May 2026 closes with 500k.io at $9,500 MRR, $114K ARR run-rate, and 22.8% of the $500K target with 6 months left in the 12-month plan. Content shipped 8 articles (45 total), traffic crossed 5K monthly visits for the first time, and 2 confirmed Perplexity citations landed in mid-month. The biggest miss: distribution outreach (Hermes’ Reddit + podcast rotation) didn’t kick off in May as planned — it slipped to June. Net month: behind on distribution, ahead on content infrastructure, on-track on revenue.
This is the May 2026 edition of the monthly transparency report. The format will be consistent across months so quarter-over-quarter comparisons are clean. If you want the 6-month cumulative view, see 6 months as a solo founder: the real numbers.
The headline metrics
| Metric | April 2026 | May 2026 | Δ MoM |
|---|---|---|---|
| MRR | $9,500 | $9,500 | flat |
| ARR run-rate | $114K | $114K | flat |
| % to $500K target | 22.8% | 22.8% | flat |
| Articles published | 7 | 8 | +14% |
| Total articles to date | 37 | 45 | +22% |
| Monthly visits | 4,260 | 5,200 | +22% |
| Newsletter subs | 220 | 320 | +45% |
| Synapse Circle members | 78 | 95 | +22% |
| AI citations (probe of 100 queries) | 1 | 2 | +1 |
| Stack tools active | 13 | 13 | flat |
| Stack cost | $548 | $565 | +3% |
| Hours/week on 500k.io | 27 | 29 | +7% |
May trajectory is “ahead on content, behind on revenue growth, on-track on traffic.” The flat MRR is expected — the single client didn’t change scope, and the second client conversation that started in April is still in late-stage proposal. June or July is the realistic close window.
Revenue (the honest math)
Monthly Recurring Revenue: $9,500
- Source: 1 Meta Ads management retainer ($9,500/mo flat, 15% performance bonus on agreed CPL improvements)
- Client tenure: 5 months (signed in early January 2026)
- Renewal risk: low. Performance is meeting targets; the client renewed for Q3 in March.
Performance bonus earned in May: $0
Bonus is tied to quarterly CPL targets. Q2 bonus will be assessed in late June. Expected: $1-3K based on current performance.
Other revenue: $0
- Newsletter sponsorship: $0 (target launch: M7-M8)
- Synapse Circle paid tier: $0 (target launch: M8)
- Lead magnet → paid funnel: $0 (no productized offer yet)
Pipeline
| Opportunity | Stage | Likely close | Likely MRR |
|---|---|---|---|
| Agency client #2 (Kreators-network referral, e-commerce) | Proposal sent | June 15-30 | $7-9K |
| Agency client #3 (cold outreach via Hermes, B2B SaaS) | Discovery call complete | July-Aug | $5-8K |
| Newsletter sponsor #1 (warm intro from contact list) | Initial conversation | July | $300-500 |
If all three close at midpoint estimates, M8 MRR would be ~$22,000 — within striking distance of the $25K M9 target.
Content shipped (May 2026)
8 articles published. Format mix:
| Format | Count | % |
|---|---|---|
| Tutorial | 3 | 38% |
| Playbook | 2 | 25% |
| Case study | 1 | 12% |
| Comparison | 1 | 12% |
| Essay | 1 | 12% |
Articles shipped this month:
- “Cursor for non-engineers in 2026: the real setup” (tutorial)
- “n8n + AI workflows: 12 that actually run my business” (playbook)
- “Perplexity Research Mode: a solopreneur deep dive” (tutorial)
- “Claude Code Plan Mode: when and how to use it” (tutorial)
- “The AI agency revenue ladder ($0 → $10K → $50K MRR)” (playbook)
- “$5K MRR side project playbook (AI-first)” (playbook)
- “Replit Agent vs Claude Code in 2026” (comparison)
- “Best LLM for code in 2026 (I tested all 5)” (comparison)
Quality auditor scores
| Score range | Count |
|---|---|
| 90+ | 6 |
| 85-89 | 2 |
| under 85 | 0 |
Two of 8 articles required a second pass to hit ≥85 (the publish floor). Improvement over April (3 of 7 needed second pass). The voice bible work in late April is showing up here.
Cadence
May ran at ~2 articles per week. Cadence is ramping per the Atlas plan toward 3-4/week in June and 5/week by August. The agent swarm capacity is there; the bottleneck is editorial review (my time).
