Channel selection: meaningful reach versus raw audience size
Reviewed: 2026-10-08. Read this before recommending social platforms, allocating posting effort, or claiming that any network will deliver the most views. This is a strategy reference, not a publishing capability declaration. See the bundled platform playbooks, measurement guidance, and the full repository’s platform matrix where available. A copied standalone skill may not include repository-level docs.
Non-negotiable decision rule
Section titled “Non-negotiable decision rule”Do not equate a platform’s users, registered accounts, ad reach, or total video plays with the reachable audience of a new personal account. Optimize for the owner’s intended audience and outcomes, not just a popularity table. Confirm brand, audience, region, language, content/production budget, account authorization and primary desired action before promising a plan; state safe assumptions if unavailable.
- Choose one primary outcome: qualified professional inquiry, GitHub contributor, technical adopter, paper readership, website signup or broad consumer awareness. Keep raw impressions/views as diagnostic unless broad awareness is explicitly the goal.
- Choose one flagship medium (e.g., real screen-recorded demo + useful canonical documentation), then adapt native pieces instead of identical spammy cross-posting.
- Evaluate channels on audience fit (most important), discovery potential, proof/media fit, observable outcomes, production time and actual permission to publish. Mark unknown reach or conversion as unknown, not 0 or a fabricated score.
- Pick two main channels + one bounded experiment channel. Re-evaluate on native evidence at 7 and 28 days. Do not stretch limited capacity over twelve profiles.
- For HN, Reddit and private communities, human participation and community rules outrank reach. No automatic generated submissions, brigading or mass DM.
2026 audience scale: not interchangeable metrics
Section titled “2026 audience scale: not interchangeable metrics”| Channel | Checked scope of scale | What the count means | Primary source |
|---|---|---|---|
| YouTube | Billions of monthly logged-in users; Shorts averaged 200 billion daily views in its 2026 CEO update | Views are repeated plays, not unique people | YouTube 2026 letter, press data |
| 3 billion monthly active users (announced September 2025) | Worldwide MAU, not organic impressions per account | Meta announcement | |
| More than 1.3 billion members (April 2026) | Registered members, not monthly active users | LinkedIn Q3 FY26 update | |
| 514.6 million weekly active uniques; 130.3 million daily active uniques (Q2 2026) | WAUq and DAUq, not MAU | Reddit Q2 2026 | |
| Threads | 500 million monthly active users (June 2026) | Worldwide MAU | Meta Threads update |
| Snapchat | 971 million monthly active users (Q2 2026) | Worldwide MAU; less direct fit for specialized B2B tech | Snap Q2 2026 |
| Over 600 million monthly active users (July 2026 corporate statement) | Worldwide MAU; visual-search intent differs from technical networking | Pinterest company update | |
| TikTok | Over 200 million monthly users in Europe (September 2025) | Europe-only, not a global MAU claim | TikTok European newsroom |
| Facebook, WhatsApp | Widely used globally; no like-for-like platform count asserted in this report | Consult the same methodology/time window for relative scale; messaging does not equal public discovery | DataReportal 2026 midyear |
| X, Bluesky, Mastodon, Telegram, Hacker News, DEV | No independently verified comparable MAU in this assessment | Do not invent a count or treat account registrations as active attention | Official provider/community sources before use |
Cross-platform context: DataReportal’s April 2026 report estimates 5.79 billion social media user identities globally; identities are not necessarily distinct people. Its Similarweb mobile app sample finds YouTube the leading social app by active handset users, followed by WhatsApp, Instagram, Facebook and TikTok; its China exclusion, panel bias and measurement methods matter. Platform ranking is not account-level forecast. DataReportal midyear 2026.
Recommended channel allocation for engineering, AI, open source and research
Section titled “Recommended channel allocation for engineering, AI, open source and research”Priority indicates likely qualified attention for this kind of work, not global popularity. Confirm actual audience and revise after measurement.
