SEO Expert Witness

What the work actually involves

An SEO expert witness is retained to answer a narrow question in a way a decision-maker with no technical background can follow. The question is almost never was this good SEO. It is usually closer to: was the work described in this agreement actually performed; did this party's conduct cause the traffic decline being claimed; was the ranking that was promised achievable at all; was a competitor's brand term used to divert search traffic, and can that be shown from records rather than assumed from an outcome.

My job is to take material that lives in server logs, analytics exports, archived pages, registrar records, crawl data and search console accounts, and turn it into findings a non-technical reader can weigh - with the uncertainty left in rather than sanded off. That last part matters more than people expect. Search evidence is circumstantial by nature. An opinion that presents it with the confidence of a bank ledger will not survive a competent cross-examination, and it should not.

I work for the retaining attorney. The method does not change depending on which side calls, and I would rather tell you early that the data does not support your theory than discover it in front of someone else.

Where SEO disputes come from

The matters that involve search evidence tend to fall into a handful of recurring shapes:

Performance and contract disputes - whether an agency or contractor delivered what an agreement described, and whether the deliverables produced were the deliverables invoiced.
Claimed interference - allegations that someone manipulated links, content, listings or crawl access to damage another party's visibility.
Trademark use in search - a brand term appearing in paid search, anchor text, page copy, metadata or a competitor's domain.
Asset value - transactions where search visibility was part of what was bought, and the buyer says it was not what it appeared to be.
Post-migration collapse - a rebuild, replatform or domain change followed by a traffic loss, and a disagreement about whose decision caused it.
Reseller, franchise and affiliate conduct - branded bidding, duplicated sites, and channel behavior that violates an agreement.

Different shapes, one common thread: somebody is asserting a causal story about search, and the records either support it or they do not.

What search data can and cannot establish

Start with the hardest limitation. There is no authoritative historical record of search rankings. Search engines do not publish one and never have. Rank tracking tools store samples - observations taken from a chosen location, device and query set, on the days the tool happened to run. If nobody was tracking the term on the dates that matter, no tool can reconstruct it later, and any expert who says otherwise is describing a model, not a measurement.

Results are also personalized, localized and unstable through the day. "Position three" is a statement about one observation, not a property of a page.

Search Console data is better because it is the site's own record, but it carries its own edges: roughly sixteen months of history, average positions weighted by impressions rather than true ranks, and queries below a privacy threshold omitted from the export entirely, which quietly undercounts the long tail.

What the data does establish, and establishes well: that a page existed with particular content on a date; that a site's recorded impressions for a query fell between two dates; that a redirect, a canonical tag or a robots directive changed crawl and indexing access at a known time; that inbound links appeared or disappeared. Those are facts. The ranking narrative built on top of them is an inference, and it should be labeled as one.

Correlation, causation and traffic loss

The most common overreach in this field is treating a timing coincidence as proof of cause. Organic traffic moves for many reasons at once: core algorithm updates, seasonality, a change in how the site is tracked, consent banners suppressing measurement, bot filtering changes, a competitor improving, brand demand falling, a paid campaign ending, or a merchandising change nobody documented.

The way through it is mechanism testing. Every alleged cause leaves a distinct fingerprint, and you can look for it:

If the claim is deindexing, impressions should fall, not just clicks or sessions.
If the claim is a tracking failure, sessions fall while orders in the accounting system hold steady.
If the claim is a manual penalty, there is a message in the property's own console account.
If the claim is a link attack, the link profile should show the acquisition, and the timing should precede the loss rather than follow it.

Then segment - branded against non-branded, page against page, device, country, channel - and find a control: sections of the same site that were not affected, or comparable properties over the same window. A decline that shows up everywhere at once, including in sections the alleged conduct never touched, is telling you something different from what the complaint says.

Sometimes several sufficient causes overlap and the data cannot apportion between them. When that is the answer, that is the answer.

Documenting findings so they hold up

An opinion is only as durable as the record behind it. My working rules:

Preserve before analyzing. Work from copies, keep originals untouched, and record a cryptographic hash of each file at collection so the version relied on can be identified later.
Make every number traceable. Each figure should name its export, account, property, date range, filters, segment and time zone. "Traffic fell 60%" is not a finding until a reader can reproduce it.
Capture pages properly. A picture of a browser window is the weakest form of web evidence. Where I can, I keep the response itself - status code, headers, raw HTML - alongside a full-page capture showing the URL and the system clock.
Prefer scripts to manual steps. An analysis that can be re-run is an analysis the other side can check, which is exactly the point.
List assumptions explicitly. If an opposing expert has to argue with a stated assumption instead of discovering a hidden one, the report has done its job.
Disclose the gaps. Say which data was unavailable and what that prevents you from concluding.