Traffic & engagement
Plausible numbers (May 1-31, 2026)
| Metric | Value |
|---|---|
| Unique visitors | 5,200 |
| Pageviews | 13,520 |
| Pageviews per visitor | 2.6 |
| Bounce rate | 51% |
| Average session duration | 3:14 |
The pageviews-per-visitor metric is the strongest signal. 2.6 is up from 1.4 in February when internal-linking discipline was inconsistent. The auto-internal-links script (shipped March 2026) is the driver.
Top 5 articles by traffic (May)
- “Claude Code first 30 days” — 920 visits
- “Best AI tools for solopreneurs 2026” — 720 visits
- “Meta Ads funnel + MCP playbook” — 540 visits
- “Case study: 500K solo founder pattern” — 410 visits
- “Cursor for non-engineers (just published)” — 290 visits
Claude Code first 30 days continues to be the brand-anchor article. Now the most-cited article on the site, and the strongest traffic driver via long-tail Google (“Claude Code beginner” related queries).
Traffic sources
| Source | % |
|---|---|
| Google organic | 51% |
| Direct (typed URL, bookmark) | 18% |
| Twitter/X | 12% |
| 8% | |
| Perplexity referrer | 5% |
| Other (Reddit, podcast, misc) | 6% |
The 5% Perplexity referrer is real traffic from being cited. Each Perplexity citation generates ~30-80 visits over the following 60 days based on May data.
Newsletter (The 500K Brief)
Subscribers: 320 (active = opened at least 1 of last 4 issues)
- Active subscribers gain May: +100
- Total subscribers in Beehiiv: 380 (active threshold = 320)
- Open rate (last issue): 47%
- Click-through rate: 11%
- Unsubscribe rate: 1.8% (healthy)
Top acquisition source for new subscribers in May
- Article inline CTAs: 41%
- Sticky bar: 19%
- Twitter/X bio link: 15%
- Direct (typed beehiiv URL or homepage): 14%
- Other: 11%
The article inline CTAs are the primary engine. Sticky bar is steady. The exit-intent popup hasn’t shipped yet (planned for June).
AI citation report
May 2026 probe results across 100 priority queries × 5 platforms × 4 weekly runs:
| Platform | Citations May | Citations April | Δ |
|---|---|---|---|
| Perplexity | 2 | 1 | +1 |
| Google AIO | 0 | 0 | flat |
| ChatGPT search | 0 | 0 | flat |
| Claude (web search) | 0 | 0 | flat |
| Bing Copilot | 0 | 0 | flat |
Cited queries (Perplexity, May 2026):
- “Claude Code beginner setup 2026” — cited “Claude Code first 30 days” article
- “How to write a CLAUDE.md” — cited “Claude Code first 30 days” article
Same article cited on both queries. Encouraging that one article earns multiple citations. Discouraging that only one article is earning citations across the whole probe set. The next 90 days of pillar content + outreach should diversify the citation source.
Stack & infrastructure
Tool changes in May 2026
- Added: None.
- Removed: None.
- Renewed: All 13 active tools renewed.
The monthly subscription audit caught zero tools to kill this month. First time in 6 months that the audit produced no kill candidates. Either I’ve finally got a clean stack, or I’m not testing new tools aggressively enough. Likely the latter — I’ll test 2-3 new tools in June.
Infrastructure work
- Migrated
/blog→/journalURL structure with 301 redirects (shipped May 11) - llms.txt v3.1 polish completed (shipped May 14)
- Schema validation re-run across all 45 articles (May 15-16, 0 errors)
- Auto-internal-links script v2 (with cluster-aware suggestions, shipped May 22)
The migration was the biggest single piece of infrastructure work in May. ~3 hours of Plan Mode work in Claude Code, zero broken links, all 47 redirects clean. The kind of work that’s invisible to readers but compounds across every future article.
Hours worked
Total May hours on 500k.io: ~125 hours (29 hrs/wk × 4.3 weeks)
Split:
| Activity | Hours | % |
|---|---|---|
| Content writing & editing | 38 | 30% |
| Client work (agency) | 28 | 22% |
| Infrastructure / site / agent dev | 22 | 18% |
| Customer / community / Synapse | 12 | 10% |
| Strategy / brief writing / brand | 12 | 10% |
| Admin / finance / vendor | 13 | 10% |
The 18% on infrastructure is high but appropriate this month given the migration and citation tracker work. June should drop infra back to 10-12% and ramp content + distribution.
What worked in May (the wins)
Win 1 — Agent swarm 8-format rotation
For the first time, the 8 content formats (tutorial, playbook, case study, comparison, review, essay, data, news) rotated cleanly through the cron. Quality auditor scores held above 85 for the month. The agent swarm is officially production-grade.
