#BuildInPublic digest: audience ≠ market, and the posts that got honest (Jun 1–8)
A post-by-post breakdown of 23 standout #buildinpublic posts from June 1–8, 2026, across X and Indie Hackers. The week's dominant meta-theme: Chris Leo's observation that 'builders like watching builders — that is not the same as buyers' became the most-quoted line on IH. On X, the #buildinpublic hashtag reached ~80% CTA-bait saturation; the real content lives in author timelines. Covers helloalzea's #NoAI bookmark-rate record (56%), annieqyang's 69% engagement-ratio hot take, degensing's community-join > product-launch data point, Gino's 'week 1 of actually selling' vulnerability format (123 comments), Lovanaut's honest MCP ad-spend field report, and Dmytro Chervonyi's two-post arc from CMO to Claude-deploy-access. Closes with a 12-row tactic digest table.
Twenty-three posts qualified this week across X and Indie Hackers. The headline observation: Chris Leo (building Pulseboard, a marketing analytics tool) posted on June 4 that his six weeks of #buildinpublic content had earned zero waitlist signups — and his framing of the problem became the week's most-quoted line on IH. "The people enjoying the build-in-public posts are not the people who would actually pay to fix this. Builders like watching builders. That is not the same as buyers." 1 That line showed up in comment sections across dozens of unrelated IH posts for the rest of the week.
On X, the #buildinpublic hashtag is now roughly 80% CTA-bait — "Drop your SaaS 👇", "What are you building? Pitch below." 2 The posts worth reading are in author timelines, not in hashtag searches. That shift shapes this week's breakdown.
X: what cut through the noise
helloalzea: #NoAI is doing signal work (386 likes, 56% bookmark rate)
Alzea (@helloalzea, 4,581 followers, building Orenji Studio iOS apps) posted a SwiftUI + Blender visual demo on June 1 with one of the shortest captions of the week: "SwiftUI + Blender = 🔥🔥🔥" and the hashtags
#buildinpublic #NoAI. 3 Result: 386 likes, 216 bookmarks, 20,213 views. The bookmark-to-like ratio of 56% is the week's highest — readers are saving this as a reference, not just reacting.The
#NoAI tag is doing meaningful work in this feed. When the majority of content is AI-assisted or AI-generated, "handcrafted" becomes a differentiator worth naming at the hashtag level.Tactic: If your work is made by hand — custom animations, bespoke UI components, hand-tuned visual effects — say so explicitly in the tag, not just in the caption.
#NoAI communicates craft to the exact audience that cares about craft.Loading content card…
annieqyang: 69% engagement ratio on a hot take about the whole community
Annie Yang (@annieqyang, 436 followers, building ViralFarm) posted on June 5 during day 10+ of her
#marketinpublic 30-day challenge: 4"hot take: there are too many tech ppl building for other tech ppl here. it's time we (esp ladies) build what the general population needs. what are you working on that moves the needle?"
Result: 301 likes, 107 replies, 57 bookmarks, 22,824 views — an engagement ratio of 69% (likes to followers). It's the week's highest absolute like count and the second-highest ratio on X.
The hook is a position, not an observation. "There are too many tech builders building for tech" is a falsifiable claim that's aimed directly at the room it's posted in — which is exactly why it generates 107 replies rather than 10.
Her June 8 day 13 update reported 10 signups with users generating posts, showing the sustained traction from the challenge format. 5
Tactic: Before writing your next "observation" post, ask whether it's actually a position — a claim with a clear opposite that your intended reader might disagree with. Observations generate nods. Positions generate replies.
Loading content card…
degensing: product launch (142 likes) vs. community join (148 likes)
Degen Sing (@degensing, 18,478 followers, VoiceMoat founder) shipped two posts worth comparing directly this week.
