Spotting Fake, Bought and Doomed Traffic Before You Buy

You can buy 100,000 pageviews for less than the price of dinner. You cannot buy 100,000 people who wanted to be there. Traffic diligence is entirely about telling those apart.

1. Look at the sources, not the total

Pull traffic by source for 12 months. Healthy small sites have a messy mix: organic search, direct, a couple of referrers, maybe social. Warning shapes:

• One referrer over 40% — the business depends on somebody else's editorial decision.

• Direct traffic over 60% on a site with no brand — direct is where bot traffic and mislabelled traffic go to hide.

• Social over 50% from one account — check whether that account transfers. Often it doesn't.

2. Check engagement, not arrivals

Bought traffic bounces. Compare bounce rate, pages per session and average duration against the site's own historical baseline. A month where sessions doubled and average duration halved is a month somebody bought traffic.

3. Cross-check with search data

Ask for read access to search console. Impressions and clicks there should broadly track the organic line in analytics. If analytics shows 40,000 organic sessions and search console shows 6,000 clicks, one of those numbers is wrong, and it isn't the one Google owns.

4. Look for the algorithm cliff

Plot 24 months, not 12. Core updates leave a visible step-change. A site that lost 45% in a single week eight months ago and has been flat since is not a stable business — it's the survivor of an event that can recur.

5. Check keyword concentration

If one keyword cluster drives most of the value, price the business as though that cluster could halve. Sometimes that's still a good deal. It should never be a surprise.

6. Check the geography against the monetisation

A site monetised by US-market advertising whose traffic is 70% from markets with a tenth of the ad rate is not the business the topline implies. Revenue per thousand sessions is the number that reconciles them — compute it and compare it against the category norm.

The one-hour version

1. Analytics: 24 months, by source. Look for cliffs and concentration.

2. Search console: 16 months of clicks. Does it agree with analytics?

3. Engagement metrics, month over month. Any sudden divergence?

4. Top 20 landing pages. Is it one page or a real site?

5. Revenue per 1,000 sessions vs the category norm.

6. Ask directly: "has any traffic to this site ever been paid for or exchanged?" and watch how the answer is phrased.