Campaign dates: 29 September to 2 October 2026. All metrics as of 3 October 2026. This GameFi KOL campaign took BattleTown, the legendary arcade tank game reimagined for Web3, from 16.2K to 20.7K followers on its X account (+27.8%) in three days. Thirty-one KOL posts drew 299,623 views at a 2.14% engagement rate, about 20% above the 1.78% those same KOLs average on their own feeds.
BattleTown is a play, win and earn tokens game: the NES tank shooter many players remember, rebuilt on-chain. When the campaign started, its $BTOWN presale was in Round 1 at $0.015 per token. The brief was short on time. Everything was deployed in two days, and the campaign itself ran for three. The headline numbers below are percentages because that is how the client compares periods; the absolute numbers sit under every one of them, with screenshots.
GameFi KOL campaign results in numbers
The engagement rate is the figure we watch most closely. The 31 posts collected 6,400 interactions on 299,623 views, which is 2.14%. The KOLs who published them average 1.78% on their own posts, so the campaign ran about 20% above the baseline of the accounts that carried it. We track this ratio because a paid post that performs below the account’s own average usually means the audience recognised an ad and scrolled on.
The combined audience of the KOLs was 1,855,100 followers. The views came in over three posting days, an average of 9,665 per post. The table shows each channel that ran and what it returned.
A note on the baseline. We compare each campaign against the average engagement rate of the same accounts, not against a market benchmark, because a benchmark mixes accounts of every size and niche. The question a client needs answered is narrower: did these particular KOLs do better or worse for this project than they do for themselves? Here the answer is better, by about a fifth.
| Channel | What ran | Result as of 3 Oct 2026 |
|---|---|---|
| X KOL posts | 31 posts in 3 waves, 5 content angles | 299,623 views, 6,400 interactions, 2.14% ER |
| BattleTown X profile | Posts on the profile: 90 to 142 | 16.2K to 20.7K followers (+27.8%) |
| Web2 display | 4 GIF banners, A/B test on a Web3 audience | 89,945 impressions, 0.21% average CTR |
| YouTube | 18 channels in 5 languages selected | In progress, not counted in this case |
How the messaging was built: five content angles
The campaign was built on content angles rather than one shared script. Each KOL got a BattleTown post that was relevant to their own feed to quote, and an angle matched to the type of content they normally publish. An NFT account and a market-data account do not win over their followers with the same argument, so we did not ask them to use one. These are the five angles, in the words we briefed them.
Quoting a project post, rather than writing a standalone promotional post, keeps the original BattleTown post inside the KOL’s message. A reader who agrees with the KOL sees the project’s own words and is one tap away from its account, which is also where the follower growth shows up.
A. Not pay-to-win
The NFT tank is cosmetic. Tiers change how a tank looks, holders share the tournament pools, and none of it is ever an edge in a match. This angle answers the first objection a gamer raises about a Web3 game before it is asked. All three KOL posts shown further down quote BattleTown posts built on it.
B. Transparency
The team states openly what is built and what is switched off, and anything not yet released is marked as planned. For a presale that is the message in itself: the KOL is not vouching for a roadmap, only pointing at a list the project published about its own product.
C. Tokenomics for players
Unsold presale tokens go to the player rewards pool: 80,000,000 BTOWN, 26.67% of supply, with payouts stepping down 20% every two years across a decade. The angle turns a tokenomics table into a plain statement about who gets paid, which is the version a player audience reads to the end.
D. Nostalgia
The NES tank game, rebuilt on-chain. It is the shortest angle and needs no explanation for anyone who played the original, and it fits accounts that post gaming and pixel-art content better than market commentary.
E. Market data
Seven million wallets open a blockchain game every day. 93% of projects launched since 2020 are gone, and 141 gaming tokens were delisted in H1. The thesis we gave market-focused KOLs: players stopped paying for promises, and a playable beta before listing is a different proposition.
Every KOL also received an individual link with a UTM tag. Each account’s traffic to the BattleTown site can then be separated from the others, and the next wave can be planned on which accounts sent people to the site, not only on which ones collected likes. We describe the briefing format in how to brief a crypto KOL.
Three waves on X: 31 posts and 299,623 views
The posts went out in three waves, on 29 September, 30 September and 2 October 2026. Splitting the roster this way kept BattleTown in the feed across the whole campaign window instead of producing a single spike on one day, and because each group posted on its own day, the groups can be compared with each other afterwards. The interactions broke down as follows.
| Interaction | Count | Share of total |
|---|---|---|
| Likes | 4,284 | 66.9% |
| Replies | 1,620 | 25.3% |
| Reposts | 321 | 5.0% |
| Saves | 168 | 2.6% |
| Quotes | 7 | 0.1% |
| Total | 6,400 | 100% |
Replies make up a quarter of all interactions. A like costs a reader nothing; 1,620 replies means the posts gave people something to answer, which is what quoting a specific project post with a specific angle was meant to do. The campaign summary from our tracking sheet is below.
