Four days after launching my latest project:
• 945 Google Search impressions
• 3 organic clicks
Small numbers, but seeing people discover something I built is exciting.
Next: improve the tools, learn from real search queries, and keep building.
1.69K Google impressions. 4 organic clicks.
I went through the search queries and updated my guides and tools based on them.
Now I’m curious to see whether those changes translate into better rankings and more clicks.
Small experiments. Real data. Let’s see.
Four days after launching my latest project:
• 945 Google Search impressions
• 3 organic clicks
Small numbers, but seeing people discover something I built is exciting.
Next: improve the tools, learn from real search queries, and keep building.
@viktoroddy I still like seeing the actual product somewhere near the top. Doesn't have to be the whole dashboard. One useful screen can explain more than another headline.
@frederickjames I went web-first for my app. For something people might use once a month, being able to open a link and start without installing anything felt like a better fit.
@gemimatt@framer Got my site's mobile lab score to 100 recently too. Deferring analytics helped, but the bigger fix was letting the HTML show before startup JS finished.
@synopsi Are those visitors landing on the app homepage or the free tools? Working on SEO for my app too, and trying to separate tool traffic from people looking for the app.
@ItsPavanGK Curious whether Reddit still brought any installs from supported devices, or if the whole campaign was a loss. The click numbers alone would hide that.
@LeoBuilds_ Repo docs for decisions, a short handoff for the current task. What changed, what passed, what's left. Still need the next agent to check the actual files though.
@1Umairshaikh Been working on this too. Keyword research showed me I was describing my app differently from how people search for it. Had to fix that before writing more pages.