Skip to main content
Visionnment Estimate value
Menu

Buy & Due Diligence

How to Verify Website Traffic Before an Acquisition

Verify website traffic before buying with read-only analytics, Search Console, channel and page checks, bot review, seasonality, and revenue reconciliation.

A magnifying glass inspecting an anomalous segment in a website traffic graph

To verify website traffic before buying, replace screenshots with controlled source access, compare systems that measure different parts of the journey, and explain the asset’s concentration, seasonality, anomalies, and connection to revenue.

The objective is not to make every report match. Web analytics, search reporting, server delivery records, and financial systems use different scopes and rules. The objective is to decide whether the evidence tells a coherent story—and whether that story still works after the buyer takes control.

This is an acquisition-review method, not a forensic audit or a guarantee of future traffic. Use technical, financial, legal, or privacy specialists when the deal and data require them.

The short answer

A credible traffic review should answer:

  • Which accounts and properties generated the reports?
  • What changed in measurement during the review period?
  • Which channels, pages, queries, geographies, and devices drive the audience?
  • How concentrated is the asset?
  • Which changes are seasonal, technical, algorithmic, promotional, or unexplained?
  • Does measured attention correspond with leads, sales, advertising, or affiliate records?
  • Could automated or fabricated events be distorting the story?
  • What access and measurement setup will transfer?

If the seller will not provide reasonable read-only evidence for a material claim, record that as an unresolved acquisition risk rather than estimating around it.

What to request before the review

Request source access with the least privilege that still allows investigation. Google Analytics supports Viewer and Analyst roles as well as higher-permission roles; Google’s access-management documentation defines the current capabilities. During diligence, viewing data should not require control over users or settings.

Ask for:

  • read-only web-analytics access;
  • verified Search Console access appropriate to the review;
  • server, host, or CDN request exports where available;
  • advertising, affiliate, commerce, lead, or subscription reports;
  • a record of tagging, consent, migration, and site changes;
  • the sitemap and indexable URL inventory;
  • campaign calendars and material referral arrangements;
  • property, stream, time-zone, currency, and attribution settings; and
  • monthly revenue and operating records for reconciliation.

Flippa’s current due-diligence checklist recommends full Google Analytics read-only access rather than relying only on marketplace-displayed figures. That is useful first-party marketplace guidance, not proof that the data is accurate.

Step 1: preserve the measurement context

Before filtering reports, capture a measurement cover sheet:

Field Why it matters
Property and stream identifiers Confirms you are reviewing the claimed asset
Time zone and currency Changes date boundaries and monetary reporting
Data retention and start date Defines what history is actually available
Internal/developer traffic rules Can change reported audience
Consent and privacy configuration Can change coverage over time and by region
Tagging method and measurement ID Exposes duplicate, missing, or replaced implementations
Key-event definitions Shows what a reported “conversion” actually means
Attribution/reporting changes Prevents false before-and-after conclusions
Site migrations and redesigns Provides alternative explanations for traffic shifts

Capture change dates and annotate the monthly trend. A break in tracking is not a traffic decline; a new duplicate tag is not growth.

Do not ask the seller to remove filters, change tags, or alter users merely to make the report easier to read. Preserve evidence during diligence and test changes in a separate environment where possible.

Step 2: choose comparison periods that expose seasonality

Review daily and monthly data. A short, recent window can hide a seasonal decline or exaggerate one campaign.

When history exists, use:

  • trailing 12 months for a complete operating cycle;
  • at least 16 months for clearer year-over-year comparison;
  • same-month or same-week comparisons for seasonal assets; and
  • separate views before and after known measurement or site changes.

Google Search Central’s current guide to debugging Search traffic drops recommends using 16 months of data to identify seasonality and comparing equivalent periods. It also distinguishes algorithmic, technical, security, spam, migration, demand, and seasonal causes.

Do not annualize the strongest month unless a defensible model explains why it represents the next 12 months.

Step 3: reconcile analytics, search, delivery, and outcomes

Build a monthly reconciliation table:

Month Analytics sessions Search Console clicks Server/CDN requests or sessions Leads/sales/actions Revenue Explanation
Month 1
Month 2

These columns should move in ways you can explain, but they should not be equal:

  • Analytics records instrumented user activity under its collection, consent, and processing rules.
  • Search Console reports activity from Google Search and applies privacy and reporting limits.
  • Server or CDN records capture delivered requests, including activity that may be automated or may not run analytics code.
  • Revenue systems record financial events on their own recognition and payout timelines.

