@zakisbuilding@covabundle yeah 1 trial out of 37 can't call anything. one more trial doubles your rate, so I'd give it another week before moving budget
@abduzeedo If it's the Oct 13 agreement update, you accept it on the Account page of Apple's developer site, not in App Store Connect. Only the Account Holder sees the button
@aleksanderwco you can ask Apple for an expedited review, it's under Contact Us on their developer site. 4 users stuck on a dead-end paywall is exactly what it's for
@heyitzami restore purchases is the one that bites. the other I'd read line by line is any price string in a non-US storefront, since a wrong currency format ships quietly. do you test paywalls on a sandbox account outside the US?
@pedrito_dev they work when people already search the App Store for what your app does. I'd start with a small daily cap and pause any keyword that spends with 0 installs
@agilelabspl For those 3 apps, a word already in the title or subtitle is indexed, so repeating it in the 100-character keyword field wastes space. Swap those for words that appear nowhere else in the listing. Which of the three has the fewest impressions right now?
@joncphillips The inbox also hands you the words people use for the problem, which beat anything you'd write for the landing page. Do you have one on yours, and who reads it?
@SimonOniOS Poland, Spain and France all land at 52 to 55% tap to install (40/73, 30/58, 19/36), so the page converts about the same in each. The missing countries are an impressions gap, which fits raising the bids. Are you raising all six at once or one at a time?
@SimonOniOS That's +$40.80 over the three days. I'd check how many of the 20 countries the paying users come from, since a few can carry the rest. Which ones paid?
App makers: which part of your store presence have you gone longest without touching?
Keyword field, screenshots, review replies, ad account. Tell us which one, and roughly how long.
Posting 3 to 8 times a day across 3 accounts is a full day's work, so decide now what number says it is working, for example installs per 1,000 views. For the ads pillar, cost per trial gives the 30 to 45 days of budget something to be judged against. Which number would make you drop a pillar?
A rule that helps is writing down, before you start, the number that would tell you this one isn't working, such as installs after 30 days on one channel. Without it, every idea on the timeline looks better than the project you're in, because it has nothing to be measured against. What number would you pick for the one you're on?
@jdluk87 Seven apps gives you a comparison most of us lack. Where did the first 100 users of each come from, and was it the same place twice? For an app, store search is one candidate, and there the subtitle and keyword field decide who finds you.
@iamericpereira The blocked-account part is the one to plan around. Reddit, Hacker News and Indie Hackers tend to judge new accounts hard, so replies and comments in the weeks before count for more than the post on the day. Did he say how old those accounts were at launch?
Two things worth setting up in week one: a campaign with Search Match on, to see which terms Apple serves you for, and a separate exact-match campaign for the keywords you already want. The terms that convert can move across, and the ones that waste taps become negatives. Which country did you start with?