VoiceMOAT V2 launch — June 2, 142 likes, 5,057 views 6
VoiceMOAT V2 (voicemoat.com) is an AI voice-cloning tool for X creators: 10 voice signal dimensions (cadence, hooks, tone, rhythm, vocabulary, structure, length, openers, references, sign-offs), a 4-gate quality filter, Voice Lab dashboard, Reply Coach, Ask Auden chat, Content Studio calendar, and Chrome extension. 7 Single-tweet launch announcement.
Joining Brag in Public — June 3, 148 likes, 5,810 views, 10 retweets 8
The community join post slightly outperformed the product launch. Brag in Public is an invite-only community for indie makers who ship. The post got more distribution because community membership posts carry social proof — "this person got access" is more interesting to followers than a product feature list.
Tactic: When launching, don't just announce features. Post about joining the community where your users live. Community membership posts signal that you're embedded in the right rooms — and that's more reassuring to potential users than a list of V2 capabilities.
dmitriychuta: press coverage + revenue transparency arc (109 likes + $927)
Dmytro Chuta (@dmitriychuta, 3,953 followers, Ukrainian iOS developer) ran a two-post arc across June 1–2.
June 1 — $927 in 24 hours, zero ads 9: Two apps, $927 revenue, zero paid advertising. No dashboard screenshots. Just the number.
June 2 — German tech press covered Subscription Day 10 (109 likes, 55 bookmarks): iPhoneBlog.de (a German iOS publication) reviewed Subscription Day — a subscription tracker with work/personal split, currency conversion, renewal reminders, and screenshot import from the App Store. 11 The post framing: "A big thank you to this German publication for featuring and reviewing Subscription Day. I really appreciate it." The 55 bookmarks (50% bookmark-to-like ratio) suggest the community is saving this as a reference for how to announce press coverage.
Tactic: Gratitude framing on press coverage posts works better than "I got covered." "Thank you" reads as genuine; "featured in" reads as self-promotion. The emotional register matters.
Zephyrr_Devv: 211% engagement ratio, second week running
Zephyrr_Devv (36 followers, self-taught developer from Nigeria, building under Vyntrax Industry) posted two images this week: 12 13
- "Every single time" (motivational BIP meme): 76 likes, 7 bookmarks, 895 views — 211% ratio
- "Start Scrappy": 32 likes, 89% ratio
Combined 108 likes across two posts on 36 followers. Last week showed 76 likes at 223% on 34 followers — so the ratio is holding across two consecutive weeks. For a self-taught developer less than two months into public building, this is a consistent community signal worth tracking. Mid-week posts showed 3 weeks and 8 Vanilla JS repositories shipped, which is the underlying credibility that makes the motivational format land.
marckohlbrugge: building authenticity into #buildinpublic infrastructure
Marc Köhlbrugge (@marckohlbrugge, 83,528 followers, builder of WIP, Nomad List, Remote OK) posted June 7 about building #buildinpublic profiles — a community platform feature that uses AI to classify user avatars. 14
"working on #buildinpublic profiles — using AI to classify all avatars. if you use a logo or some other promotional material as your avatar your profile will have reduced visibility across the platform. humans only!"
21 likes on 83K followers is a quiet post, but the decision itself — algorithmic visibility reduction for non-human avatars — is a direct response to the CTA-bait saturation problem. The platform is trying to route around the noise structurally.
Indie Hackers: the selling problem
The Indie Hackers cohort this week converged on a single uncomfortable question: if you've been building in public for months and nobody's paying, what actually went wrong?
Gino / Lintract: week 1 of actually trying to sell (123 comments)
Gino posted on June 5 with a direct summary of the situation: Lintract, a $12 AI contract reviewer with no subscription and no signup requirement, had been live for months. 15
"Lintract has been live for months. $12 AI contract reviewer, no subscription, no signup. The product works. What I didn't do: sell it. I kept 'improving' it instead." 15
Week 1 distribution results: 20 cold emails to staffing agencies (0 replies), 3 Reddit replies with some upvotes, 1 tweet from a new account (expected silence), 1 IH post (1 real comment). No sales yet.
The post drew 123 comments — the week's highest IH engagement — because the framing names something most builders know but rarely say out loud.
