Just got back from korea, there were two KBWs and most people dont even know it One was on the floor, the one everyone has access to - the one every one sees then there is the other one, the hidden one, the one in the invite-only rooms. If you only did the first, the week looked quiet and boring however the reality is far from it The expo was in Korea’s Wall Street, trad fi, banks, securities firms, payment companies etc. suits The ppl that decide which products get built or which tokens get listed were in the invite only rooms, not the expo floor. They usually had a fully booked calendar and didnt really go to the conference floor. If your calendar was only the official programme, you definitely missed the important parts of kbw They had specific things to show. Checkout flows, merchant settlement, cross-border tests. The conversation if youre in crypto you already know has moved past whether blockchain might be useful. Banks are going full throttle into crypto, stablecoins, stripe all show this: Who accepts the payment? How does the merchant get paid? What happens between the two? Korean institutions want a won stablecoin. Banks are already running internal tests, all are aiming for q1 2027 If you’re still waiting for an official announcement you’re already late. Trading-product builders were in the same rooms. Tokenization is not the hard part anymore. “We can tokenize this” was the start of the conversation. The harder questions i heard were who buys it, where it trades, and whether anyone can borrow against it. Korea’s version ran both ways. Global assets into Korea, and Korean stocks made easier for overseas investors. DeFi isnt dead, it kept coming back through lending and liquidity. Issuing an asset is one step. Giving people a reason to hold it and use it is another. The banks have the products. The rooms have the users. Retail remains strong. The institutional focus of kbw did not erase the trading crowd. Live competitions still drew people. Plasma ran cashback at Olive Young, so visitors had a reason to try a crypto card while shopping. Paid posts were not the examples people named. Privacy came up as a trading concern. People didn’t want others seeing their positions and trading ahead of them. Most western teams flew in, met each other, and called it Korea. But the audience that forms opinion sits in private Telegram rooms. Those do not open because you bought a badge. If the relationship was not there before the week, the week did not create it… and this is what most people miss. However the week showed progress, and how much is still being tested as an industry. Gg. If youre reading this and need some advice on korea, pls shoot me a dm, always happy to talk korea w crypto teams

Jumper had 26 Korean mentions in the first 8 months of 2026, then 259 in the 23 days to its Legion sale close. Squid ran the previous Legion sale, and Korea was its #2 country by participants. Squid had 161 in its own 23 days. Commitments: - Squid $26.7M - Jumper $48.6M

Venice just raised at a billion dollars. Profitable, three million users, and it never spied on a single one of them. Rare to see a company get that big without giving up the thing it was built to protect.

Big day for Venice. One of the most legitimate products in crypto for years now. Congrats to the team. Well deserved.

