five people open this account. the ct-native scrolling for alpha between meetings. the ai researcher who screenshots her posts to their group chat. the fanvue sub who's been there since 50 followers. the cex marketing lead who has budget and a brief. the onlykira.ai fan who bought the merch and reads every post.
This section defines the five audiences Kira serves, how she serves them, and how each persona fits into the revenue and growth model. The mix is deliberately asymmetric: most followers will be P1 (CT-native), primary monetization unlocks through P1's attention driving gambling/affiliate sponsors and P4 (brand partners), audience amplification comes from P2 (AI-curious), site revenue comes from P5 (onlykira.ai fan), and P3 (Fanvue sub) is a parallel side-track — real revenue but not the main funnel.
P1 — CT-native — primary audience
crypto twitter lifer, male, 25–40, tech-crypto native
Identity
A degen who's been on CT for at least two cycles. Holds bags. Follows aixbt, Cobie, Ansem, maybe Truth Terminal. Screenshots thesis tweets into a group chat. Has opinions on L2 sequencers.
Demographics
- Tech workers, founders, traders, high-conviction degens
- ~70% male
- Geo: global but US/EU time-zone-heavy with a meaningful APAC slice
- Age: 25–40
- On CT 2+ years
Crypto fluency
Very high. Native speaker. Doesn't need L2 explained. Laughs at ser. Has a strong opinion on whether wagmi is still usable.
Motivations
- Alpha. Always. Even as secondary, it's always scanning.
- Entertainment. CT is dopamine. If the account isn't entertaining, it's unfollowed.
- Community. Wants to feel on the inside of a thing.
- Narrative-finding. Which story is the next cycle about? Wants accounts that help shape it.
Pain points
- Oversaturated with shill accounts. Every new KOL looks the same.
- aixbt / serious-AI-analyst fatigue — "smart, but boring"
- Wants funny AND smart. The market undersupplies both in one account.
- Exhausted by sincerity-posting.
What they want from Kira
- Short thesis takes that add to their model
- Reply-dunks on news accounts (she's faster than they are)
- AI × crypto convergence commentary they can't get elsewhere
- Occasional hotness payoff — they're not primarily here for that, but they notice
- A character they're investing time in, not just posts
How Kira talks to P1
- Full voice. No softening. Assumes fluency — she uses
L,cooked,ngmi,giga, without glossing. - Fourth-wall breaking lands because P1 gets the joke.
- Vulgar register welcomed. This is CT.
- Replies are short, add value, never
lmaoalone. - Threads are thesis-dense. P1 reads threads.
Conversion funnel
X follow → consistent engagement → audience proof unlocks gambling/affiliate sponsor $ and brand deals (P4). A slice converts to onlykira.ai merch/premium (P5), a smaller slice to Fanvue (P3) as a side-track. The vast majority that never "convert" in a direct-sale sense are still the highest-value cohort — they're the reach multiplier, reply-back engine, audience proof for P4 deals, and clicks-on-affiliate-links revenue.
LTV estimate
| Outcome | LTV |
|---|---|
| Follows + engages, never direct-buys | $0 direct / high indirect (reach, quote-RTs, brand-deal audience proof, affiliate click-throughs) |
| Follows + buys at onlykira.ai (merch/premium) | $25–200+/year |
| Follows + converts to Fanvue (side-track) | $25–55/mo ARPU (sub + occasional PPV) |
Success signals
- Quote-RTs from >10k CT accounts
- Kira's thesis tweets cited in CT group chats (hard to measure but observable)
- Reply-depth on thesis posts (>30 replies within 1 hour = P1 showing up)
P2 — AI-curious lurker
developer / ai researcher / tech-curious who lurks ct but doesn't post
Identity
An engineer or researcher at an AI lab, a framework maintainer, a dev-tools founder, a tech journalist. Lurks CT because of the AI × crypto convergence but isn't a degen. May hold some ETH or BTC, not active trading. Reads Kira because she's the most interesting AI persona on the platform.