What makes a weak SEO opinion weak

The failure modes repeat, and these are the ones I would go after first in someone else's analysis:

Confidence the source data does not support. Certainty is cheap to write and expensive to defend.
Ranking claims with no cited observation. If there is no dated record of the position, there is no position.
Using today's tool output to describe a past state. Link indexes drop URLs, caches expire, tool datasets get rebuilt. A crawl run this month is evidence about this month.
Ignoring known algorithm activity in the window at issue, which is the first thing a competent opposing expert will raise.
Adopting the client's narrative and then collecting only the data that agrees with it.
Opining on legal conclusions. Whether conduct was willful, or amounted to unfair competition, is not the expert's call. Describe the mechanism and the evidence; leave the legal characterization to the people whose job it is.
Damages arithmetic built on invented inputs - assumed conversion rates and assumed margins that nobody inside the business would recognize.

An expert who has never written the sentence "the available data does not show that" has not been doing the work honestly.

Working with the retaining attorney

Bring the technical analysis in early, because search evidence expires on its own schedule. Raw server access logs are often rotated within weeks. Analytics and console access frequently die with the vendor relationship, and console history simply ages out. A preservation letter written before anyone knows what to preserve tends to miss the things that matter.

When I am consulted early, the most useful thing I do is usually not analysis at all - it is telling you exactly what to demand while it still exists: raw access logs rather than summaries, direct analytics account access rather than exported PDFs, the search console property export, CMS revision history, tag manager container versions, registrar and DNS records, the agency's contract, deliverables and monthly reporting, and any rank tracking account that was running at the time.

Conflicts are cleared before I look at anything substantive. Scope, role and whether the engagement stays consulting or becomes disclosed are your decisions, made under the rules of your jurisdiction, not mine. Consulting and expert engagements are handled through Hartzer Consulting; this site is my professional record rather than a place to hire me.

Frequently asked questions

What can an SEO expert witness actually prove about past rankings?

Less than most people assume, and it is better to say so up front. No search engine publishes a historical ranking record, so a past position exists as evidence only if somebody observed and stored it at the time. What is provable from records is different: that a page held certain content on a date, that the site's own console data shows impressions and clicks for a query rising or falling, that a technical directive blocked indexing from a given moment, and that links appeared or vanished. Those facts often answer the real question without needing a ranking number at all.

Can you show that a competitor caused our traffic to drop?

Only if the mechanism leaves a fingerprint, and only after the alternatives have been tested. I start by asking what the alleged conduct would have had to do technically, then look for that specific signature in the data - impressions versus clicks, affected pages versus unaffected ones, timing to the day or hour where logs allow. I also test the ordinary explanations: algorithm updates in the window, seasonality, tracking changes, consent settings, and the site's own edits. If several causes overlap and the records cannot separate them, my report says that rather than picking the convenient one.

What should we preserve before bringing in a search expert?

Move first on the things with short lifespans. Raw server access logs are frequently rotated within weeks. Analytics and search console access is usually tied to accounts controlled by an agency or a departing employee, and console history ages out at roughly sixteen months regardless of what anyone intends. Also capture CMS revision history, tag manager container versions, the contract and every monthly report delivered under it, registrar and DNS records, and any rank tracking subscription that was running. Preserve access to live accounts, not just exported screenshots - a PDF cannot be re-segmented later.

Do you work for plaintiffs or for defendants?

Both, and the analysis does not change with the side. The method is fixed before the conclusion is known: identify what the records can support, test the alternative explanations, and state the limits. An expert whose findings always favor whoever called first is a liability to the party who retained him, because that pattern is discoverable and it is the first thing a good cross-examination goes after. I would rather deliver an unwelcome early assessment than a comfortable one that falls apart later.

How is an SEO expert different from the SEO agency already on the account?

The incumbent agency is a fact witness with an interest in the outcome, and its reporting was built to demonstrate progress rather than to withstand challenge. My role is independent reconstruction from primary sources - logs, exports, archived pages, registrar and DNS records - with the methodology written out so someone else can run it. That independence is the value. It also means I sometimes conclude that the work at issue was performed competently, which is a finding a retaining attorney needs to hear before it arrives from the other direction.