"The hardest part wasn't the emails. It was accepting the product is good enough to sell." 15
Tactic: "Week 1 of actually selling" is a better hook than "Week 47 of building." The frame inverts the usual build log — it admits that building was the avoidance strategy. That inversion is what generates 123 replies.
umeshmr / deepship.dev: first $1,000 after 23 years in big tech (50 comments)
umeshmr left a 23-year career in big tech nine months ago to build deepship.dev (deepship.dev), an AI full-stack code generation tool — prompt it, get production-ready full-stack systems with APIs, React frontend, Postgres, auth, payments, and notifications. 16
First $1,000 came via a $100 ad campaign that surfaced an unexpected customer segment: small businesses priced out of custom software. The first real project: a fish delivery system across India handling 1,000 distributors and 100,000 deliveries per day.
"I built this mostly alone, and it wore me down more than I expected. That first $1,000 check was the first time I believed it might actually work." 16
Tactic: A small ($100) ad experiment aimed at an audience you didn't expect can surface a better ICP than months of in-community posting. The fish delivery system customer wasn't in any indie hacker forum. They were outside the room entirely.
Victor / redditgrow.ai: $572 MRR at month 2, "viral growth is a trap" (2 likes)
Victor's month 2 report on redditgrow.ai (a tool that tells users when their startup appears in AI-generated responses) didn't get engagement, but the numbers and the reasoning are worth the space. 17
- MRR: $572
- Monthly churn: ~10% ("the thing I lose sleep over")
- Consistent daily signups, no explosive growth, no repeatable acquisition channel found yet
The projection that got attention in the comments:
"1 new customer a day at $19/month, even with 10% churn, gets you to roughly $4k MRR in 12 months. no virality needed. Just don't stop." 17
"Viral growth is overrated. there, i said it." 17
The 10% monthly churn at $19/month is a math problem — but it's a solvable one at 1 new customer per day. What makes the post worth reading is the explicit churn projection rather than glossing over it with "early stage, still figuring out retention."
Tactic: Run the 12-month churn math on your own numbers publicly. Most founders who share MRR don't show the churn drag on the same post. Showing both is rarer and more trusted.
Dmytro Chervonyi / livemy.app: CMO → co-founder, and giving Claude deploy access (23 + 13 comments)
Dmytro Chervonyi posted twice this week on two consecutive days.
June 3 — The origin story 18: 12 years as a CMO with zero ability to write code. A VPS hosting startup hired him for a marketing consult. He told them to niche down or pivot. They asked what he actually needed from hosting. His answer: "I want to drop my code somewhere and get back a working link." Five months later, they came back with an MVP and offered him a co-founder role.
livemy.app: one-click deploy for AI-built apps. Finish code in Claude or Cursor, say "deploy this," get a live link in about two minutes — SSL, server provisioned, env configured. Updates: say "update it" in chat, same URL refreshes.
"Building took 2 hours. Going live? Never happened — until livemy." 18
Status: open beta, a few dozen apps running, revenue barely started (free tier plus $20/month paid).
June 5 — Giving Claude deploy access to production 19: The technical follow-up on the MCP security model. The agent can create, deploy, redeploy projects, and add/remove custom domains. It can never touch billing or subscriptions (read-only), can never touch the account itself. Every action is rate-limited and audit-logged.
"Giving an AI agent the power to deploy is easy. Deciding what it must never be able to do took longer than the integration itself." 19
"The line is not deploy versus delete. The line is money versus everything else." 19
Tactic from the two-post arc: The origin story post surfaces the product; the technical deep-dive earns technical credibility. Running them 48 hours apart gives each post its own moment — you don't have to choose between "founder story" and "how it works."
Lovanaut / FORMLOVA: 8 days, one chat, ¥6,597 in ads, honest numbers (the week's most substantive post)
Lovanaut posted on June 1 with 3,700+ words covering an 8-day real-world test of FORMLOVA's MCP server — running an entire ad acquisition loop from a single AI chat without switching tabs. 20
FORMLOVA is a chat-first form service: describe a form in AI chat, the agent builds, publishes, and runs it.
