Coach Mike, Michael Reyes, and M. Reyes, CSCS Read as Three Different People to Google
A strength coach in Redondo Beach has appeared on fourteen podcasts in the last two years. He's quoted in three fitness publications, has a book listed on Amazon, and holds a CSCS certification through the NSCA. By any reasonable measure, he's an authority in his niche. Ask ChatGPT or Perplexity who the top strength coaches in the South Bay are, though, and his name doesn't come up. A competitor with half his media footprint gets cited instead.
The difference isn't output. It's identity. His podcast appearances list him as "Coach Mike." His LinkedIn says "Michael Reyes." His NSCA certification record, which shows up in a handful of directory searches, says "Michael A. Reyes, CSCS." His book cover says "Mike Reyes." Four data points, four slightly different names, zero explicit statement anywhere on the internet that these are the same person. To a human reader, obviously the same guy. To a search engine or a language model building an entity graph from web text, these are four weak, unconnected signals — and none of them individually carries enough weight to be treated as a confident, citable authority.
Search Engines Don't Recognize People — They Recognize Confirmed Entities
Google's Knowledge Graph and the retrieval systems behind AI answer engines don't work by reading a name and understanding "this is a person." They work by clustering co-occurring signals — name strings, topic associations, linked profiles, structured data — into a node, and then assigning that node a confidence score based on how consistently and how often those signals reinforce each other. When the same person's work is scattered across the web under three or four name variants with no bridge connecting them, the system doesn't merge the mentions into one high-confidence node. It creates several low-confidence nodes instead, each one too thin on its own to surface in a knowledge panel, a featured snippet, or an AI-generated answer that requires a named source.
This is different from having no online presence at all. The Redondo Beach coach has plenty of presence — arguably more than most competitors in his category. The problem is that his presence is fragmented at the identity layer, which is the layer machines actually use to decide who gets cited. Volume of content doesn't fix this. Fourteen more podcast appearances under "Coach Mike" just add mass to a node that was never going to be selected as the authoritative one, while his certification body — the single highest-trust signal available in his entire footprint — sits attached to a name variant nobody links back to his main site.
Why This Happens to Almost Every Coach and Consultant
Nobody sits down and decides to fragment their identity. It happens because personal brands get built in layers, over years, by different versions of the same person making reasonable decisions in the moment. The nickname gets used for podcast bookings because it's warmer and more memorable for a host introducing a guest. The full legal name gets used on LinkedIn because that's the professional convention. The certification record uses whatever name was on file when the exam was taken, sometimes years before the public-facing brand existed. Nobody is coordinating these across platforms because nobody thinks of their own name as a piece of technical infrastructure that needs a canonical, declared version.
That's the actual failure: name consistency is treated as a stylistic choice instead of a structural requirement. A site's personal-brand website is supposed to function as the hub that reconciles every one of these mentions into a single, verifiable identity — the place where Google and every AI crawler can go to confirm that "Coach Mike," "Michael Reyes," and "Michael A. Reyes, CSCS" are the same node. Most personal-brand sites never do this work. The homepage says "Coach Mike." The About page repeats the nickname with no mention of the legal name at all. There's no structured markup declaring an alternate name, no explicit sentence anywhere stating the relationship between the public brand and the credentialed identity. The one property on the internet with the authority and the incentive to make this connection stays silent about it.
The Architecture That Resolves One Entity Instead of Three
Fixing this isn't a rebrand. Nobody needs to give up "Coach Mike" — plenty of strong personal brands run on a nickname. What has to exist is an explicit, machine-readable bridge between every version of the name, anchored on the one site the person fully controls.
That starts with Person schema on the homepage or About page, using name for the primary public-facing identity, alternateName for every variant in active use, and a sameAs array linking out to every verified profile where the name appears — LinkedIn, the NSCA or relevant certification directory, the publisher page for the book, the show pages for the podcasts that will stay live long-term. This is the single piece of structured data doing the most work in the entire fix, because it's the explicit, unambiguous statement that a search engine's entity-resolution system is built to look for.
Second, the About page copy itself has to state the connection in plain language, not just imply it. A sentence like "Michael Reyes, known professionally as Coach Mike, is a CSCS-certified strength coach based in the South Bay" does more for entity resolution than a hundred more Instagram posts under the nickname alone, because it gives both Google and an LLM parsing the page a direct textual confirmation to pair with the structured data.
Third, every future byline — guest podcast show notes, contributor bios on other sites, press mentions — needs to use the same canonical name-and-title pairing and link back to the same URL every time. Inconsistency going forward re-fragments the entity even after the historical mess is patched. A simple internal rule (one canonical name, one canonical bio sentence, one canonical link, used everywhere without exception) does more long-term entity-building work than any single piece of content.
Fourth, a consolidated media or press page on the site should list every appearance with the same name and same linking pattern, functioning as a single dense cluster of confirmed association — this is the page that gives both Googlebot and an AI crawler the clearest possible signal that all these mentions belong to one entity, and it's the page most personal-brand sites either don't have or built as a scattered logo wall with no text and no links doing the actual identity work.
The Real Cost Isn't a Missing Knowledge Panel
The cost of this problem isn't cosmetic. When an answer engine is synthesizing a response to "who should I talk to about strength training in the South Bay," it's selecting from entities it has enough confidence to name. A fragmented identity doesn't get excluded out of spite — it simply never crosses the confidence threshold required to be cited by name. The coach with fewer credentials and less media exposure but one consistent, well-declared identity wins that citation, every time, because the system can verify him faster and with less ambiguity.
Every credential, every media hit, every guest appearance a coach or consultant has accumulated is only as valuable as the system's ability to confirm it belongs to one person. Split across three name variants with no declared bridge, that accumulated authority is sitting on the internet, unclaimed in the one form that would actually let it compound.
If your name shows up differently depending on which platform someone finds you on, that's not a branding quirk — it's an unresolved technical problem sitting on the one asset that's supposed to consolidate your authority. Worth a conversation with Axesris before the next podcast, guest post, or certification renewal adds one more variant to the pile.