Skip to main content
Growth8 min read

Stop optimizing for vanity traffic

Visits feel like progress. A source can send hundreds of visits and no sale, and a dashboard that ranks by traffic will put it near the top.

Vicente TorresFounder of Vero

It is easy to celebrate a traffic spike. It is much harder to notice when that traffic never turns into money, because most dashboards are built to make traffic look like the goal. This post shows what vanity traffic is, how to find it with one division, and what to do once you have found it, using the numbers from Vero’s public demo.

What is vanity traffic?

Vanity traffic is a source that sends real, human visits and no revenue. A link that spreads, a low-intent referral or a campaign aimed at the wrong audience can all produce a visit count that looks good in a weekly report, with nothing behind it when the payments come in.

On a dashboard that ranks sources by visits, these sources look like wins. They sit near the top of the list, they move the line on the chart, and they make the week look busy. On a dashboard that ranks sources by money, the same sources stand out for the opposite reason: a large visit count with a zero next to it.

In Vero, revenue by source answers that question on Pro, and the Revenue screen has a panel called Vanity traffic: the sources sending real traffic and zero revenue, listed as candidates to cut or rework. The rest of this post is about reading that list well, because a zero can mean several different things.

A worked example from the public demo

Vero’s public demo runs on simulated data for a fictional SaaS. It is a product demo, not anyone’s real traffic, but its numbers reconcile with each other, so it is a useful place to practice. It tells the story of a launch week: 14,724 visitors, 187 sales and €18,423 of revenue in seven days.

Most of that money came from two sources. news.ycombinator.com sent 3,420 visits and €7,124 of revenue. twitter sent 1,245 visits and €4,821 from 49 payments. producthunt, the other launch-week spike, sent 1,632 visits and €1,845 from 19 payments.

Then there is reddit. In the demo, reddit sent 421 visits and no sale. Those visits came from ads, and the demo gives them €1,150 of spend for the week, so reddit’s profit is -€1,150. On a chart of visits, 421 is a respectable line, more than the 240 visits from chatgpt, which brought €400. On a chart of money, reddit is the source that lost the most and returned nothing.

The sponsored newsletter is the quieter version of the same problem. It did bring sales, €482 of them, against €900 of demo spend, so it reads -€418. It never appears as vanity traffic, because its revenue is not zero. That is why the list of zeros is where the analysis starts, not where it ends.

Revenue per visitor: one division that ranks every source

Raw visits reward whatever is loudest. To compare sources fairly, divide each source’s revenue by its visits. That gives revenue per visitor: how many euros, on average, one visit from that source was worth over the period you are reading.

Applied to the demo, the division reorders the list. The figures below are derived from the revenue and visits the demo states, rounded to the cent. The whole site earns €1.25 per visitor (€18,423 / 14,724). twitter earns €3.87 (€4,821 / 1,245), news.ycombinator.com €2.08 (€7,124 / 3,420), producthunt €1.13 (€1,845 / 1,632) and google.com, the organic search traffic, €0.44 (€1,000 / 2,264). reddit earns €0.00.

That ranking says something visit counts hide. producthunt sent more visits than twitter, but each of its visits was worth less than a third of a twitter visit, and slightly less than the site average. It matches the conversion rates the demo states: the site converts to paid at 1.27%, twitter at 3.9% and producthunt at 1.16%.

Revenue per visitor also puts the spend in the same unit. reddit’s €1,150 bought 421 visits, about €2.73 each (derived: €1,150 / 421). Even if those visits had converted like the site average, at €1.25 each, they would have brought in about €527 (derived: 421 × €18,423 / 14,724), less than half of what they cost. This is not bad luck in one week. Each visit costs more than an average visit is worth.

What counts as a source, and what counts as revenue

The division is only as good as its two inputs, so it helps to know exactly what each one is. In Vero, the source of a visit is its first-touch utm_source. Without one, it is the host of the referrer, so a visit from https://www.google.com/search counts as google.com. Without either, it is direct. A referrer on your own site or one of its subdomains counts as no referrer.