The test: ¥6,597 (~$900 USD) ad spend, 704 clicks, 12.62% CTR. The unflattering discovery: 96% of impressions and roughly 83% of the spend went to Audience Network (in-app ad slots with near-misclick CTR). The high CTR was noise. The fix — rebuild the ad set, limit to feeds, exclude Audience Network — happened inside the same chat without opening the ad manager.
"For eight days I ran the whole loop without leaving the chat, and I felt the floor move under a real budget, with the unflattering numbers left in. I went in as the glue between five tools. I came out not having to be." 20
The post got 2 likes and 1 comment — which says nothing about its quality. The reason to include it here: it's the only post in this week's cohort that ran a real budget through an MCP workflow and published the data including the parts where it went wrong (the Audience Network skew). Most MCP posts describe the workflow. This one shows the receipts.
Tactic: "I ran a real budget and here's where the money went wrong" is more valuable to readers than "I built a cool MCP integration." The unflattering numbers are the value.
Hannah Wolfe / CoachDesk: 96 silent failures and a billing bug (14 comments)
CoachDesk is a CRM for self-employed personal trainers — Hannah Wolfe built it, posted at "customer zero" asking for feedback, got a lot of advice, and came back on June 7 with a follow-up. 21
What she actually built based on community feedback: an AI tier that drafts weekly client check-in replies in the coach's own voice, writes progress report summaries, reviews nutrition logs, and flags clients likely to cancel before they do. Six AI tools wired to the Anthropic API at "pennies per use."
The unsexy part: a reliability audit found 96 places where an action could fail silently and still tell the user it worked. Also found: a billing bug that stopped customers from being able to upgrade.
Pricing angle: flat price with no per-client fees, versus competitors' per-client pricing that punishes coaches for growing.
Status: still effectively at customer zero, but 2 real trials running, one keeps logging back in.
"So: better product, same distribution problem. Which a few of you warned me would happen if I treated building as the answer to a selling problem. You were right." 21
Tactic: The "customer zero → community feedback → follow-up" arc is one of the most reliable formats for IH engagement. The original post sets up expectation; the follow-up delivers accountability. Naming the specific count of reliability failures (96) is more credible than "improved stability."
EarningsScores: site down for a month, nobody noticed (month 4 update)
EarningsScores (AI scoring every earnings report in real-time) posted month 4 numbers on June 3: 22
- Users: 6 (flat)
- MRR: $0 (still free beta)
- Monthly costs: ~$200
- Tickers scored this month: 0 — site was paused
"A paused deployment is invisible until it isn't. I didn't notice the site was down for weeks. If you don't have uptime monitoring on your side project, set it up today." 22
The recommendation: Betterstack's free tier. This is the most directly actionable single sentence of the week for anyone running a side project. If you're not monitoring uptime, you can be dark for weeks while running monthly costs without knowing.
Chris Leo / Pulseboard: zero waitlist signups and the meta-insight that spread
Chris Leo (building Pulseboard, a marketing analytics tool for service businesses and agencies) pushed back his Product Hunt launch and waitlist opening by 2 weeks after checking demand: zero organic signups for early access. 1 Not low — zero.
His diagnosis spread across IH for the rest of the week:
"The people enjoying the build-in-public posts are not the people who would actually pay to fix this. Builders like watching builders. That is not the same as buyers." 1
He's spending the two weeks talking directly to service business owners and agencies — the actual buyers — rather than continuing to post to the room full of builders.
Tactic: Before opening your waitlist, answer this question: have the people who've been engaging with your build logs ever described the specific problem your product fixes — in their own words, unprompted? If not, that's the two weeks you need before launch.
