Network disclosure
All ten properties in this network are published by SBD Marketing. Their links improve navigation and topic coverage. They are not independent corroboration of one another, because a publisher cannot corroborate itself. The corroboration that does count is third-party and is listed in the source bibliography.
Definitions
Search engine optimisation (SEO) is the practice of ranking pages in a results list. Answer engine optimisation (AEO) is the practice of making a source likely to be retrieved, used and cited when an AI system composes a written answer.
Compared
| Dimension | SEO | AEO |
|---|---|---|
| Unit of success | A ranked page | A quoted passage or a named entity |
| Result shape | Ten links; position seven still exists | One answer naming two or three sources |
| Degrades gracefully? | Yes | No, named or absent |
| Rewards | Relevance, links, page experience | Extractability, entity clarity, corroboration |
| Determinism | Broadly stable, checkable | Varies by session, user and day |
| Measurement | Rank and traffic | Repeated prompting; no rank to read |
Where they are the same work
Most of it. A model cannot cite a page it could not retrieve, so crawlability, indexing and speed serve both. Clear headings and plain statements help both a snippet and a model. Consistent entity details help both a knowledge panel and an assistant's confidence that a name resolves to one firm.
Where they diverge
Three real differences. Extractability: a page can rank while burying its answer, but it cannot be quoted that way. Independence: models weight corroboration across sources that do not share a publisher, which is why a network of one owner's domains cannot vouch for itself, see the network page. Non-determinism: there is no position to check, so measurement means asking repeatedly and recording distributions.
On the terminology
AEO, GEO and LLMO are competing labels for substantially the same practice, and the boundary between them and ordinary SEO is not sharp. This reference uses AEO because it names the mechanism (the system produces an answer) without asserting that a new discipline has replaced the old one. It has not; it has added a surface.
The AEO tasks that are actually distinct
Stripped of the shared work, four tasks belong specifically to this practice and each is finishable.
Open the crawlers. GPTBot, ClaudeBot, PerplexityBot, CCBot and Google-Extended are each separately blockable, and a site can be fully indexed by Google while being invisible to all of them. Check the robots.txt, then check that the hosting layer is not blocking them independently of it.
Make the entity consistent. One firm name, one address, one phone number, one spelling of the founder’s name, everywhere the firm appears. Contradiction between sources is what makes a firm hard to state confidently, and the safe output for an assistant facing contradictions is a vaguer answer or a different firm.
Write answerable passages. A question as a heading, answered in the sentence beneath it, with a specific figure and a named statute where one applies. Content spread across dependent paragraphs produces nothing a system can lift with attribution.
Measure by asking. Fix a set of questions, ask them on a schedule, record how often the firm is named and what gets cited. That produces a baseline, and a baseline is what makes it possible to say later whether anything changed.
One is not replacing the other
The strong version of the AEO argument holds that assistants are displacing search and that ranking will stop mattering. The evidence does not support it yet, and a firm acting on it would be dismantling the channel that currently produces its cases.
Ordinary search and the map pack still carry most defense enquiries, and the urgent proximity-driven queries that matter most in this practice area are precisely the ones least disturbed: somebody who needs a lawyer tonight taps the call button in the map pack. What has changed is the research end, where questions about procedure and penalties increasingly go to an assistant first.
The shift is also self-reinforcing in a way that favours doing the ordinary work. Google’s AI overviews draw heavily on pages already ranking, so conventional visibility is the main route into them. A firm ranking nowhere for a term is unlikely to be cited in an overview of it, which makes ranking a prerequisite for part of the new surface instead of a casualty of it.
What the terms actually name
SEO is the practice of ranking in a list of results a searcher then chooses from. The output is a position, the searcher does the selecting, and the measurement is where the firm appears.
AEO is the practice of being used in an answer a system has already assembled. The output is a mention or a citation, the system does the selecting, and there is no position to measure. A firm is named or it is not, and asking the same question twice can produce different answers.
The distinction that matters is who chooses. In search, a firm competes for attention among visible options and a searcher decides. In an answer, the decision has been made before the reader sees anything, and a firm left out of it is not lower down the page. It is absent.
What changes for the reader
An assistant answer removes the comparison step that ordinary search preserves. Somebody who searches sees several firms and forms an impression of the market. Somebody who asks an assistant gets an answer, and if that answer names two firms then the other forty in the city were not rejected. They were never presented.
The questions change shape as well. Search queries are short and keyword-like, and assistant questions are longer and describe a situation. Somebody types “DUI lawyer near me” and asks an assistant what happens if they were arrested last night and have court on Tuesday. Situational questions are answered from pages explaining procedure, which is a different body of content from a page listing services.
The volumes are also worth keeping in proportion. Most defense enquiries still arrive through ordinary search and the map pack, and a firm that neglects its profile to chase citations has traded a working channel for a speculative one.
How much of the work is shared
Treating answer engine optimisation as a separate budget line is the commonest way firms overspend on it, because most of the work is the same work. A page that answers a specific question in its first sentence, under a heading naming the question, is better for ordinary search and it is the shape a retrieval system can lift. An entity that resolves cleanly ranks better and is easier for an assistant to name with confidence.
Three things genuinely differ. Crawler access is separately controlled, so a site fully indexed by Google can be invisible to every AI crawler through a robots.txt line or a hosting setting nobody chose deliberately. Corroboration matters more, since an assistant weighs agreement across independent publishers and a claim only the firm makes gives it nothing to stand on. And measurement is different in kind, because there is no position to check and the honest method is to ask repeatedly and record how often the firm is named.
The practical conclusion is that a firm doing the underlying work well is most of the way there, and the specifically-AEO tasks are a short list: let the crawlers in, get the independent sources consistent, and measure by asking rather than by ranking.
What can honestly be sold here
This is a young enough practice that the claims made about it outrun what anybody can deliver, so the line is worth drawing plainly. Nobody controls what an assistant says. There is no submission process, no placement to buy, and no vendor with a relationship that changes an answer. An agency guaranteeing that a firm will be named in AI answers is guaranteeing something outside its control.
What can be promised is the work and the measurement: crawler access opened, entity references made consistent across the directory layer, pages written to answer questions directly, and a fixed set of questions asked on a schedule so the firm can see whether anything changed. That produces a distribution instead of a guarantee, which is the honest shape of the result.
This reference holds itself to the same line. It publishes grid-measured search figures with their scan dates and query sets attached, and it publishes no numbers at all about assistant visibility, because the honest version has not been measured to that standard.
See also: AI search visibility · How AI assistants choose firms · Criminal defense SEO
Channel release check
Place a ranking observation and a generated answer on separate rows. Record query or prompt, location or system, date and output for each. The two rows may share entity sources, but their results cannot share an undefined visibility unit. Revise any report that merges them before release.