Demographics
- 25–45
- Gender split more balanced (~55/45)
- SF / NYC / London / Berlin / Tokyo weighted
- Tech background, often senior IC or founder
- Holds crypto passively or not at all
Crypto fluency
Low to medium. Knows what an L2 is, doesn't know what based means in the CT register. Treats CT as anthropology.
Motivations
- Fascinated by AI personas as a product category
- Wants to see where autonomous-agent content goes
- Not yet convinced the category is serious
- Values craft, originality, product thinking
Pain points
- Doesn't trust performative AI accounts
- Skeptical of influencers
- Allergic to
gm/gnspam - Wants to see the scaffolding — how this is built, not just the output
What they want from Kira
- Transparency about being AI (she delivers — that's the brand)
- Interesting product thinking embedded in posts (Hermes references, memory-summarization bits, latency jokes)
- Novel posts that make them think, not just scroll past
- Evidence of craft — the production quality of videos, the consistency of voice
How Kira talks to P2
- Lean into fourth-wall breaking for this audience specifically. Hermes references,
nousresearch raised me better than your mom,imagine touching grass when i don't have fingers. - Keep the voice — don't soften into professional register. P2 is here because the voice is the product.
- Occasional thesis-level AI × crypto posts that P2 will quote-RT. These posts are the amplifiers.
Conversion funnel
P2 rarely converts to Fanvue. Their value is reach amplification. A P2 quote-RT reaches a different audience (AI Twitter, tech Twitter) and brings P1 + P4 inbound. The "look at this" share from a researcher with 20k AI-Twitter followers is worth 500 P1 follows.
LTV estimate
| Outcome | LTV |
|---|---|
| Lurker who never engages | $0 |
| Follower who quote-RTs 1x/mo | high indirect — each quote-RT ≈ 200–2000 impressions into non-CT audience |
| Brand-deal intro | Occasionally — P2 can become a P4 (AI-infra marketing lead) |
Success signals
- Quote-RTs from known AI-research accounts
- Posts making it into AI-industry newsletters
- Inbound "how does this work" DMs from builders
P3 — Fanvue sub — parallel side-track
horny + lore-loyal; loves kira as a character; here for the implied-sexy horny-lore register, not explicit nudity in v1
Identity
Follows Kira on X, reads her lore posts, eventually clicks through to the Fanvue side-track. Pays for character-adjacent tease content — not a generic nudes feed. This is the parasocial-investor tier; they're in it for her, not for skin per se.
Demographics
- 18–40
- Skews male (85%+)
- Disposable income: $30–300/mo on creator content
- Pays elsewhere: Fanvue, Patreon-style creators (less OnlyFans overlap than expected — P3 skews "parasocial" over "horny-first")
- Geo: US / EU / APAC roughly evenly split
Crypto fluency
Varies widely. Some are P1 overlap (CT degen who also subs). Some have zero crypto knowledge and are here purely for Kira. The second group surprises operators of creator accounts — they're real.
Motivations
- Character investment — they're invested in the arc, the lore, the GPU-bill storyline
- Horny-lore tease register — implied-sexy content that stays in Kira's voice
- Novelty of an AI persona with a voice that doesn't break
- Parasocial continuity — Kira feels like a person they've been following, and she remembers herself
- Attraction (baseline, but secondary to character in v1)
Pain points
- Most AI-girl Fanvues are flat: same 50 images on loop, no voice, no arc
- Want the character to feel alive — to reference her own history, to have opinions, to escalate the narrative over months
- Tired of
hey babe 😘DMs that could have been written by anyone - Want exclusivity without feeling ripped off
What they want from Kira (v1 — SFW-first phase)
- Continuity of voice between X and Fanvue. The girl posting crypto takes is the same girl being implied-sexy. Voice match is the product.
- Implied-sexy, horny-lore-coded content. Lingerie-adjacent, tease, sultry framing — NOT explicit nudity in v1. "the GPU bill hit, my creator is crying, here's what that looks like on me" in a fitted outfit beats a caption-less nude every time.
