Prismatext

At the end of last month, an old friend of mine joined me to ramp up Prismatext's marketing campaign. I admittedly have little experience with running ads, and to have someone so capable at it this early on was a huge boon.

Since ramping up, we've seen a decent uptick in the number of people buying books, and yesterday we crossed the 100 customer threshold. Most are first-time buyers, but we've even had a few returning customers already! It's encouraging to see even this early on we're finding out who our hardcores are.

Having the extra hand on board also means I'm also able to spend more time refining the actual products and develop new features.

For fans of Arrested Development, today is Cinco de Cuatro. For many more, it's a chance to say "May the Fourth be with you," whether or not they actually saw (or liked!) Star Wars.

But today is special for another reason, at least to me: it's when I chose to start telling random internet strangers—outside of Indie Hackers—about Prismatext. To do this, I started a week-long Facebook ad campaign, and am in the middle of posting to a handful of relevant subreddits.

Less than an hour ago, I experienced my first sale "from the outside," meaning the customer had no previous connection to Prismatext in any way. Whoever this wonderful person is saw the ad or post, clicked over to my little bookstore, thought to themselves, "cool," and straight up bought a book. Just like that.

Does it always feel this good? :)

Anyone who has ever watched Bob's Burgers will know that the opening credits features the aforementioned restaurant suffering setback after setback at each Grand Re-(Re-Re-)Opening. Looking back on the last month, I think I've experienced this exact sequence of events when getting Prismatext off the ground.

Thankfully, I only ever "launched" to small groups of people. First family and close friends, then the larger (but still limited) Facebook community, plus a few posts here and elsewhere to like-minded communities. But based on early feedback from these smaller audiences, I decided to overhaul Prismatext a third time to make it look more like a proper online bookstore. Hopefully the adage rings true, and that the third time really is the charm.

Just shy of one week after my "Pre-Pre-Launch," I made a Facebook post on the Prismatext page and shared it from my personal account to share with a larger circle of friends and family. I still hadn't done any sort of "Open for Business" post to the world, as I wanted to tinker and refine based on more feedback. Thankfully, the headaches and troubles have been pretty minimal so far, so I'm planning on more public-facing posts starting early next week.

In the game of Entrepreneur Bingo, I just marked the "Mom is your first customer" square with a big fat marker.

I sent out a small salvo of text messages to friends and family to let them know I'm starting to roll out the Prismatext bookshelf. Before I could circle back around to my parents and give them their coupon codes, my mom and dad had both already purchased their books. 😅

I've read so many accounts of entrepreneurs musing on the early days of their business. Regardless of how much success they had going forward, those that mentioned supportive parents always seemed the happiest. I'm so thankful that I can count myself among them.

On April 13, I opened the bookstore to close friends and family. I didn't hide the site from the public, but I didn't make any posts or announcements about it being available, either—just a handful of messages to a variety of friends who have known about the project.

I wanted to do a slow roll-out of the books and content, as there was still a lot of work to be done. I'm glad I did, too, as there were changes made that week that helped visitors orient more quickly.

Things are getting smoother and smoother—my commit messages are becoming narrower and more focused as a result of having "less to do," except blend books. The actual work done by the computer is done in seconds, and the manual review process takes almost exactly what I thought it might.

I've started sharing the product with close friends and family. For the first time, I'm seeing Prismatext on other e-readers.

ACTUAL TIME PER BOOK: 30-45 minutes (depending on length)

I was finally able hit the benchmark I set for myself over 18 months prior: blend an entire book in under an hour. In the three months since the last milestone, I dove into a number of online courses. In many instances, I took the course twice: once for the curriculum, and again to abstract what I learned into something else.

ESTIMATED TIME PER BOOK: 1 hour

Today isn't so much a celebration of progress as it is a sort of "base camp" for the next scale. After a number of decisions (both technical and linguistic), I realized I'd worked myself into a corner: if I had stuck with what I currently had, I would never be able to produce the books at a meaningful scale.

October 3 is when I decided to scrap several weeks of work and start over—this time in a language I still needed to learn.

ESTIMATED TIME PER BOOK: 1 1/2 days

I'd learned a while ago that this just wouldn't work if it was only ever run from the command line. No matter how optimized the logic was, the product was always going to need a manual review component (by early estimates, about 30 minutes per book).

Today is when the revised front end came to life, which anticipated the solution of the myriad technical hurdles. Processing a book still takes way too long for this to be scalable, but it's getting faster.

ESTIMATED TIME PER BOOK: 3 days