One umbrella brand: Ghost Team and Agent Discovery / agentdiscoverability.com should merge under one brand. The name is still open; the operating direction is a single authority for the product + services, not two parallel brands.
Name criteria: short, brandable, pivot-proof, ideally one word / ~5-6 letters, and broad enough to survive a small or major pivot. A .ai domain is acceptable and may fit better than forcing a .com; strong .com options are likely expensive. A good secondary-market domain may be worth a few hundred dollars.
Domain-search process: brainstorming is only useful if availability is checked quickly. Need an API/plugin-assisted search loop for .ai and .com domains. Cloudflare only helped with .com; Porkbun may support .ai but is cumbersome; GoDaddy/ChatGPT-style domain search was floated as a workable route.
Candidate/name notes: “Tonic” was liked as a feel but likely unavailable/premium. “Addzest” is brandable but may have rights/conflict risk with PFG. Fruit/food/animal-style names show that the brand can be abstract, but the team does not want to blindly copy the Silicon Valley food-name trend. Later candidates (2026-06-23): “Actra” and “Elevate”.
Active shortlist:Berilla, Axolo, Quinn — the name is the main blocker (logo is considered a solvable follow-on). Mavra is OFF the table (2026-07-07, Vincent): the domain isn’t available — it was provisionally locked on the evening of 07-01, then killed on the domain check. Per-candidate read:
Berilla (Vincent’s favourite): story = the first-ever glasses (13th-c Italy) were made from beryl crystal → “seeing” / first-to-see, which maps cleanly onto agent discoverability; the color is almost the existing brand color. Claude language check: means “readings” in Slovenian, close to “laundry” in Croatian, otherwise clean; sounds close to “barrel”/the gem in ES/PT/IT (intended). Risk = Barilla pasta brand, but spelled differently (B-A-R-I-L-L-A vs B-E-R-I-L-L-A). Preferred spelling Berilla over “Berillode”. Logo should not be a literal crystal — rings/circles/abstract forms.
Axolo (Vincent also likes): from axolotl — Aztec origin, regeneration story, water themes to play with. The X gives strong visual distinctiveness / faster cognitive recall; few companies use an “Axo” prefix (seen as a plus).
Quinn: single syllable, four letters, easy recall; “win” embedded → tagline “Win the prompt. Powered by Quinn.” Concerns: reads Polish to some (Andy’s note, sample of one); with the “Quin/Q” spelling the domain isn’t available.
Mavra: cool, memorable word, but means “black” in Greek → risks reading dark/gloomy; would need a light/italic/handwritten logo treatment to offset.
Food/spice names (Harissa etc.) are lower preference — common foods feel low-effort; only obscure spices/exotic fruits stay interesting enough to keep memorability.
Word-linked naming method (2026-06-23): start from a strong, evocative word that fits the brand story (e.g. Elevate), then derive a brandable ~5–6-letter variant within the name criteria — rather than brainstorming abstract strings cold. Mountains are the ideal visual association (e.g. “Alp-” roots), which is awkward because Alpic already owns that space.
Secondary-domain-marketplace tip (2026-06-23): good .coms are largely claimed; many are held by investors reselling them. Worth ~$500 for the right one. To find reputable secondary marketplaces, ask on Reddit (the founders’ working route to marketplace leads).
Primary positioning candidate:“Show up and sell where your customers already are.” It avoids technical jargon, lets visuals carry ChatGPT / Claude / Gemini context, and speaks to the business outcome: distribution and conversion, not just “discovery.”
Supporting language: “Make AI your number one channel” and “build AI distribution in ChatGPT, Claude, and Gemini” are directionally right but more generic. “Get found” is clearer than “discoverability” for marketers, but risks blurring into AEO unless tied to product/tool surfacing and use.
Audience: primary buyers are marketing, growth, product marketing, and product leaders who see AI clients as a new channel; engineering is secondary and likely involved after the buyer has already bought into the channel.
Build is an on-ramp, not the headline: the landing page should not lead with “build a ChatGPT app” because that can shrink the perceived value and lose marketers. Build belongs lower in the journey as the get-present / starter step; the core value is continuous monitoring and optimization of a new distribution channel.
