Momentum is building on multiple fronts.

Topics for this week:

• Wrong Shortcut

• Brand Growth

• Binary Management

• Validated

• New Bananas

Wrong Shortcut

With Wave finally live on both platforms, iOS and Android, it’s on to the next big milestone.

Payments.

Stripe is the goto choice, but Superwall may be an option depending on the final payment implementations.

As far as transactions go, Stripe seems like the clear winner.

I was hoping to get a really quick prototype up and running by using Stripe’s Payment Links.

Create one product with a set price, and Stripe provides a link to a website that they host.

Wave would merely have to route to that link instead, no payment processing code in the app at all.

But 2 major problems with that.

First is flexibility.

Each venue will have it’s own payment options like “Table” and “Cover”, and they’ll need to manage their own inventory or capacity based since Wave won’t be fully integrated into their venue’s occupancy… yet.

So the venue needs to keep a count and decided when and how much they can allow to be sold on the app.

Plus they need the freedom to change the price.

Stripe Links are all predefined products with set prices, so that would require a ton of duplicate generic products at interval prices:

• Venue Cover - $15

• Venue Cover - $20

• Venue Cover - $25

• Venue Cover - $30

• and so on…

Then the venue manager would select the price from a list rather than enter it themselves.

A simple enough system to get started, but there’s one other major flaw here.

All money would go through the Wave Stripe account.

Then we would have to keep track of every transaction, the percent cut Wave would keep, and send the rest of the money to the venue regularly.

This would be a nightmare, and according Gemini, potentially a legal issue.

So Stripe Links are out, and the old-fashioned Stripe API is in.

Using the API we can have the venue sign up for their own Stripe Express account that can connect to Wave’s Stripe account.

With that, every transaction can be appropriately handled by Stripe at the time of processing, splitting Wave’s cut and the venue’s remaining transaction amount to our separate accounts without the money passing through 2 different bank accounts, avoiding any legal issues, and debts to pay ever week, 2 weeks, or month.

Sometimes a shortcut seems like a quick win in the beginning but could lead to way more work in the long run.

Brand Growth

Keeping the live stream streak alive on all the platforms, with I believe the 5th week in a row.

Still enjoying streaming at this point too, and they have remained surprisingly active.

Been on a good stretch where at least 1 chatter shows up for the stream which helps a lot.

I’ve been streaming off and on for years now which is crazy to say, so at this point I’m pretty comfortable rambling to myself, pretending someone is there.

But it is much easier when someone is actually chatting.

And the regular streams are leading to slow growth.

YouTube is still carrying most of the weight but it’s going well.

As of today:

• YouTube: 950 subscribers

• Skool Members: 294

• X Followers: 96

• Twitch Followers: 18 (with 3 Prime Subs!)

I recorded a video for YouTube yesterday but somehow completely forgot to finish editing and upload it so that will be happening today, and should hopefully get a good push to 1000 subs for next week 🤞

But it’s hard to believe it’s almost there.

Also crazy to see almost 300 members in Skool.

The YouTube traffic is working.

If I can stay consistent, I’ll be looking back on these days fondly when my numbers were below 4 digits.

Binary Management

Speaking of consistency, I’ve been fairly consistent frustrated with n8n lately.

It might be an expectation issue on my end but the last few times I’ve sat down to build something, I think it will only take 10 or 20 minutes, and more than an hour later I’m not much further than where I started.

I had a horrible time trying to get n8n to let me add 3 images to an API call to Nano Banana (Gemini 2.5 Flash Image Preview).

n8n just does not have a good way to handle multiple binaries at the same time without duplicating them too many times or not giving a decent way to split them out again reliably.

The goal was take these images:

And combine them:

And the only way I could get this to happen was starting the flow like this:

As a programmer, this pains me deeply.

DRY is a general guideline of “Don’t Repeat Yourself”.

If you are going to use something more than twice, make it a variable.

Having to create 3 explicit routes to pull out “Image_0”, “Image_1”, and “Image_2” when it should very easily be Image_{{$itemIndex}} was driving me insane, but there just wasn’t a straight forward way to do this otherwise that I could find, even using the “Code” node in n8n which let’s you run custom javascript.

Plus the more Code nodes I use, more I wonder why I’m even using n8n in the first place.

So after over an hour, I settled for that 3 image path to get the proper result.

As mentioned the video is recorded going through the full workflow and mentioning my frustrations.

Should be uploaded by Monday at the latest.

But lesson learned, don’t count on n8n for binary management.

Validated

Somehow it’s been over a year since referencing Gary Vee’s predictions for live streaming opportunities in the future.

I’ve been banking on it more and more, and finally getting into a groove of actually streaming myself before worrying so much about repurposing all the live stream content to other socials.

And this week my commitment to live streaming was further validated as the right direction.

Alex Hormozi did his first full live multistream on YouTube where committed to focusing on regular live streams.

He’s streamed on TikTok and Instagram before, but not usually on YouTube, except for his recent book launch.

But now he’s multistreaming.

And he decided to focus on streaming after talking to Mr. Beast.

Mr. Beast noticed while hosting a number of live events, when introducing the various creators, short-form creators get some excitement from the crowd, long form creators like YouTubers get a much bigger reaction, but live streamers get the biggest response.

Which makes sense, viewers become much more invested with live streamers because it feels more personal even if it is “parasocial” or one-sided.

It builds trust more than any other form, but it also takes the most time and effort.

From this I’m still seeing a major opportunity in building tools to help with this idea.

Make live streams as valuable as possible.

I’ll keep streaming for now with even more motivation to get repurposing tools up and running.

Seeming more and more like the right direction.

New Bananas

Nano Banana has been the image model hype for the last couple weeks.

That’s what I was using in the section above, and there’s a ton of search traffic for it on YouTube at this point.

But there’s arguably a new image model winner: ByteDance’s Seedream 4.0

Artificial Analysis has Seedream 4.0 as #1 and LMArena has it at #2 but it’s closing in on #1 for them as well.

I’m sticking with videos about Nano Banana for now, specially since it can be accessed through OpenRouter making it very easy to plug into an n8n workflow, but I will give Seedream a shot in the near future.

Never hurts to have multiple high quality image options.

And that’s it for this week. More features, more live streaming, more image models.

If you want to start a newsletter like this on beehiiv and support me in the process, here’s my referral link: https://www.beehiiv.com/?via=jay-peters.

I also have a Free Skool community if you want to join for n8n workflows and more AI talk: https://www.skool.com/learn-automation-ai/about?ref=1c7ba6137dfc45878406f6f6fcf2c316

Let me know what else I missed! I’d love to hear your feedback at @jaypetersdotdev or email [email protected].

Thanks for reading!