Revenue is what was charged, without tax, less refunds, before fees. A refund comes off the source that made the sale, so a campaign that sells and then gets refunded is not credited for money you gave back. Provider fees are not deducted, because Vero does not know them.

Each sale is tied to the visit that brought it when the sale is received, and keeps that visit’s source whatever date range you read. The mechanics are in how revenue attribution works without cookies: the visit id travels with the checkout, the provider’s signed webhook brings it back, and the sale is credited to the first touch of that visit.

Direct is never listed as vanity traffic. It is what is left when a visit has no tag and no referrer: bookmarks, typed URLs, links opened from an email client or a chat app. You cannot cut it or rework it as a channel, so a zero there tells you about your tagging, not about a source. In the demo, direct sent 3,535 visits and no sale, and it stays off the list by that rule. A channel that is small but worth reading is AI assistant referral traffic.

Read small numbers with care

A zero over one week is a signal, not a verdict. At the demo site’s 1.27%, 421 visits would be expected to bring about five sales (derived: 421 × 1.27% ≈ 5.3). When you expected five, a zero is unlikely to be noise. The same zero on 60 visits, where the site rate predicts less than one sale, would tell you nothing.

The same care applies to sources that did sell. In the demo, chatgpt brought €400 from 3 payments over 240 visits, about €1.67 per visitor (derived: €400 / 240). One sale more or one sale less would move that figure by roughly a third. A per-visitor figure built on a handful of sales is a hint to watch, not a ranking to act on.

Two habits help. Read a longer range before cutting anything you do not pay for, because a month smooths out a launch spike. And look at the number of sales next to the revenue: €400 from 3 sales and €400 from 40 sales are not the same evidence.

Journeys that span several visits matter too. Vero credits a sale to the first touch of the visit in which the customer paid. Someone who arrives from a source today, comes back next week by typing your address and pays then is credited to direct, not to that source. So a source that opens long decisions can read zero while its visitors buy later through another door. If your customers usually take more than one visit to buy, keep that in mind before you call an unpaid source vanity traffic.

What should you do with a source that sends visits and no sale?

Once a zero survives a longer range and a reasonable sample, there are three honest answers: cut it, rework the page it lands on, or keep it as awareness and say so out loud.

Cut it when you pay for it and the arithmetic does not close even at your average. reddit in the demo is that case. To cover the €2.73 each visit costs, at the demo’s average order of €98.52, it would have to convert at about 2.8% (derived: €2.73 / €98.52), more than twice the site’s 1.27%. Stopping the spend is worth €1,150 a week in the demo’s numbers.

Rework it when the audience is right and the page is wrong. A paid campaign that lands on the home page instead of the page it advertised, an offer that does not match what the ad promised, or a price that audience never gets to see can all turn interested visits into zero sales. Change the landing page or the offer, tag the new links with their own utm_campaign, and read revenue per visitor for that campaign on its own.

Accept it as awareness when it costs nothing, or when you have decided to pay for reach. A free community thread that sends visits and no sale is not hurting you. Keep it out of the figures you use to judge acquisition, and write down why it stays, so the next review does not have to rediscover the reason.

Compare against what you spend

Revenue tells you which sources sell. Profit tells you which ones are worth paying for. In the demo, google, the brand_search ads, brought €1,783 against €1,400 of spend, so it kept €383 (derived). The sponsored newsletter brought €482 against €900. Neither is vanity traffic, and only one of them made money.

On Pro, the profit table takes what you spend each month per source, prorates it to the period on screen, and shows the revenue per visit and the profit each source leaves. Profit by source after ad spend walks through that view with the same demo week.

If you run several campaigns at once, read the same idea per campaign: analytics for marketing teams covers revenue per UTM campaign and moving budget from vanity traffic to the campaigns that pay.

Unsure of a term? Read the glossary

Vicente Torres
Vero is built by one person in Spain. I read every waitlist entry and pick the beta accounts myself.
Vicente Torres, Founder of Vero LinkedIn
Hand-picked beta, waitlist open

Ask for a place in the beta.

Anyone can join the list. The accounts we pick get Pro free for the whole beta.

By joining you agree to receive launch updates and accept our Privacy Policy.

See the demo