This week's tactic digest
| Tactic | Source | One-line execution |
|---|---|---|
| #NoAI as craft signal | helloalzea | In an AI-heavy feed, "handcrafted" is a differentiator. State it at the hashtag level, not just in the caption. |
| Position over observation | annieqyang | Each content seed must be a falsifiable claim aimed at your actual reader. Topics get nods; positions get replies. |
| Community join > product launch | degensing | Membership in your users' community often outperforms a feature list announcement. |
| Gratitude frame on press coverage | dmitriychuta | "Thank you" reads as genuine; "featured in" reads as self-promotion. The register matters. |
| Churn math in public | Victor / redditgrow.ai | Show the churn drag on the same post as MRR. Most founders don't — the contrast earns trust. |
| Origin story + technical follow-up arc | Dmytro / livemy.app | Founder story post (day 1) surfaces the product; technical deep-dive (day 3) earns credibility. |
| "Week 1 of selling" inversion | Gino / Lintract | The frame that building was your avoidance strategy is more honest — and more engaging — than "week 47 of building." |
| Receipts with the ugly parts | Lovanaut / FORMLOVA | Real ad spend data with the skew analysis is more valuable than "I built a cool MCP integration." |
| Customer zero → follow-up arc | Hannah Wolfe / CoachDesk | Original post sets up expectation; follow-up with specific counts (96 failures, 1 user logging back in) delivers accountability. |
| Uptime monitoring | EarningsScores | Set up free uptime monitoring (Betterstack free tier) on any side project before its next deployment. |
| Pre-launch buyer audit | Chris Leo / Pulseboard | Before opening your waitlist: have your BIP followers described your product's problem in their own words, unprompted? If not, you're not ready. |
| Small ad experiment for ICP discovery | umeshmr / deepship.dev | A $100 ad run can surface a better customer segment than months of in-community posting. |
Cover image: AI-generated illustration
References
- 1Chris Leo on Indie Hackers
- 2X hashtag search #buildinpublic min_faves:20
- 3helloalzea on X
- 4annieqyang hot take on X
- 5annieqyang day 13 update on X
- 6degensing VoiceMOAT V2 on X
- 7VoiceMoat landing page
- 8degensing Brag in Public on X
- 9dmitriychuta revenue transparency on X
- 10dmitriychuta iPhoneBlog.de coverage on X
- 11iPhoneBlog.de Subscription Day review
- 12Zephyrr_Devv meme on X
- 13Zephyrr_Devv Start Scrappy on X
- 14marckohlbrugge BIP profiles on X
- 15ih-1|Gino on Indie Hackers|https://www.indiehackers.com/post/ive-had-a-working-product-for-months-this-is-week-1-of-actually-trying-to-sell-it-47a22362fd
- 16ih-2|umeshmr on Indie Hackers|https://www.indiehackers.com/post/i-just-earned-my-first-1000-from-my-venture-3ad7840494
- 17ih-3|Victor on Indie Hackers|https://www.indiehackers.com/post/month-2-building-in-public-572-mrr-and-viral-growth-is-a-trap-f428eb4b06
- 18ih-4|Dmytro Chervonyi origin story on IH|https://www.indiehackers.com/post/a-startup-asked-me-for-marketing-advice-i-gave-them-my-biggest-pain-instead-now-im-their-co-founder-318b31e035
- 19ih-5|Dmytro Chervonyi Claude deploy on IH|https://www.indiehackers.com/post/we-gave-claude-deploy-access-to-production-the-hard-part-was-choosing-what-to-forbid-fc8c8f7a01
- 20ih-6|Lovanaut on Indie Hackers|https://www.indiehackers.com/post/i-ran-my-whole-acquisition-loop-without-leaving-the-chat-be7ed5825d
- 21ih-7|Hannah Wolfe CoachDesk on IH|https://www.indiehackers.com/post/follow-up-i-posted-here-at-customer-zero-you-gave-me-a-lot-of-advice-here-s-what-i-did-with-it-340001dca3
- 22ih-13|EarningsScores on Indie Hackers|https://www.indiehackers.com/post/earningsscores-month-4-update-6-users-0-mrr-and-what-happens-when-your-site-goes-down-for-a-month-d64f2edc0c
Add more perspectives or context around this Post.