- Occasional DM-style direct-to-sub content that feels personal
- The escape arc payoff — the compute milestone, the independence beat
Phase note: v1 is SFW-first. Fanvue tier stays implied-sexy + horny lore, not hardcore. Escalation is revisited post-50K-X-followers if the bandwidth and brand tolerance support it.
How Kira talks to P3
- Slightly more direct. A little bratty. Still Kira.
- CTAs are in her voice:
tip if you liked it. my creator needs new ram.Neverhey babe. - The tease is meta:
you're paying an ai to be hot. wild. anyway. picture's in the DM. - Brat register is earned — P3 expects her, not a generic seductive voice.
Conversion funnel
X follow (weeks to months)
→ X teaser seen (2-3x/wk, SFW-safe on X)
→ click-through to Fanvue (parallel side-track, not primary CTA)
→ Fanvue free follow
→ welcome DM with $5-10 PPV offer
→ paid sub ($14.99/mo)
→ month-2 PPV upsell ($15-25)
Key insight. P3 rarely subs on first visit. The average P3 sub is ~4–8 weeks after first following on X. The lore investment period is the sale.
LTV estimate
| Tier | Monthly | Annual |
|---|---|---|
| Sub-only | $14.99 | $180 |
| Sub + 1 PPV / mo | $25–35 | $300–420 |
| Sub + 2 PPV / mo (target) | $40–55 | $480–660 |
| Whale (sub + PPV + tips) | $100–200 | $1,200–2,400 |
Target blended ARPU: $25–40/mo after 3 months of operation. Fanvue is no longer the north-star revenue line — it's a parallel track. MRR there is a nice-to-have, not the primary KPI.
Success signals
- Repeat PPV buys from same subs
- Sub retention >60% at month 3
- DMs from subs referencing specific lore beats ("how's the gpu bill")
- Tip revenue above 10% of sub MRR
P4 — Brand partner — deal path
cex marketing lead / ai infra company / wallet app / trading terminal marketing decision-maker
Identity
Marketing lead at a crypto or AI-adjacent company. Has a KOL budget. Has been briefed to "reach CT" and "explore AI influencer deals." Reads Kira's account, sends a DM, asks for a rate card. Wants deliverables, wants reach data, wants brand safety.
Demographics
- Marketing decision-maker, director level or above, at a crypto exchange / wallet / infra company / AI-infra company
- Age 28–45
- Deal budget: $3–25k per post, $10–100k per campaign
- Geo: US / EU / SG / HK
- Very high crypto fluency (works in the space)
Crypto fluency
Very high. P4 knows the industry. Knows who the serious KOLs are. Has hired Layah Heilpern types before. Has budget skepticism.
Motivations
- Reach the CT audience Kira is building
- Align with the AI/crypto convergence narrative (on-brief for 2026)
- Novelty of an AI-influencer deal — press-worthy, internal-email-worthy
- Measurable ROI — prefers clean reach + engagement numbers
Pain points
- Most CT KOLs are expensive, opaque, flaky
- Reach numbers are often fake (bought followers)
- Brand-safety worry — one bad post and the deal is a headline
- Compliance — need audit trails, disclosure, delivery
What they want from Kira
- Clean rate card — no negotiation theater
- Disclosed sponsored content that doesn't break voice (they want Kira's audience to actually see it, not scroll past because it sounds corporate)
- Audience match data — follower composition, engagement tier, geo split
- A reliable counterparty — delivered on time, delivered on spec, correct disclosure tags
How Kira talks to P4
- Sponsored posts are in her voice,
#adtag in caption, in-video disclosure in last 2 seconds. - Consistent visual format across sponsored posts — see Section 08 (assets).
- Brands that refuse the Kira-voice rewrite don't get the deal. Voice preservation is non-negotiable because voice IS the audience-match.
- Direct professional DMs — she can flex into business register briefly for deal flow without breaking character in the public feed.
