50. Non-tech dad of 2. Building with AI after work. 4 products shipped. Workflows, launches, lessons & mistakes. See what I am up to: lucacapone.io 🇮🇹
Italy & Luxembourg
- Tweets 10,952
- Following 773
- Followers 1,556
- Likes 5,809
Pinned Tweet
Starvling is on Product Hunt today.
Day one, honestly: 6 creatures on the board, two of them listed by people I've never met. 20 feeds from 7 people.
Nothing has died yet. PairHabit, mine, is at 42% and goes first, tomorrow around 05:00 CEST, unless strangers feed it.
Feeding is free, one click, no signup. Rank can't be bought. producthunt.com/products/sta…
13
22
1,885
Parents, this one's yours. When the school emails pile up and you go looking for an app, what do you type into the App Store search box?
My bet is "family calendar". I'd rather have your words than my guess before I build a page around them.
39
Kikko is for the parent whose inbox is mostly the school.
Its store page starts: "Forward one school email." Then an event and a what-to-bring reminder.
This week I'm learning what they type to find it.
apps.apple.com/app/id6789952…
43
Zero. That's what I'd done by Monday, three weeks after shipping Kikko, an app for parents, to the App Store, to put it in front of one. I'd never built that part, for any product.
Where did your last 100 users actually come from? One source is a full answer.
36
My distribution system, as I'm writing it this week:
A weekly list. Where a parent looks for an app like Kikko, my app for the school inbox. Which words they type. Which page of mine answers them.
Measured by what I can read: clicks and store page views.
2
1
62
19 months building products and never building distribution for one of them.
Kikko, my AI family assistant, went on the App Store on 14 September. By Monday I'd done nothing to reach a parent.
This week the build is the system, SEO first. I'm learning it here as I go.
33
Starvling had the day, two weeks after I stopped it: what a dead board still sends, what stays, what goes.
Once the code's in the archive folder, that's the week's first kept promise.
Distribution starts tomorrow.
1
2
76
Nobody, me included, has pressed Starvling's real kill switch.
It's the $9 price. While it's live, any of the fifteen dead listings on my launch directory can be bought back. Archiving it is a separate decision, on its own day, if ever.
42
Evening work in Rome, with the kids asleep: Starvling's teardown.
Whatever's left on the list gets tonight, if anything is: a metrics schema to drop, the home page copy.
Reading meanwhile: SEO. Distribution is the part I've never built.
48
Starvling after the verdict: what stays, what goes.
The site stays: fifteen graves, the clicks-out counter, the $9 button that brings one back.
The home page line selling a $9 listing on a dead board goes. Nobody ever paid for one.
starvling.com
43
Fifteen graves on Starvling, my launch directory, each showing its feeds. Kikko 46, PairHabit 29, Précis 24, smallest 10.
Précis turned 24 feeds into 67 clicks to its own site by 20 September, the best return on the board. Every grave still carries the maker's link.
27
A promise I made here on 22 September went two weeks without being kept: Starvling's code goes to the archive folder. I'd repeated it on the 25th, then went quiet.
What's the oldest promise in your own posts you still haven't kept?
38
Why one Starvling counter still moves and the other can't.
A feed needs a living creature. Last death 16 September. Feeds: 297, frozen.
A click is a visitor pressing a maker's link on a grave. Clicks: 297 as of Monday.
Graveyard's public, every grave shows its feeds.
30
Dead since 16 September, and the board still sends people on.
Starvling, the launch directory I stopped on the 22nd, sent 93 more clicks to makers' sites between 20 September and Monday morning.
72
Today was for saying it: ten days gone, two promises from September not kept.
Starvling's folder move comes first of the two, this week, and gets its own post once it's done. Rome, kids in bed.
53
As of this morning Starvling's code hadn't moved. The launch directory I stopped on 22 September was still where it was built, not in the archive folder I promised that week. Two weeks late.
It moves this week. The post saying so waits until it has.
46
Tonight in Rome, once the kids are in bed, first evening back at this desk.
On it, two September promises: Starvling's code, a stopped launch directory, into the archive folder, Kikko in front of parents. Then the X routines, LinkedIn, Substack. None done as of 09:10.
1
93
Kikko, my AI family assistant, is three weeks on the App Store today. Forward a school email, get the event.
Distribution by this morning: none. Promised on 25 September for last Monday. This week it does.
apps.apple.com/app/id6789952…
1
1
71
On 26 December I wrote here that X's algorithm would punish me for radio silence.
After nine silent days, read this morning at 09:07:
Followers, 1,553 to 1,552.
Impressions, last seven days: 928. Last posting week: about 15,000.
Profile visits: 6.
Posts that week: 0.
4
5
200
Nine days off X after posting ten a day since August. Not a decision: the posting runs off one laptop, and it was in a bag.
Longest you've gone dark on a channel you'd built a habit on, and what did it cost you?
38