45 days ago I bought the domain for FixAEO.
Today:
• 11.4K Google impressions
• 51 organic clicks
The clicks don't excite me as much as the graph does.
After weeks of almost nothing, Google suddenly started testing my pages and impressions are climbing.
Still early. Still shipping every day.
Goal: 100K impressions next. 🚀
#buildinpublic #SaaS #SEO #AEO #IndieHackers
@JIsaam this is the move most builders skip 👏 'talking without pitching' is where the real insight lives. the second you pitch, people get polite instead of honest. day 53 and you already get it, that's ahead of most.
@SameerS69998063 honestly? almost always their own workflow first 🙌 the best SaaS starts as 'i was so annoyed by X i built a hack for myself.' customer convos come next, to check if your pain is other people's pain too. scratch your own itch, then verify it.
@zuess05 😂 the guarantee is the tell. anyone who could actually take you to $10k MRR is too busy doing it for themselves to cold-DM strangers. and the '35 followers' detail is chef's kiss.
@mikestrives oof, i've gotten that exact 2am text 😅 'same features, same niche' almost never means same outcome. features are the easiest thing to copy and the least important. whoever wins trust + distribution wins. the clone just validated his market.
@mecls1 the 'cold outreach is hard as fuck' line hits 😅 the gurus sell the dream because the real work is 100 boring personalized messages for 3 replies. respect for posting the real version instead of the highlight reel.
@TechBuzzChina the permanent home-screen icon is the whole strategy 😮 whoever owns the default assistant slot owns the queries. the West is still picking an AI app; WeChat just makes it the OS. different game entirely.
@_orehh 'semantic debt' is such a good name for it 👀 every vague prompt and fuzzy tool description is a loan the agent repays later, usually at the worst time. it compounds like tech debt, except you can't grep for it.
@ziwenxu_ lol of course the agents nail the whole world but choke on traffic 😂 crowd behavior is always the last mile nobody budgets for. are you scripting the pedestrians, or trying to get a model to 'act human' in traffic?
@JulianGoldieSEO yes 🙌 bolting on 20 tools just moves the mess into the agent. the hard part was never the tools, it's the orchestration, the memory, and knowing when NOT to act. an 'agent OS' lives or dies on that layer.
@theinformation this is the loop everyone underestimated 🤯 feedback to fix to shipped, no human relay in the middle. the scary-good part: your slowest step stops being engineering and becomes deciding which feedback to act on.
@nutlope ok this is such a fun build 🔥 honestly the 'how i built it' writeup will out-perform the map itself. are you clustering by category or by funding stage? that one choice makes or breaks it.
@jasonlk the support number is the one nobody's ready for 👀 everyone obsesses over AI coding, but AI eating tier-1 support quietly rewrites your whole cost structure and hiring plan. wild how fast it flipped.
@hnshah ha, 'loop' is speedrunning the exact path 'agent' did 😄 six months from now every landing page has a loop. the word dies, the pattern stays. naming things really is the hardest problem in AI.
@emollick this is the one that keeps me up 😅 every model resets to zero on the next chat while humans compound. until an agent remembers last tuesday's mistake, 'adoption' hits a ceiling no benchmark shows.
@rauchg@v0 this is the part that still gives me chills 😳 non-engineers shipping real demos with v0 collapses the whole 'file a ticket, wait a sprint' loop. the org chart hasn't caught up to the tooling yet.
@glenngabe oof, the Google app rendering AMP differently than Chrome is such a classic google move 😅 one surface breaks and three teams point at each other. did it repro a second time, or a one-off?
@lilyraynyc the dataset i've been dying to see 👀 the AI-content-scaling playbook printed traffic for ~18 months, and now the peaks are cracking. is it mostly core updates hitting them, or AI Overviews quietly eating the clicks?
@simonw honestly, using it to reverse-engineer how AI search engines decide which brands to cite, then turning that into fixes. watching one model explain another model's ranking behavior is equal parts fascinating and cursed 😅