| Tier | Channel | Reason to prioritize | Initial creative asset | Delivery / consent note |
|---|---|---|---|---|
| A1 | LinkedIn (personal professional presence) | Reach engineering peers, research collaborators, business decision makers and prospective clients | Verified field lesson, working demo, architecture trade-off, useful PDF document | Manual post in current ReachWeaver; access for member/company APIs not verified |
| A2 | YouTube (search + Shorts) | High overall discovery, evergreen searchable tutorials, proof-by-demo | 3–10-minute reproducible tutorial + derived original short clips | Script/storyboard only until video actually produced and authorized uploader verified |
| A3 | Reddit (relevant subreddits) | Questions and feedback from intent-rich niche practitioners | Non-promotional troubleshooting answer, reproducible demo or community-approved project review | Human-led; check each subreddit’s rules, disclose affiliation, never mass-crosspost |
| A4 | Hacker News / Show HN | Concentrated technical early adopters; launch events can create short-lived discovery spikes | A finished, accessible, nontrivial interactive project with author-written explanation | HUMAN-ONLY writing and submission; HN disallows generated/AI-edited posts and automation |
| B1 | Instagram Reels | Very large, broader non-specialist discovery if visual storytelling works | Original 20–60-second on-screen before/after demonstration with subtitles | Requires real media and proper Meta permissions; current skill drafts only |
| B2 | TikTok | Large interest-based discovery beyond existing follower graph, especially short demos | Short visible transformation + searchable plain-language explanation | Current skill scripts only; audit/app rules may prohibit public API posts |
| B3 | X and Threads | Short-form public conversation about AI, open source, research and startups | Self-contained proof-led finding, conversational follow-up | Draft/manual; platform API scope and fees require recheck; Threads reported 500M MAU |
| B4 | Bluesky / Mastodon | Smaller but sometimes dense software and research communities | Concise reproducible note, honest question, source link | ReachWeaver text adapters mock-tested; real accounts not yet verified |
| B5 | DEV Community, GitHub Discussions, relevant project forums | Evergreen problem-solving + valuable developer discovery, not general mass exposure | Original technical tutorial with runnable steps and affiliation disclosure | Human editorial review and community rules; don’t flood identical articles |
| C | Facebook Groups, Pinterest, Snapchat | Worth a bounded test only when an audience or media format specifically matches | Targeted community explanation, infographic or demo | Do not prioritize solely because MAU is large |
| Retain | Discord, Telegram channels, WhatsApp groups, email newsletter | Community depth, subscriber retention and conversion | Opt-in release notes, demos and high-signal updates | Mostly retention/distribution, not cold public discovery. Avoid unsolicited invites/DMs |
Regional exception: For Chinese-speaking users consider Bilibili (technical video), Zhihu (technical Q&A), WeChat public accounts (retention) and Xiaohongshu (visual discovery), only after validating localization, account/legal eligibility and platform rules. No automatic integration is currently claimed.
Specialist exception: open-source contributors when Reddit/HN are constrained
Section titled “Specialist exception: open-source contributors when Reddit/HN are constrained”If the primary goal is GitHub-qualified visits, first-time issue authors and genuine PR contributors, and Hacker News is flagged or subreddit self-promotion is restricted, read open-source-growth.md first. Start with DEV technical posts, LinkedIn professional teaching, GitHub contributor onboarding and Bluesky FOSS discussions; test a real YouTube demo. These are goal-specific experiments, not a claim that DEV ranks highest for all social reach. In the full repository see docs/OPEN-SOURCE-DEVELOPER-GROWTH-PLAYBOOK.md.
Goal-specific default choices
Section titled “Goal-specific default choices”- Consulting/B2B leads (e.g., AI infrastructure services): LinkedIn + a proof-backed YouTube demo; small Reddit/DEV engagement experiment; call-to-action to a trustworthy project page.
- Open-source adoption (e.g., Kubernetes, AI tooling, dictation software): YouTube tutorial + relevant developer community; LinkedIn for professional trust; one community-compliant HN launch only if the project is meaningfully usable.
- Research/paper impact: LinkedIn technical note + a clear explainer on YouTube; specialist research communities where authors can answer questions directly.
- Broad consumer attention: YouTube Shorts + Instagram Reels, test TikTok once actual video production is sustainable; compare qualified traffic not only loops.
- Existing subscribers and community: Keep Telegram/Discord/newsletter as opt-in updates while prioritizing other platforms for acquisition.
Practical first 14 days (experiment template, not universal posting prescription)
Section titled “Practical first 14 days (experiment template, not universal posting prescription)”- Day 0: record per-account baseline and origin of metrics; collect source proof, audience questions, legal/brand approvals and current API restrictions.
- Week 1: publish two different, genuine LinkedIn learning artifacts; one real YouTube demo (or a production-ready storyboard when video cannot be made); contribute one human-written answer to a relevant community only where rules permit.
- Week 2: publish a distinct follow-up case study or honest limitation, and test one original 15–60-second video adaptation on a permitted short-form channel. Reserve time for responses to real questions.
- Compare native metrics at equal age (24h and 7d). Record watch time / engaged views, qualified clicks with consented UTM where available, genuine GitHub referrals and collaborations. For YouTube, see the August 24, 2026 views definition change in YouTube Analytics.
- Continue / pivot / stop: define the decision beforehand, compare per hour of effort, acknowledge data gaps; never claim causal uplift from two organic posts.
Output format for Claude
Section titled “Output format for Claude”Return a transparent choice table: channel, target persona, reason, feasible asset, official rule/publishing status, measured baseline, goal KPI, owner effort, confidence (documented / extrapolation / hypothesis), source URL/date, and next test. Explicitly list deprioritized popular channels and explain why. Say if account-level analytics are unavailable; global MAU is never a substitute.