twelve70
Men's Outfit Calculator & Closet Inventory System
Over the past 6 years (crazy to say that) building in a niche consumer SaaS space, something became clear.
Users don't get excited about the shiny new feature that we spent months developing. They get excited when things feel simpler.
No joke, we built an 'outfit calculator' to generate outfits for you using an algorithm... it got decent use. Then we put a button to just randomize the clothes on an illustrated mannequin...boom, they loved it.
Not saying that we solved our retention problem, but it got us thinking about our plan differently.
TLDR: Clarity beats capability.
Have you noticed anything similar? Anything that actually moved retention for you?
A few weeks ago, I received an email from ClearDB, the database service we’d been using through Heroku. They announced that they were shutting down in a month, and we’d need to migrate our data somewhere else. Panic set in pretty quickly. One month felt like such short notice, especially for a live app with customer data. It was hard not to worry about downtime, data loss, or getting stuck with an overly complex setup.
For some time, I’d considered moving off Heroku. While it was reliable, the high costs made it tempting to explore other options. But every time I’d think about it, I’d fall back on the old saying: “If it isn’t broken, don’t fix it.” Now, thanks to ClearDB, it was broken.
Exploring the Alternatives: Stackhero, AWS, and DigitalOcean
Once the need for a new database solution was clear, I started looking at other options.
Stackhero: My first thought was to find another database add-on for Heroku, and Stackhero seemed like a solid option. It was easy to set up and would let us keep the rest of our Heroku stack intact. However, to use it securely, we’d need to add a static IP add-on for secure database access, and that bumped up the cost. Stackhero’s customer support was excellent, though, and they were quick to answer my questions. Ultimately, though, the added costs made me hesitate. The whole point was to reduce our hosting expenses, so Stackhero ended up off the table.
AWS: Next up was the all and mighty AWS, a powerful and widely used option. However, as someone without a lot of DevOps experience, AWS felt overwhelming at times. I managed to spin up a database, configure it, and connect it to our app, but within just two days, I noticed charges of about $10 for minimal data usage. I realized AWS might be more than we needed at this stage, but it’s something I’ll keep in mind for future scaling.
DigitalOcean: Then came a recommendation from a friend — DigitalOcean. Signing up felt like a turning point. Setting up a database and app server was seamless, and DigitalOcean’s straightforward pricing model made it easy to estimate monthly costs. For a self-funded startup, knowing what you’ll be spending each month is invaluable, and I kept waiting to find a catch. There wasn’t one — a week in, and it’s still smooth sailing.
Lessons Learned from the Transition
Looking back, I’m almost grateful for ClearDB’s shutdown notice. It pushed me to make a move that I’d been putting off for too long. In the process, I picked up a few important lessons:
Prioritize Security: During this migration, I learned the importance of securing database access, only allowing trusted sources. This might sound basic, but it’s something I overlooked initially. Thankfully, Cloudflare had already been providing some security, which probably saved me from potential issues.
Budget-Friendly Choices Matter: As a self-funded startup, keeping costs under control is crucial. Heroku’s ease of use was tempting, but DigitalOcean’s pricing transparency and functionality were ultimately a better fit for our needs.
Be Prepared to Adapt: Even when your setup seems to be working well, don’t assume things won’t change. Having a migration plan (or at least an idea of other options) can save you a lot of stress.
In the end, I’m glad ClearDB’s shutdown forced me to evaluate our options and make a move that was long overdue.
By the way, this isn’t a sponsored post. I didn’t even use a referral code to get any credits :( — though if anyone from DigitalOcean sees this, I wouldn’t say no to some credits. Cheers to learning, adapting, and finding the right tools as we grow.
Here's a code for you if you decide to go with DO : https://m.do.co/c/110d41caa6cc.
It gets you $200 and will give me $25.
Yes, this is another 'I left my six figure job to pursue my startup' :) And YES, I am excited to be the one writing such a post. I always knew that I would have something similar to tell, but I didn't realize how hard it would be to go through the process.
It officially happened on May 10, but the gears started turning last September when my company called us back into the office 3x per week. I've realized that I am the type of person to complete the tasks I have committed to within the timeframe that is assigned. I do not believe in the '8 hours of desk work' mentality, and it was something I brought up to my manager during my interview. It worked out well because he was on the same page. It was the most honest I've been during an interview process and I was in a place where I felt like, 'if I'm not honest in the things I want now, it will just build up resentment and I will hate my job in 6 months'.
So pro tip, be brutally honest to yourself and the company early on, and you will put yourself in a position that you enjoy being in for who knows how long.
Back to the original point, once the recall happened, I knew I did not want to be commuting 45 min in traffic each way, and felt like twelve70 deserved attention that has been lacking since Mayan and I took full time jobs.