Three posts from the campaign as examples. Each KOL quoted a BattleTown post and added their own reading of it: @ManLyNFT on Round 1 founder paints that change looks but never stats, @valerioshi on the founder badge and paint pick at $0.015 per token, and @ChemistDeFi on ranked matches where a tier 10 and a tier 1 tank play on the same terms.
BattleTown X account: 16.2K to 20.7K followers
Over the campaign window the BattleTown profile on X grew from 16.2K to 20.7K followers, about 4,500 new followers or +27.8%, roughly 1,500 a day across the three posting days. Posts on the profile went from 90 to 142 (+58%) over the same period. The screenshots are the before and after: the header, bio and verification are identical, only the counters moved.
The profile was opened in July 2026, so the campaign landed on a young account. We give the growth as a percentage in the headline because the absolute gain depends on where an account starts; the absolute figures are in the screenshots for anyone who wants to check them.
Beyond X: a Web2 banner test and 18 YouTube channels
Four GIF banners, A/B tested on a Web3 audience
Alongside the KOL waves we created four GIF banners and A/B tested them on a Web3 audience. They served 89,945 impressions with an average CTR of 0.21% across the four variants. The frame below is from one of them: the one-screen tank battle, the Round 1 price and a Play free button. The banners use the game’s own pixel art and the same $0.015 price the KOLs quoted, so someone who sees both meets one picture and one number.
0.21% is a display number and should be read as one. The test is there to find which creative and which message pull clicks before more budget goes into display, not to compete with KOL engagement, which is measured on a different base.
YouTube: 18 channels in 5 languages, work ongoing
For YouTube we selected the 18 most relevant channels from a base of 11,694 crypto channels, with the US, India and Pakistan excluded by design. The 18 channels cover five languages, English, Portuguese, Turkish, Russian and Vietnamese, with 3.3M combined subscribers and about 183K views per wave. Work there is ongoing, so none of the results above include YouTube. Our list of GameFi marketing channels where the real users are explains why video sits next to X for gaming projects.
Client feedback
During the campaign we also advised the team on the presale purchase flow: add BNB and USDC on BSC as payment options, and put a swap or on-ramp widget on the purchase page for buyers who hold nothing on BSC. The reply is below.
The second screenshot is from the same chat. Above the client’s message is our tracking sheet, with a brief and a UTM link for each KOL; below it is the wave plan for 29 September, 30 September and 2 October.
What this case study does not prove
It does not prove sales. This case reports reach, engagement and follower growth over three days; it does not report presale purchases. The UTM links exist so that BattleTown can attribute purchases to individual KOLs, and those numbers belong to the client. Nor does it say anything yet about YouTube, which is still running and will be judged on its own numbers when the waves there are complete.
It also does not prove that all of the roughly 4,500 new followers came from KOL posts. The profile published 52 posts of its own in the same window, and the banner test ran at the same time. What the numbers do show is that a GameFi KOL campaign of 31 paid posts, built on angles instead of a shared script, ran 20% above the engagement baseline of the accounts that published it, and that it went from brief to live in two days.
KOL selection, angle planning and per-account tracking as a service are on our crypto influencer marketing page and the crypto Twitter KOL marketing service. The screening method is in how to vet a crypto KOL. Earlier X campaigns: the VAIX KOL case study and the Kojiru thought leadership case.
Frequently Asked Questions
How long did the BattleTown GameFi KOL campaign take?
Everything was deployed in two days: KOL selection, the five content angles, a matched quote post for each account and an individual UTM link per KOL. The campaign itself ran for three days, with waves on 29 September, 30 September and 2 October 2026. All results in this case were measured as of 3 October 2026.
How was the 2.14% engagement rate calculated?
Engagement rate is interactions divided by views. The 31 KOL posts collected 6,400 interactions (4,284 likes, 1,620 replies, 321 reposts, 168 saves and 7 quotes) on 299,623 views, which gives 2.14%. The same KOLs average 1.78% on their own posts, so the campaign ran about 20% above their baseline.
Why did each KOL get a different content angle?
Because their audiences are different. An NFT account, a DeFi account and an account that covers market data persuade their followers with different arguments. Each KOL received a relevant BattleTown post to quote and one of five angles matched to the content they normally publish: not pay-to-win, transparency, tokenomics for players, nostalgia or market data.
How were results tracked for each KOL?
Every KOL received an individual link to the BattleTown site with a UTM tag, recorded in a shared tracking sheet next to the brief for that account. This separates the traffic each account sends from the rest of the campaign, so the next wave can be planned on visits per account instead of likes alone.
Did the campaign include YouTube?
Yes, and that part is still running. We selected the 18 most relevant channels from a base of 11,694 crypto channels, with the US, India and Pakistan excluded by design. They cover English, Portuguese, Turkish, Russian and Vietnamese, with 3.3M combined subscribers and about 183K views per wave. YouTube results are not part of the numbers in this case.
What did the Web2 banner test show?
Four GIF banners were created and A/B tested on a Web3 audience. Together they served 89,945 impressions at an average CTR of 0.21% across the four variants. The test exists to pick the creative and message that earn clicks before more budget goes into display, and it is reported separately from the KOL results.