Every app you install is a monthly cost, a page-weight cost and a dependency.
A new store has no traffic data, so most app decisions at launch are guesses — defer everything that optimises something you cannot yet measure.
Four categories genuinely earn a place on day one; almost everything else can wait until you have real behaviour to act on.
Apps you install and abandon frequently leave code in your theme, so trialling freely has a cost that outlasts the trial.
Shopify covers more natively than most new merchants realise, and the first check should always be whether the platform already does it.
Fewer than most start with. Four or five covering genuinely necessary functions is a reasonable launch position.
The full guide has the formats, the deadlines and the worked examples.