Google Analytics explains that traffic-source dimensions exist at first-user, session, and event scopes. Comparing “first user source” with “session source” can therefore answer different questions.

Search Console’s performance filtering guidance notes that anonymized queries are omitted from tables and that rows and filters can affect displayed totals. Avoid presenting the visible query rows as a complete keyword inventory.

Investigate material divergence. Examples include missing tags, consent changes, timezone boundaries, crawler load, offline conversions, cross-domain tracking, returns, payout lags, and genuinely different scope.

Step 4: measure channel concentration

Export monthly sessions or comparable traffic by default channel group and source/medium. Keep paid, organic, direct, referral, email, social, and unassigned activity separate before combining anything.

Calculate simple concentration measures:

top-channel share = sessions from largest channel / total sessions

top-source share = sessions from largest source-medium pair / total sessions

Repeat by month. A yearly average can hide a sharp late-period dependency.

Then inspect the quality of each material channel:

  • Does its landing-page mix make sense?
  • Do geography, device, and engagement patterns fit the audience?
  • Does the channel create the actions or revenue claimed?
  • Is it dependent on the seller’s personal brand, another owned site, or a contract?
  • Is the acquisition method documented and permitted?
  • Will the account, creative, list, relationship, or referral continue after transfer?

Do not treat “direct” as proof of brand demand. Direct can contain unattributed visits. Do not treat “organic” as durable merely because it is unpaid.

Step 5: measure page and query concentration

Export landing pages from analytics and pages and queries from Search Console. Review absolute performance and share.

top-page share = sessions landing on highest-traffic page / total sessions

top-5-page share = sessions landing on five highest-traffic pages / total sessions

For search traffic, compare top pages with their material queries. Search Console allows comparisons and filters by page, query, country, device, and search type. Use those views to determine whether a “diverse” site actually depends on several pages answering the same underlying demand.

For high-concentration pages, record:

  • intent and subject;
  • publication and substantive-update dates;
  • ranking query groups;
  • link and internal-link support;
  • monetization role;
  • evidence and author requirements;
  • recent trend; and
  • likely consequence of a decline.

Concentration is not automatically bad. It is a risk that should change diligence depth, downside modeling, and price.

Step 6: segment geography, device, and audience behavior

Review countries, regions where reliable, devices, browsers, landing pages, and session timing. Compare each segment with the business model.

Questions to ask:

  • Does traffic come from the markets used in the revenue model?
  • Are sudden geographic shifts explained by campaigns or demand?
  • Does device mix fit the page experience and conversion path?
  • Are important pages usable on the dominant device?
  • Do new and returning audience patterns support the seller’s story?
  • Are referral hostnames or language settings plausible?

Avoid declaring an entire geography invalid simply because its monetization is lower. The issue is whether the audience is genuine, relevant, economically understood, and likely to remain.

Step 7: investigate bot, spam, and event-manipulation signals

There is no universal bounce-rate, engagement-time, or geography threshold that proves bot traffic. Look for combinations of evidence:

  • abrupt spikes without a credible campaign, referral, or search explanation;
  • repeated visits to a narrow set of paths at regular intervals;
  • high request counts with little corresponding instrumented activity;
  • improbable geography, language, device, or browser clusters;
  • sudden key-event growth without downstream leads, orders, or revenue;
  • session growth isolated to one source/medium label;
  • near-identical event patterns or payload values; and
  • server, analytics, Search Console, and financial trends that diverge materially.

Google’s Measurement Protocol permits server-to-server event collection. Its protocol security guidance warns that a publicly exposed API secret can allow unauthorized parties to send spam events to a property. This does not prove that a suspicious property was manipulated; it explains why analytics access alone is not conclusive.

Review tag and protocol implementations, secrets handling, bot controls, hostnames, and server records with a qualified technician when the signal is material. Preserve logs lawfully and minimize access to personal data.

Step 8: explain every material rise and fall

Overlay the traffic trend with:

  • publishing and update history;
  • Google Search changes and manual-action/security records;
  • site migrations, redirects, outages, and template changes;
  • tracking and consent changes;
  • paid or partner campaigns;
  • newsletter sends and promotions;
  • news, seasonality, and demand shifts; and
  • changes in products, prices, inventory, or monetization.

