Check this page with an assistantOpens a chat asking it to summarise this article and name the evidence behind each claim.
Claude opens with the prompt on your clipboard: Anthropic does not support prefilled prompts on the web, and we would rather copy it than ship a button that drops it.
You Are Buying the Traffic, Not the Site
For a business whose revenue arrives through search, the website is a delivery mechanism and the traffic is the asset. Most acquisitions verify revenue carefully and take the traffic on trust, which inverts the risk: the revenue is a consequence, and the thing producing it is what could disappear.
The failure mode is specific. A buyer confirms twelve months of consistent profit, agrees a multiple, completes, and then watches the traffic decline for reasons that were visible in the data all along: concentration in two queries, a step-change nobody could explain, or a domain with a history.
Everything here is data the seller already has and can grant access to in minutes. A seller who will not is either disorganised or hiding something, and both are material. Read-only access to the accounts is a reasonable condition of proceeding, not an imposition.
You can explain, from data you pulled yourself, where the traffic comes from and what would have to happen for it to stop.
What to Demand, and Why Screenshots Are Worthless
Exported images and PDFs can be edited in a minute and routinely are. Insist on read-only access to the live accounts and pull everything yourself. If that is refused, the diligence has already produced its most useful finding.
- Search Console, read-only, sixteen months.
The most important single item. It reports from Google's side rather than from a tag on the site, which makes it considerably harder to manipulate than analytics.
- Analytics, read-only, same period.
For behaviour and conversion. Expect it to disagree with Search Console, which is normal and explained in why the two disagree. A gap that is wildly out of the ordinary is worth asking about.
- Revenue exports, reconciled.
Affiliate network, ad network, or payment processor data directly, not a spreadsheet. Then check it reconciles against the traffic pattern rather than merely existing.
- Domain registration history.
What the domain was before. A previously expired domain with a life in an unrelated industry is a materially different asset from one that has always been what it is now.
- Any contracts that carry the revenue.
Affiliate agreements, advertiser deals, and anything terminable on change of ownership. Traffic that survives the sale is worth nothing if the contract monetising it does not.
Reading the Traffic Honestly
Pull queries and pages over sixteen months and look for the shape rather than the total. What you are testing is whether the traffic is explainable: does its growth correspond to things the site did, and is it spread across enough surface to be resilient?
| Check | Healthy | Investigate |
|---|---|---|
| Traffic trend | Gradual, with explainable movements | A step-change with no content or link event behind it |
| Query distribution | Revenue spread across many queries and topics | Most of it resting on a handful of terms |
| Page distribution | Several pages carrying meaningful traffic | One page carrying most of the business |
| Indexed pages vs real pages | Roughly comparable | Index bloat, per duplicate content |
| Branded vs non-branded | Both present and growing | Almost entirely branded, meaning no discovery |
| Recent core updates | Rode them without much movement | Fell at each one; see page-level authority |
The step-change row is where purchased links usually reveal themselves. Traffic that rose sharply without a corresponding publication or coverage event rose because something was bought, and bought advantages are reversible in a way the price rarely reflects.
Separating Commercial Demand From Volume
A site can show impressive growth entirely from informational queries asked by people who will never buy. That volume is real and its commercial value is close to zero, which makes it the most common way an asking price gets inflated honestly.
Segment the query export by intent and see where the money actually comes from. If commercial and transactional queries are flat while informational volume tripled, you are buying a traffic chart rather than a business, and the multiple should reflect the commercial segment rather than the total. The classification method is in search intent.
Then check the reverse: revenue per session by page. A page carrying a fifth of the traffic and none of the revenue is a liability in maintenance terms and a distraction in valuation terms. What you want to establish is which pages are the business, and whether those specific pages are durable, which is a much narrower question than whether the site is.
Concentration and the Domain's Prior Life
Two risks change what you are buying rather than merely reducing its value. Revenue concentrated in a few queries can vanish in a single update. A domain with a history in another industry carries that history whether or not the seller mentions it.
Concentration is measurable and should be priced explicitly. Rank the queries by attributable revenue and calculate what share the top five represent. A site where they carry most of it is a different asset from one where they carry a fifth, and the difference is not a discount, it is a different risk category.
History requires actively looking. Check archived versions of the domain, look for a prior industry, and treat any evidence of purchased links or a previous penalty as material. Buying a domain with a past means inheriting it, and the relevant background on how that surfaces is in the link spam and disavow guide and the site reputation crackdown.
Turning Findings Into a Number
Diligence produces adjustments, not a verdict. Almost nothing found here makes a site unbuyable; it makes it worth a different amount, and the useful output is a list of specific risks with a price attached to each.
Three questions convert findings into terms. What would it cost to fix, for issues like index bloat or thin content that are remediable. What share of revenue is exposed, for concentration risk. And what is genuinely unfixable, such as an arrangement that policy targets structurally rather than a quality problem you can improve.
The third category is where buyers get hurt, because it looks like the second. Content that declined because it was thin can be improved; content that declined because the arrangement itself is what policy targets has no editorial route back, and paying for it as though it were a fixable quality problem is the most expensive mistake available in this category.
Questions Buyers Ask
- What should I ask for before buying a website?
Read-only Search Console and analytics access covering at least twelve months, ideally sixteen so you can see full seasonality. Also the revenue exports for the same period, domain registration history, and any affiliate or advertiser contracts. A seller unwilling to grant read-only access to their own data has told you something important.
- How do I verify traffic claims?
Never from a screenshot or an exported PDF, both of which are trivially edited. Insist on read-only access to the accounts themselves and pull the data yourself. Search Console is the harder of the two to manipulate because it reports from Google's side rather than from a tag on the site.
- What is vanity traffic?
Volume from queries that will never buy anything. A site can show impressive growth built entirely on informational searches from people outside the customer profile, which inflates the apparent value while contributing nothing to revenue. Separating commercial from informational demand is frequently the difference between the asking price and the real one.
- What are the biggest red flags?
Revenue concentrated in a handful of keywords, a traffic step-change with no corresponding content or link explanation, index bloat far exceeding the real page count, a domain with a prior life in a different industry, and any history of purchased links. Each of those changes what you are buying rather than merely reducing its value.
- Does an algorithm update risk make a site unbuyable?
No, it makes it differently priced. Every site dependent on search carries that exposure. What matters is whether the traffic is diversified across many queries and topics or concentrated in a few, because the first survives an update as a dip and the second can disappear in a week.