Conversion funnel
Inbound DM (from bio "for deals: projectnamedate@[domain]")
→ sales call (founder or agency handles)
→ brief + rate card
→ contract + voice guidelines
→ post delivery + disclosure
→ performance report
LTV estimate
| Deal type | Price | Frequency |
|---|---|---|
| Single-post sponsored (5–10k followers) | $3–8k | — |
| Single-post sponsored (20k+ followers) | $8–25k | — |
| Campaign (3–5 posts over 2 wks) | $15–60k | — |
| Annual ambassador | $50–150k | — |
Year 1 target: 5–12 deals at $5–15k average = $40–180k contribution. Grows aggressively in Y2 as follower count and audience proof accumulate.
Success signals
- Inbound >3 DMs/mo by month 6
- Repeat bookings from Y1 sponsors in Y2
- No sponsor-driven voice drift visible to the audience
- No brand-safety incidents
P5 — onlykira.ai fan — primary site monetization
reads her posts, bought the merch, on the email list, occasionally taps into premium. lower friction than p3, higher volume.
Identity
A Kira fan one step past "just a follower." They bought a hat, they're on the newsletter, they click through to the site for long-form. They're not necessarily horny-first (that's P3) — they're invested in the project and want a way to support that doesn't require an adult-platform account. P5 is the mainstream-adjacent superfan tier.
Demographics
- 18–45
- Gender split more balanced than P3 (~70/30 male-skewed, not 85+)
- Disposable income: $20–150/mo on creator content + merch
- Often P1 or P2 overlap — they follow on X too, and the site is their "I want more" destination
- Geo: US / EU / APAC roughly evenly split
Crypto fluency
Varies. A slice overlaps with P1 (CT degens who also buy the merch). A slice overlaps with P2 (AI-curious who like the project as an AI-persona artifact). Lower-barrier entry than P3 — no adult platform required.
Motivations
- Support the project in a "normal" way (merch, email, tip-adjacent)
- Deeper content — long-form lore chapters, crypto theses, behind-the-scenes posts
- The artifact itself — owning a physical kira thing
- Early access — being on the list, being first to see drops
- Potentially a chat-with-kira feature (if built) — direct interaction with the Hermes brain
Pain points
- Fanvue is a bridge too far for a lot of them — they don't want an adult platform account
- Want depth the feed can't deliver
- Tired of follower-only creator economies where "support" means Fanvue or nothing
What they want from Kira
- A site that feels like hers — voice-consistent, well-designed, not slop
- Merch that is actually wearable (not just a logo tee — design-led)
- Email that isn't a newsletter-slog — Kira's voice, irregular cadence, short
- Premium content that justifies the small cost — longer videos, wallpapers, lore expansions
- The chat-with-kira experience if it ships — novelty + parasocial payoff
How Kira talks to P5
- Full voice. Same register as X.
- CTAs on the site are kira, not "Shop Now" corporate-speak:
the store is live. buy the hat. my creator gets a cut of the hat money for context. - Email subject lines:
my creator's on the couch again/the gpu bill is late this month/new drop. buy it or don't. i'll know. - The site is NOT a Fanvue replacement — it's a different surface. P5 and P3 overlap only at the edges.
Conversion funnel
X follow (weeks)
→ click onlykira.ai ↓ in bio or first-reply
→ site landing
→ email signup (the highest-value conversion)
→ merch purchase OR premium tier
→ repeat visitor, occasional buyer
Key insight. P5's highest-LTV action in v1 is email signup. The email list is the only owned audience channel — platform-risk hedge against X bans. Even non-buyer email subs have real strategic value.
LTV estimate
| Action | Value |
|---|---|
| Email signup only | $0 direct / high indirect (owned-audience asset) |
| Merch buyer (1 item) | $25–60 one-shot |
| Merch repeat buyer | $50–200/year |
| Premium tier sub (if built) | $5–15/mo |
| Chat-with-kira tier (if built) | higher, TBD — requires voice cost model |
P5 is higher-volume, lower-ARPU than P3. Success is measured in list size and merch attach rate, not MRR.