The fastest way to lose $100k in Korea is to activate some KOLs. That phrase is the tell. It means the team has already decided how this works before they understand how Korea actually works. Here's what they think they're buying: > Access to creators > Posts from those creators > Engagement from those posts > Signups from that engagement Four clean steps. Transactional, Measurable, Familiar. None of it is how the market moves. — Korean market entry with KOLs has 𝟱 𝗺𝗼𝘃𝗶𝗻𝗴 𝗽𝗮𝗿𝘁𝘀: > 𝗦𝗲𝗹𝗲𝗰𝘁𝗶𝗼𝗻 — which specific creators fit this specific project > 𝗦𝗲𝗾𝘂𝗲𝗻𝗰𝗶𝗻𝗴 — the order coverage unfolds in over 6-8 weeks > 𝗜𝗻𝗰𝗲𝗻𝘁𝗶𝘃𝗲 𝗱𝗲𝘀𝗶𝗴𝗻 — the structure that gets creators posting again without only being paid to (pay your creators) > 𝗖𝘂𝗹𝘁𝘂𝗿𝗮𝗹 𝘁𝗿𝗮𝗻𝘀𝗹𝗮𝘁𝗶𝗼𝗻 — rebuilding positioning for a Korean audience instead of translating it > 𝗠𝗮𝗰𝗿𝗼 𝗿𝗲𝗮𝗱𝗶𝗻𝗴 — tracking Korean market trends and conditions weekly to time every push All 5 are required. Each one compensates for the blind spots of the others. Miss two and you stall. Miss three and nobody in Korea will care about you. Most teams cover one of the 5. Sometimes two. Here's how they end up there. — 𝗗𝗲𝗰𝗶𝘀𝗶𝗼𝗻 𝟭: 𝗛𝗶𝗿𝗲 𝗶𝗻-𝗵𝗼𝘂𝘀𝗲, 𝗼𝗿 𝗳𝗶𝗻𝗱 𝗮 𝘀𝗽𝗲𝗰𝗶𝗮𝗹𝗶𝘀𝘁. The default move is an in-house hire. A Korean speaking ops person. Someone with KOL relationships. One hire, One salary, Region covered. That hire covers piece 1 of the 5 at best. No single person runs selection, sequencing, incentive design, cultural translation, and macro reading simultaneously. 𝗗𝗲𝗰𝗶𝘀𝗶𝗼𝗻 𝟮: 𝗪𝗵𝗶𝗰𝗵 𝘀𝗽𝗲𝗰𝗶𝗮𝗹𝗶𝘀𝘁. Every service provider in this market works from roughly the same list of Korean KOLs sitting in a Google Drive somewhere: > Some have direct relationships with the creators > Some hired a research team that built the sheet last quarter > Some have worked with specific creators before > Most know 𝘴𝘰𝘮𝘦𝘰𝘯𝘦 𝘸𝘩𝘰 𝘬𝘯𝘰𝘸𝘴 the creators The list isn't proprietary. Access to Korean KOLs is a commodity. What's proprietary is operational depth across all 5 pieces of the stack. That's the entire ballgame. — Three of the 5 pieces almost nobody handles correctly: 𝟭. 𝗦𝗲𝗹𝗲𝗰𝘁𝗶𝗼𝗻 A tier-one DeFi creator covering a gaming project looks forced. A creator covering anything outside their focus reads as paid. Korean audiences pattern-match creator-to-project fit within a post or two. Wrong fit turns your coverage into an advertisement instead of a discovery. The right selection matches editorial voice to project, not follower count to budget. 𝟮. 𝗦𝗲𝗾𝘂𝗲𝗻𝗰𝗶𝗻𝗴 An 8-creator drop in one day is how teams convince themselves they ran a campaign. Audiences read it as a coordinated paid push and discount every post in the sequence. Real sequencing looks boring: > Creator one posts > Time passes > Creator two picks it up from a different angle > Niche channels start forwarding both > Creator three adds their take Each post earns the amplification of the next. Your taget audience sees this unfold and reads it as organic diffusion vs some agency choreography. 𝟯. 𝗜𝗻𝗰𝗲𝗻𝘁𝗶𝘃𝗲 𝗱𝗲𝘀𝗶𝗴𝗻 A creator paid for one post produces one post. A creator given tokens they have to stake, trade, or use on the platform produces content about real experiences. Features they tested, friction they hit, details no brief could have written for them. 𝗣𝗮𝗶𝗱 𝘁𝗼 𝘀𝗮𝘆 𝗶𝘁 𝗮𝗻𝗱 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘂𝘀𝗲𝗱 𝗶𝘁 𝗽𝗿𝗼𝗱𝘂𝗰𝗲 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗰𝗼𝗻𝘁𝗲𝗻𝘁 𝘀𝗵𝗮𝗽𝗲𝘀. 𝗞𝗼𝗿𝗲𝗮𝗻 𝗮𝘂𝗱𝗶𝗲𝗻𝗰𝗲𝘀 𝗰𝗮𝗻 𝘁𝗲𝗹𝗹 𝘄𝗶𝘁𝗵𝗶𝗻 𝘁𝘄𝗼 𝗽𝗼𝘀𝘁𝘀. — Here's the diagnostic that separates operators from access sellers. If you're evaluating a service provider, or questioning whether your in-house hire is doing the work, skip "who do you have access to." Everyone has access in a way. Ask this instead: 𝘞𝘢𝘭𝘬 𝘮𝘦 𝘵𝘩𝘳𝘰𝘶𝘨𝘩 𝘵𝘩𝘦 𝘤𝘳𝘦𝘢𝘵𝘰𝘳 𝘴𝘦𝘭𝘦𝘤𝘵𝘪𝘰𝘯 𝘢𝘯𝘥 8-𝘸𝘦𝘦𝘬 𝘤𝘰𝘷𝘦𝘳𝘢𝘨𝘦 𝘴𝘦𝘲𝘶𝘦𝘯𝘤𝘦 𝘺𝘰𝘶'𝘥 𝘳𝘶𝘯 𝘧𝘰𝘳 𝘢 𝘩𝘺𝘱𝘰𝘵𝘩𝘦𝘵𝘪𝘤𝘢𝘭 𝘱𝘳𝘰𝘫𝘦𝘤𝘵 𝘪𝘯 𝘮𝘺 𝘤𝘢𝘵𝘦𝘨𝘰𝘳𝘺. Vague answer = access seller. Specific answer with category-fit logic, sequencing reasoning, and incentive structure = operator. Activating KOLs isn't a switch. It's a 5-part stack. Budget the rest accordingly.

Most Korea campaigns that look like they're working aren't. Four reasons: #1 — Community means audience, not a Telegram group A team says "we need to build community in Korea" and the people on the ground hear "get a Telegram group to 1,000 members." Six weeks later there's a 1,000-member channel, conversion is zero, and nobody can explain why. Because community was never the goal. Audience was. Audience is the people who know who you are and care enough about what you're doing to actually move when you give them a reason. Telegram is one place your audience lives. Not the only place. #2 — Most teams operate from a partial map There are five real channels for building audience in Korea. Most teams only see one or two. The other three are usually where their actual customers are. 1. Telegram presence in the groups where active crypto-natives spend their day 2. Collaborations with other projects with Korean users (points campaigns, trading comps, co-marketed events) 3. Creator partnerships matched to your specific product category 4. Events and sponsorships inside gatherings where people are already paying attention 5. Coverage on the Korean platforms users search before they participate Not every team needs all five. The right combination depends on the product, the stage, the timeline. But when you're working from a partial map, the wrong combination still feels like strategy. #3 — The teams that fail run the spike before the foundation The teams that work spend the first few weeks getting quiet presence in the right rooms. They find trusted voices who can introduce them properly. The whole point is giving people reasons to care before they're asked to do anything. No spikes yet. Then come the acquisition plays. Events, collaborations, trading mechanics, whatever fits the product. By the time the spike hits, there's already an audience ready for it. Watched a team drop a six-figure trading comp before they had any presence. The campaign hit #1 in their category for almost two weeks. The Telegram still had the same 300 people it started with. Six weeks after the comp ended, everyone forgot who they were. A project that hits #1 tomorrow and drops to zero next month hasn't built anything. A project that climbs steadily over six months has built a real position. #4 — Acquisition without brand doesn't compound Acquisition is not the same as brand. You can pay for acquisition, but brand has to be earned over time. And brand is what compounds when the acquisition stops. Two projects I watched last year ran nearly identical campaigns. Same creator partnerships, similar budgets, both hit decent numbers in the first month. One project nobody talks about anymore. The other compounded into one of the more recognized names in its category. The difference was the brand work they built alongside the campaign. That's the difference between projects that disappear after their TGE and projects that people are still talking about a year later.