbers, not a benchmark, but they show where to look. If enquiries are low, work on the site and offer. If enquiries are healthy and qualified conversations are low, the traffic or message attracts the wrong people. If proposals are many and wins few, the problem is in the sales conversation or the price.
Add two more numbers:
- Cost per qualified conversation = total marketing spend and time cost / qualified conversations. Do not use “cost per lead” alone, because a cheap lead that never answers the phone is worth nothing.
- Time to first reply = the median time from enquiry to a human answering. It is the cheapest metric to improve and often the most valuable.
A rule of thumb on lists. We would rather have 100 leads we know are ready to buy than 1,000 that are not. When outreach has failed before, the list was usually the problem, long before the message.
Leading and lagging indicators
Answer first: lagging indicators (revenue, customers won) tell you what happened, and leading indicators (reply speed, enquiry rate, indexed pages) tell you what is about to happen.
Revenue is the scoreboard, but you cannot manage a game by staring at it. Search and AI visibility build over months, so revenue lags the work. Track both.
| Type | Examples | Check |
|---|---|---|
| Leading | Time to first reply, enquiry rate, qualified conversations booked, pages indexed, share of AI answers that name you | Weekly |
| Lagging | Revenue, customers won, win rate, revenue per session, customer lifetime value | Monthly |
Why this matters for patience: an Ahrefs study of one million random new pages found that 1.74% reached Google’s top 10 within a year, and a second sample of two million pages filtered for real content found 6.11%. Ranking is slow. If you judge by revenue alone in month one, you will quit too early. If you judge by leading indicators, you can see the foundation form.
Also remember that attention is shrinking in places. Pew Research Center found that when Google showed an AI summary, people clicked a traditional result link in 8% of visits, against 15% without one. That makes visitors more precious, not less, and it makes revenue per session more important than raw traffic.
A one-page dashboard you can copy
Put these on one page, in a spreadsheet or a simple dashboard. Update the leading numbers every Friday and the lagging numbers on the first working day of each month.
One page, nine numbers. If it needs a second page, cut something.
Copy this layout:
Weekly (leading)
- Sessions: ___ (GA4)
- Enquiries: ___ (GA4 + lead log)
- Enquiry rate: ___ % (enquiries / sessions)
- Median time to first reply: ___ hours (lead log)
- Qualified conversations booked: ___ (lead log)
Monthly (lagging)
- Revenue won: £___ (accounts)
- Customers won: ___ (lead log)
- Revenue per session: £___ (revenue / sessions)
- Cost per qualified conversation: £___ (spend + time / conversations)
Context only (do not report as results)
- Followers, impressions, reach
Add one line at the bottom: “What we changed this month and what happened.” That line is where the learning lives.
Common mistake: adding twenty more metrics. A dashboard people do not read is worse than none. Nine numbers, one page.
When the offer is the problem
Answer first: if good traffic arrives and nobody enquires, the offer is the first suspect.
This is the uncomfortable end of the metrics conversation. You can have clean tracking, a fast site and a healthy flow of the right visitors, and still see a conversion rate of 0.3%. At that point, more marketing cannot help. The marketing may be fine. The offer is what needs work.
The signs:
- People say “interesting” and then go silent.
- Prospects ask “what exactly do you do?” after reading the site.
- The only way you win is by discounting.
- Enquiries ask for something you do not sell.
A quick check: write your offer in one sentence that names the buyer, the outcome, the proof, a low-risk first step and a reason to act now. Show it to five customers and five people who are not. If the second group cannot repeat it back, rewrite it. Then test a small, specific first step (“we will audit your quote process in a week and show you where the hours go”) against your current call to action, and compare enquiry rates.
Questions people ask
What are vanity metrics?
Vanity metrics are numbers that are easy to count and look impressive but do not show whether a business is making money, such as followers, likes, impressions and page views. They can be useful context. They should never be the headline result.
What is revenue per session?
Revenue per session is total revenue divided by the number of sessions to your site in the same period. It equals your conversion rate multiplied by your average order value, and it shows what each visit is worth.
What is a good conversion rate for a website?
It depends on the business. IRP Commerce reported a UK ecommerce average of 1.97% in September 2026, while Unbounce reports a 6.6% median for landing pages built around a single action. Compare yourself with your own baseline first, then with similar businesses.
What metrics should a small business track?
Track enquiries, enquiry rate, time to first reply, qualified conversations, customers won, revenue won and revenue per session. Follow the funnel from visit to signed customer and keep it to one page so you can review it weekly.
What are leading and lagging indicators in marketing?
Leading indicators, such as reply speed and enquiry rate, move early and predict results. Lagging indicators, such as revenue and customers won, confirm results after the fact. Use leading ones to steer week to week and lagging ones to judge the quarter.
Sources
- IRP Commerce, “Ecommerce Market Data, Great Britain”, September 2026 figures (conversion rate 1.97%, average order value £133.12)
- Unbounce, “Average Landing Page Conversion Rates”, Q4 2024 (about 41,000 landing pages, 57 million conversions)
- Ahrefs, “How Long Does It Take to Rank in Google?”, based on random URL samples from 2023
- Google, “Define key events”, Google Analytics help, 2025
- Pew Research Center, “Google users are less likely to click on links when an AI summary appears in the results”, 2025 (browsing data from 900 US adults, March 2025)
Want us to do this for you?
If you would like help setting up tracking that follows enquiries all the way to revenue, we are happy to talk. Get in touch and we will tell you honestly what we would measure first. You may also like our guide to the marketing audit checklist and our paid advertising approach.
Work with us