We spoke about it and created a timeline as to when would make sense to jump full on into the app.
It was a relaxed few months leading up to May. Mayan had left his job a bit earlier, while I was waiting on my bonus before submitting my resignation letter. I had spoken to my company about my plans in February so that they could plan for my departure. I developed a lot of great relationships there, and it was the first time that I was approached with questions like 'is there anything we can do to accommodate you to change your mind?'. It was really touching and had me thinking a lot about the decision I was making. It felt good to be wanted; I've never had a job that I didn't want to leave.
The people were amazing, the table tennis even better.
Anyways this is getting kind of long, I want to post here more often with update on our journey. We are currently together, developing this thing out of a co-working space in NYC for the summer.
One week in and its been really good. I look forward to waking up in the mornings to meetup with Mayan and get cracking. This is a feeling that I knew exists and just taking it all in while it's here.
Thanks for reading!
-
1
I forgot to mention, the week leading up to sending in my letter of resignation was full of anxiousness and fear. So was the week after. I was severely questioning if I was doing the right thing. But a part of me reminded me of why I was doing this...I never wanted to look back and think 'What if?'.
It does get better after the transition period. Now I wake up before my alarm ready to go, with only minimal amounts of anxiety , and that's ok :)
Oufff...I just realized the last time I posted this stat was in 2020 and figured it is time for an update. It's nice to see that something is working. This is across 25000 users so its a fairly decent average. We haven't yet purged inactive users, so we'll see how those numbers are affected once that happens.
I think changing the onboarding flow really helped with that. We are asking our users to add a minimum # of items to their closet as one step of the process. I feel like its the least someone can do to get the outfit calculator working and it allows us to deliver our value in an efficient manner.
Jumping into a full time job really diverts alot of attention from what wakes you up in the morning. It's been a bit over 2 years since I took on a 'desk' job to take care of my needs.
Twelve70 took a back seat for a while. It was there, but I wasn't feeling the fire. I guess knowing that a check is coming in, really puts out the fire that help get so much development done. It's been a comfortable two years, got to travel a fair amount while working remotely and move around way more than I did prior to that. Its been...nice.
I guess I'm coming back to this platform to get some accountability back in my life. Twelve70 is still a work in progress, like any other software project or business, there is no 'done'.
Well, somewhere along the past 6 months, the fire has been coming back and we are going to be launch v2 of the app soon. The plan is to keep the landing page on the same url, but the app will live on app.twelve70.com. Hoping this doesn't hurt our traffic too much, but we'll see. I've never done this before. Cheers to new things.
-
1
Nice concept! And I really like the clean design of the site. I would love to learn more about the story behind Twelve70 if you guys would be willing to chat/share!
I'm currently working on my own fashion-based startup called RateYourStyle (it's also just a side project for me and so I understand how difficult it is to find the time and energy after work!). Anyway, kudos to building an awesome app and excited to follow along on your journey!-
1
Thanks! Yes definitely. Are there DM's on this site?
Your project is great too! Are you building it from scratch? I like the idea of building your closet as you go. Thats one area where we are looking to improve. Not many people want to sit around and log their whole closet.
-
1
yes! I built the backend using golang and frontend using nextjs. and not sure there are DM's on here... I'll shoot you an email at the email listed on the twelve70 site if that works!
-
-
...crickets...
This was a hard pill to swallow, but it sort of fell into my lap, and the timing felt right. We wanted to keep twelve70 going as a self-funded project and this was the way to do it. This might've been naive, but its what our gut told me, so I went with it.
It felt right in the moment, and that's all I could go with.
Haven't looked at this in a while. It's still kind of wild that strangers are managing their closets and how they dress with something that we built. It's kind of a strange feeling, but gotta pause for a moment and enjoy it.
I remember when we released the closet and we were in the 100's and I was excited.
Seems like a huge number at the moment, but can't wait to see what it's like a year from now.
This is the first time we put the app out in the world. It was sort of an impulse decision on my part, but we just needed to hear what people thought of what we have been working on under covers for so long. Little did we know that it would make it to the #4 post of all time with 15.4k upvotes. It definitely pushed up our roadmap alot as we figured that there something solid here, and we need to explore the opportunity more.
I was fresh out of school, and working at my first job, living in the city. Twelve70 was on the back of our minds. My brother was grinding it out in NYC in the fashion industry, while I was in Toronto, working, learning. This pretty much was the case for the next two years.
It took a few weeks, but I'm finally done building the minimum viable product for twelve70, known then as wearbl. I coded the whole thing on PHP and mySQL since that's that what I was learning at the time. It was super basic and the design was just awful now that I look back at it. At this point it was just a school project and I used it as a portfolio item for job interviews.
Comment