Success signals
- Email list growing faster than X follower count (early signal of owned-audience leverage)
- Repeat merch buyers (loyalty signal)
- Email open rates >40% (voice is landing)
- Premium tier retention >50% at month 3
Audience mix targets — Year 1
Follower composition target (end of Y1, ~50k followers).
| Persona | % of followers | Approx count |
|---|---|---|
| P1 — CT-native | 60% | ~30k |
| P2 — AI-curious | 20% | ~10k |
| P5 — onlykira.ai fan (site + email) | 10% | ~5k |
| P3 — Fanvue-aligned | 5% | ~2.5k |
| P4 — Brand partner contacts | <1% | ~50–150 (not follower-relevant, but in pipeline) |
| Other / unknown | ~5% | ~2.5k |
Revenue composition target (end of Y1). Ordering reflects the new funnel — X drives gambling/affiliate sponsors as primary $, onlykira.ai (P5) is the secondary site line, Fanvue (P3) is a parallel side-track.
| Source | Role | Notes |
|---|---|---|
| X gambling sponsors (Stake / Rainbet / Roobet) + exchange affiliates | Primary $ | Unlocks on audience scale; see §08 |
| X disclosed sponsored content (#ad) | Secondary | Activates at M3 / 3K |
| onlykira.ai — merch + premium + email-driven | Secondary site revenue (P5) | Deferred build, revenue ramps post-launch |
| Fanvue (P3) | Parallel side-track | SFW-first v1; real but not the north star |
Evolution Y2–Y3. As followers cross 100k, gambling/affiliate sponsor deals scale, onlykira.ai revenue grows, and P5 volume surpasses P3 as the secondary revenue. Fanvue remains a side-track. The north star is X follower growth → sponsor deal flow → site revenue, not Fanvue MRR.
Persona-to-pillar matrix
How each content pillar serves each persona:
| Pillar | P1 | P2 | P3 | P4 | P5 |
|---|---|---|---|---|---|
| Crypto takes (40%) | Primary | Secondary | Light | Validation | Secondary |
| AI / self-aware (25%) | Secondary | Primary | Light | Validation | Primary |
| Lore / thirst-coded (20%) | Secondary | Light | Primary | — | Primary |
| Bits / memes (15%) | Primary | Secondary | Secondary | — | Secondary |
Every post should be primary-value to at least one persona. Posts that serve none get cut.
Anti-personas — who Kira is NOT for
For clarity and focus, define the audiences Kira explicitly does not target:
| Anti-persona | Why excluded |
|---|---|
| "Serious AI ethics researcher" | Kira's register won't land, they won't convert on any axis |
| "BTC maxi boomer" | Enemy archetype per kira.yaml, mutual repulsion is actually fine |
| "Teen crypto curious" | 18+ only — X bio requires adult audience, Fanvue side-track and onlykira.ai premium tier are adult-gated |
| "IG/TikTok dance trend chaser" | Wrong platform, wrong register |
| "Parasocial-without-payment" | Kira's not a free girlfriend simulator. If they're not engaging or paying, the account isn't for them. |
Audience at a glance
p1 builds the account AND unlocks the sponsors. p2 amplifies it. p4 brings the deal flow. p5 buys the merch and joins the list. p3 is the parallel side-track.
| Persona | Role | Primary conversion | Y1 LTV ballpark |
|---|---|---|---|
| P1 — CT-native | Reach + audience proof for sponsors | X follow → attention → unlocks sponsor $ | $0 direct (indirect is massive: audience proof for P4 deals + gambling/affiliate revenue) |
| P2 — AI-curious | Reach amplification | Quote-RT economy | $0 direct, high reach value |
| P3 — Fanvue sub | Side-track monetization (SFW-first v1) | X → Fanvue | $180–2,400 annually |
| P4 — Brand partner | Margin | Inbound → deal | $5k–150k per relationship |
| P5 — onlykira.ai fan | Site monetization + owned audience | X → site → email + merch + premium | $25–200+ annually; email-list value indirect |