FWW LIVE Web App (PWA)

This past year, my wife and I shut our fitness studio down after 7 years in business. I’ll spare you the details.

We created an online streaming fitness program called Fit Women’s Weekly LIVE . It’s a unique program. Among other things, only 300 members are allowed to be active at any one time.

We decided on this membership limit because we talk to our members and give them real training help. It’s important to us to keep a close relationship with our members.

We have another program called Habit on our horizon. Habit will have an uncapped membership. Our goal for it will be 5,000 active members. And who knows… maybe we can blow it up more!?!

Currently there are 78 active members in Fit Women’s Weekly LIVE.

The app.

The app is going to be our primary lead generation method.

And obviously… are primary member retention mechanism. There must be a better word for mechanism. (And I’m a copywriter.)

Because this is a private app… I decided to keep this a web app (PWA). I plan on making it fully offline compatible over time.

We offer three FREE 14 day training plans among the paid training plans.

The FREE training plan is being tested with a downloadable PDF and it’s converting well from prospect to lead. Lead to member can use a boost… but that is what the app is supposed to help with.

Let’s get technical, here’s the tech stack.

• SolidJS
• Cloudflare Workers
• Cloudflare D1
• Cloudflare Images
• Cloudflare Stream
• Auth is undecided

This app is build on what I call the Cloudflare stack.

In this group, we stream (and record) two LIVE workouts every weekday and one Saturday stretch session. It’s done through a private Facebook Group at the moment. (Cloudflare Stream within the next 6 months.)

The app is built around 30 day training plans that use these workouts.

However, you don’t have to use a training plan. You can structure your weekly workouts using the LIVE Workouts and recordings.

I don’t want to go into the details here too much because I have some really cool ideas to implement machine learning with the data we’re collecting.

This is not your typical workout tracking app.

I’m bringing over 10 years of personal training experience to this project… so I have some fresh ideas on how to make tracking helpful.

This is pretty basic. Just a journal or personal blog of your workout days.

At the moment, nothing fancy here. Just the ability to tag journal entries and mark a journal entry as a favorite.

This is the meat of the app.

Most training plans are 30 days long. We have two special yearly training plans that are 60 days long.

The app tracks everything day to day.

Everyday you have a series of goals to complete.

• A primary goal
• A busy woman’s goal
• A fitness challenge
• And more…

Some days the primary goal is to take a rest day. So all you have to do is check that off.

• Each goal has points.
• Each combination of completed goals has points.
• 

  You can even get negative points for not completing days.

Your goal is to get as many points as possible during the training plan and see if you can crack the top 10 leaderboard for that plan.

There’s even more that makes the training plan so cool. This feature is going to be the central feature in Habit when I build that out next.

I want our members to use and rely on this feature to help them workout more. In my eyes it’s the star, but you never know what users will do.

I do have a 97% working version of the app built with React and NextJS on the frontend.

The backend is using the Cloudflare stack as I mentioned.

• Workers for edge functions
• KV for some database needs
• 

  Durable Objects for transactional and consistent data storage needs.

But this is going to change a little.

I started to learn Solid and I really enjoy it.

Once you get your head around the way signals and stores work in Solid… it makes so much more sense to update the UI compared to React and useEffect.

React is fixing most of the issues with useEffect with a new hook. I can’t remember the name of it. I don’t even think it’s released yet. Knowing React, it will probably take another year for it to come out.

But Solid is pretty awesome. And I really like Ryan Carniato . I think he is one of the smartest coders. I trust he and his team will make Solid a beast of a framework going forward.

Plus this transition also lets me double check all my logic and make improvements to features if needed.

I’ve already found some big bugs and smashed them.

I got early access to D1… Cloudflare’s distributed SQLite database. I think they call it distributed. It should be in open beta by now.

I am going to get rid of Durable Objects and use D1 instead. I think it makes a lot more sense to solve the problems I’m having with Durable Objects.

Durable Objects are pretty cool. But I find they can be tough to work with at the moment. I still plan on using them for future features.

UPDATE - I might stick with Durable Objects. Mainly because I want total flexibility with my persistance until I know what the final version of the app will look like.

I’m still trying to find a good auth solution I like.

I wanted to go with Magic.link. I really like them, but there are so many other possibilities like Stytch I’m interested in trying.

I was looking at Stytch the other day and really like it. It has API endpoints you can hit rather than be forced to use an SDK. I really like that. So this is currently in first place. I’ll blog more about this when I start playing with it.

I might roll my own OAuth2 solution with Google, Facebook, and Apple logins. Not sure but I’ve been wanting to learn OAuth2 deeply for some time now.

This deadline is approaching FAST.

But I think I can transition the app to Solid before Jan. We have one of those “special” eight week training plans coming up in early January.

Kindal and I want to use the app to msanage it. And hopefully get our membership full by then as well.