Copy-paste before you integrate.

Strategy · 4 min read

A client wanted built-in messaging. Send a text to a vendor directly from the app. Send an RSVP reminder via SMS. Notify a team member with a push notification. All of it baked into version one.

I asked a simple question: "What if we gave them a pre-written message and a button that copies it to their clipboard?"

The room went quiet. Then: "That's... actually fine."

Integration is expensive

Building a proper SMS integration means paying per message. That cost scales with usage. If your app sends 10,000 RSVPs a month, you're paying for every single one. Before you've proven anyone wants the feature, you're carrying the infrastructure cost of running it.

Email integration is cheaper but still adds complexity. You need a sending service, deliverability monitoring, bounce handling, unsubscribe links. Push notifications need device tokens, permission flows, and a backend to manage it all.

All of that is worth building. Eventually. But not before you know whether people actually use the feature.

Templates prove demand

The manual version works like this. The app generates a pre-written message based on the context. "Hi [name], just confirming your availability for [event] on [date]. Let me know if anything changes." The user taps "Copy" and pastes it into their own messaging app. Done.

It's one extra step compared to the automated version. But it costs nothing to build, nothing to run, and gives you real data on how often people use it. If 80% of your users copy that template every time they plan an event, you know the feature is worth automating. If nobody touches it, you just saved yourself months of development on a feature nobody wanted.

Manual first, automate later

This applies to more than messaging. Any feature that involves a third-party integration, an API, or an ongoing cost should start manual. Let the user do the last step themselves. Export a PDF instead of auto-emailing it. Show a phone number instead of building a call button. Display an address instead of embedding a map.

The goal of version one is to prove the idea works. Every integration you add to the MVP is a risk. It's money spent before you have evidence it matters. Copy-paste isn't elegant. But it's cheap, it's fast, and it tells you exactly what to automate next.

Ship the manual version. Watch what people actually use. Then spend the money where it counts.

Related blog posts:

What is an MVP? →

The update nobody budgets for →

Phase two is not a graveyard →