Classify each material movement as supported, plausible but incomplete, or unexplained. Save the source evidence behind the classification.

Google cautions against making radical changes in response to a small Search decline and recommends examining pages, queries, countries, devices, and search types. Use that discipline during diligence: a seller’s explanation should be testable against the affected segments.

Step 9: reconcile traffic with monetization

Traffic has acquisition value only if its relationship to the business model is understood. Build a route map:

source → landing page → meaningful action → revenue platform → payout or receipt

For display advertising, compare sessions, page views, ad reports, geography, and receipts. For affiliate sites, compare outbound clicks where tracked, merchant reports, reversals, and payouts. For lead generation, compare form or call records with accepted leads, invoices, and collections. For commerce, compare product and checkout events with orders, refunds, fees, and bank deposits.

The numbers will often post on different dates. Document the lag and reconcile cohorts or transaction identifiers where appropriate.

Use the dedicated revenue, expenses, and add-backs verification guide for the financial side. Traffic growth without corresponding business outcomes requires an explanation, not a higher valuation assumption.

Step 10: test transferability of the traffic system

Traffic can be genuine and still be non-transferable. Identify dependencies on:

  • the seller’s name, credentials, appearances, or social accounts;
  • email lists or communities with unclear consent or ownership;
  • referral arrangements and links controlled by the seller;
  • paid-ad accounts, pixels, creative, or historical audiences;
  • authors, editors, or experts who will not continue;
  • analytics and Search Console accounts the seller controls; and
  • domains, subdomains, redirects, apps, or feeds outside the transaction.

For Search Console, a verified owner has the highest permission level. Google’s ownership guide explains verification methods and removal of prior owners’ tokens. The closing plan should establish buyer verification and then remove seller access and stale tokens after acceptance.

For each dependency, require a transfer route, buyer prerequisite, timing, test, and contingency.

Build a traffic evidence scorecard

Use pass, review, and stop—not a misleading single certainty score.

Area Pass Review Stop example
Source access Appropriate property access and identity confirmed Partial history or unresolved scope Material claim supported only by screenshots
Measurement continuity Changes logged and effects explainable Known gap can be bounded Seller cannot explain major tracking break
Channel/page concentration Within prewritten buy box Higher than planned but priceable One dependency violates stop rule
Trend and seasonality Comparable periods support the story Mixed evidence needs more history Claimed growth disappears under equivalent comparison
Automation/spam Signals plausible across systems Anomaly needs technical review Material traffic appears fabricated or unverifiable
Outcome link Attention connects to source revenue evidence Lag or attribution needs reconciliation Traffic and claimed revenue have no supportable path
Transferability Accounts and dependencies have tested paths Vendor or partner confirmation pending Critical source cannot transfer or be replaced

The website due-diligence checklist can combine this result with financial, rights, technical, operational, and transfer findings.

Common traffic-review mistakes

  • Trusting screenshots: they omit account identity, filters, roles, and surrounding history.
  • Comparing mismatched metrics: users, sessions, clicks, requests, events, and orders are not interchangeable.
  • Using only the recent 28 or 90 days: short windows conceal seasonality and tracking breaks.
  • Reading totals without concentration: a stable total can hide the collapse of a valuable page and growth in low-value traffic.
  • Calling every mismatch fraud: different systems legitimately measure different scopes.
  • Calling every visit valuable: relevance, outcomes, and transferability matter.
  • Changing the property during review: evidence should remain stable until the transaction plan authorizes changes.
  • Ignoring privacy: diligence access should be least-privilege and appropriate for the data reviewed.

The deliverable is an explanation, not a dashboard

Finish with a short evidence memorandum that states:

  1. the properties, date ranges, roles, and settings reviewed;
  2. material measurement changes and gaps;
  3. channel, source, page, query, geography, and device concentration;
  4. trend and seasonality conclusions;
  5. anomaly and automated-traffic findings;
  6. the connection between traffic and business outcomes;
  7. transfer dependencies;
  8. unresolved requests; and
  9. pass, review, or stop status.

The goal is not to prove that a traffic chart will continue. It is to know what produced it, what could break it, and whether the purchase still fits your downside plan when uncertainty is stated plainly.