Website conversion narrative: lead with FOMO and business value, then make it easy to start. The strongest funnel idea is an “enter your domain/app” experience that shows AI visibility, competitors already live, and routes prospects into either optimize (already present) or get-present/build (not live yet).
Proof should sell the channel: case studies are the primary conversion driver. Direct Booker is the anchor proof; Lucid’s reported 500% increase and a Statista case study are candidate supporting proofs. Avoid feature-heavy/bullet-heavy pages; use a simple “how it works” and clear case-study storytelling.
Alpic/report branding pressure: a public co-authored Alpic report should ideally launch under the new brand, not a throwaway Ghostteam name. The name should be short, memorable, 1-2 syllables if possible, and carry the “invisible / behind-the-scenes” Ghost idea only if it strengthens the future story.
Open questions
Final company/product name and domain.
Which headline variant wins: “Show up and sell where your customers already are” vs a more explicit ChatGPT/Claude/Gemini line?
What is the minimum viable domain-entry score flow for the homepage, given the technical challenge of matching arbitrary companies outside the known connector/app universe?
Which proof points can be used publicly: Direct Booker, Lucid, Statista, or anonymized versions?
How should “show up”, “get found”, and “sell” map to the product vocabulary without creating AEO confusion?
Can a “Ghost” continuity angle work for the new brand, or does it fight the discoverability/visibility story too much?
Timeline
[2026-07-07] (human correction — Vincent) Mavra is off the table: the domain isn’t available. Shortlist back to Berilla / Axolo / Quinn.
[2026-07-01] (internal — pre-call segment before the 17:30 Laurent call, Vincent + Elliot, recorded by Vincent) Later the same day as the shortlist stand-up: “Mavra” provisionally locked in — “let’s just lock it in so we can move forward and invest in SEO,” keeping the green/design continuity. (Superseded 2026-07-07 — domain unavailable.)
[2026-07-01] (internal — Ghost Team stand-up, Vincent + Elliot, recorded by Vincent) Narrowed the shortlist to Berilla / Axolo / Quinn / Mavra (see Current truth for the per-candidate reasoning). Agreed the name is the blocker; the logo is a solvable follow-on — and the logo should not be tied to a literal object (no crystal, no wolf). Vincent leans Berilla + Axolo; ran a Claude “does it mean anything offensive?” check on Berilla (clean bar Slovenian “readings” / Croatian “laundry”). Elliot to play with Berilla and Mavra in Paper, sleep on it, and decide.
[2026-06-23] (call — “Directbooker next steps”, Vincent + Elliot, recorded by Vincent — internal) Naming exploration in the background of the Direct Booker session: adopted a word-linked method (strong word → brandable 5–6-letter variant; candidates Actra, Elevate), with mountains as the ideal visual (the “Alp-” space awkwardly held by Alpic). Plan to spend an evening locking it down via a Codex word-document loop. Noted secondary domain marketplaces as a route (~$500 budget; ask Reddit for reputable marketplaces).
[2026-06-22] (internal discussion — Alpic report / website messaging) Alpic co-authored report made the rebrand feel urgent: publishing a serious ecosystem report under a name likely to be discarded would waste brand equity. Website messaging reinforced: headline should lead with business value/FOMO before product details; candidate phrasing “be found where your customers already are” plus AI-native distribution channel. Noodle Seed was cited as a useful reference for a clear “I see my app” moment.
[2026-06-19] (internal discussion — duplicate recordings from Elliot + Vincent) Naming and website positioning session. Agreed Ghost Team and Agent Discovery need one umbrella brand; name should be short, brandable, pivot-proof, and likely .ai or a paid secondary-market domain. Strongest headline direction: “show up and sell where your customers already are,” visually anchored by ChatGPT / Claude / Gemini. Product direction sharpened: sell the full channel outcome and continuous optimization cycle; keep build capability lower as an on-ramp; make the homepage score/domain-entry funnel the primary conversion path.