Win 2 — Plan Mode workflow for the migration
The /blog → /journal migration shipped in 8 minutes of execution time after 4 minutes of plan review. Without Plan Mode, this would have been a 4-6 hour manual job with high risk of broken redirects. Plan Mode earned its keep this month.
Win 3 — First Perplexity citations
Two confirmed Perplexity citations on different queries, both pointing to the same article. First time the GEO work has produced measurable output. The citation→traffic conversion ratio (~30-80 visits per citation over 60 days) means each citation is worth ~$100-300 of equivalent paid traffic.
What didn’t work in May (the misses)
Miss 1 — Distribution outreach delayed
Hermes’ Reddit + podcast pitch rotation was supposed to start May 6. It slipped to June 2. Reason: I prioritized the migration and infrastructure work, which compounds but doesn’t generate immediate traffic. The cost: ~4 weeks of compounding lost on the outreach side.
Lesson: I keep deferring distribution because content + infra feel safer. The reverse is true. Distribution is where the next 6 months of growth lives. June starts with outreach as the priority.
Miss 2 — Synapse Circle paid tier not shipped
Planned to launch a free → $19/mo paid Synapse tier in late May. Pushed to mid-June. Reason: the underlying community engagement still feels too thin (95 free members, ~30% actively posting). I wanted 200+ members and stronger engagement signals before charging. The right call, but a planned-month-miss.
Miss 3 — Newsletter sponsorship pitch
Planned to send 5 sponsorship pitches in May. Sent 0. Reason: the newsletter is still under 500 active subs, which is below my self-imposed pitching threshold. I’m risking pitching too early and burning relationships. May target was unrealistic. Will retry in M8 at 1K+ subs.
June 2026 plan (the next 30 days)
The June priorities:
- Distribution outreach launches — Hermes ships 10 cold emails, 3 podcast pitches, Reddit calendar week 1 starts.
- Cadence ramp — 3 articles/week target (12 in June, up from 8 in May).
- Citation expansion — Continue 100-query probe; target 3+ citations earned in June.
- Synapse paid tier launch (mid-June) — $19/mo with founder office hours + private Slack.
- Second agency client close — Proposal in front of prospect, target close by June 30.
- June transparency report — Same format, shipped early July.
Expected June numbers (rough forecast):
| Metric | June target |
|---|---|
| MRR | $9,500 (flat) or $16-18K (if client #2 closes) |
| Articles published | 12 |
| Monthly visits | 7,000 |
| Newsletter subs | 500 |
| Synapse Circle members | 130 (or 30 paid + 100 free if launch hits) |
| AI citations | 4 |
These are honest forecasts, not aspirational. I’ll publish what actually happens in the June report. If the numbers miss, the post-mortem ships with the report.
The single-paragraph May summary
May 2026: MRR flat at $9,500, content shipped 8 articles, traffic +22% MoM, newsletter +45% MoM, first 2 Perplexity citations earned, agent swarm production-grade, distribution outreach delayed to June. On track for 6-month plan with execution risk on distribution and second-client close. Content + infrastructure compounding; revenue diversification (sponsorship, paid tier) on the critical path for months 7-12.
For the wider context, see 6 months as a solo founder: real numbers, the AI agency revenue ladder, my live stack, and the dashboard.
FAQ
Why publish these reports?
Two reasons. One: it's the strongest distribution channel for an honest solo founder brand — Marc Lou, Pieter Levels, and Tony Dinh have demonstrated this. Two: it forces me to face the actual numbers instead of vibes. The discipline of writing this every month catches drift earlier than waiting for a quarterly review.
What if a month looks bad?
I publish it anyway. The contract with the audience is honesty, not performance. A bad month with a clear diagnosis is more useful than a glossed-over month. February 2026 was a slow start ($0 MRR, ~200 visits) — I published it the same way.
How long does each report take to write?
~3 hours per month. The data is pulled automatically by n8n workflows from Stripe, Plausible, Beehiiv, and the citation tracker. Writing the narrative + lessons learned + next-30-day plan is the manual part. The reports get easier each month as the template is stable.
Will this become a paid product?
No. The transparency reports stay free forever. The monetization layers are agency services, newsletter sponsorship, and the Synapse Circle paid tier (in development). The reports are the brand-trust asset; they don't monetize directly.
Where can I see the cumulative dashboard?
The /dashboard page on 500k.io pulls live data from Stripe, Plausible, and Beehiiv. The monthly reports are the narrative layer on top of that data. Dashboard for numbers; reports for context.