Intro
Hello everyone, I'm Alan. I'm 36 this year. I left a data analytics/engineer role this year to build my own SaaS. Always wanted to build my own SaaS and AI has given most of us the ability to do so.
Why write at all
Most of what I am going to write comes from my personal journey in building a multi-platform social media scheduler first, later transitioning into a lead-capture product. The intent here is so that the next builders can use the playbook filled with afterthoughts that I have for them to have a more holistic view of what it means to be a solo founder.
AI Ascendency
AI made it straightforward to ship something with features comparable to existing tools. Building is always easy, the hard problem isn't the product.
Before the product you need to have the whos, the hows, whys, whats etc - these don't come automatically, it shows up in the vaguest form where you kinda have an idea the direction you're headed though the winds may change your sails at any time.
As a solo founder who left my decent paying job I had a finite runway, and also not a clue how it's going to work out, but I guess that's the beauty and excitement of trying to run your own gig.
In the indie startup world, distribution is probably in hardcore difficulty mode right now given how the playing field is now levelled. Getting consistent visibility and turning engagement into actual conversations was not straightforward. A lot of other builders hit the same wall: the supply of new apps increased quickly; the demand side (operators who need posting and lead capture handled) did not adopt another tool at the same rate.
Building is easy.. Distribution is not...
The Journey
I'm still very much working on sprkly.app and building it daily but I'm going to document the specific points where I could have done better and what was delayed, how important competitor analysis is, what the language gap looks like, why the offer moved from “use the product” to “we handle the outcome,” and what a running distribution loop actually changes.
Similar to a hub and spoke model, the longer pieces will sit in the newsletter. Shorter versions will go out on X and the other platforms. This is the same approach used by Justin Welsh, Dan Koe, and Sahil Bloom. I am applying it to this particular sequence.
Over the next weeks, I will be sharing lessons from my own journey:
-
Hub 1 – Distribution was treated as a later step
Outreach, consistent posting, and visibility work were delayed while product work continued. The piece covers what was postponed, what the pipeline or account looked like in that period, and the concrete result (features that matched competitors but no motion). Failed assumption: distribution can be added once the product feels complete. -
Hub 2 – Supply increased faster than willingness to operate another tool
More people could ship apps. The operators who had the underlying problem (consistent posting and lead capture) did not want an additional product to run. The piece uses the signals I saw (my reaction, outreach replies, community patterns) and the resulting observation that demand sat at the handled outcome, not at another tool. Failed assumption: people with the problem will adopt a new tool to solve it. -
Hub 3 – The offer moved from tool to handled outcome
The pitch shifted from “use this product” to “we publish on schedule and hand you the people who engaged.” The piece covers what already existed in the product that made the offer deliverable, what the customer no longer had to do, and the early difference in response. Failed assumption: the product must be sold as self-serve software. -
Hub 4 – Customer language did not match the offer
My competitors and I were describing overlapping capabilities, but the wording differed from how buyers already talked about the problem. The piece uses a concrete language comparison and what changed in replies or conversion when the description moved closer to existing market language. Failed assumption: a more accurate description will be understood if the product is similar. -
Hub 5 – Outreach produced product signals, not just sales signals
Cold and warm outreach was run to acquire users. Replies contained information about what was actually needed. The piece covers specific reply patterns that indicated a direction change (including the need for a done-for-you layer) and how those differed from simple rejection. Failed assumption: outreach is only a sales channel. -
Hub 6 – Consistent visibility became the compounding layer
Once publishing and engagement capture ran on a fixed cadence, the account stayed present and a lead list became a weekly output. The piece contrasts a week that depends on someone remembering to post and reply versus a week in which the sequence runs, and notes why a running loop is harder to replicate than features. Failed assumption: any distribution activity is equivalent. -
Hub 7 – Shipping was no longer the constraint
AI assistance made a working product with comparable features buildable in a short time. The assumption that followed was that a functional product would begin to attract users. The piece covers what was built, roughly how long the core took, and the point at which feature parity produced no meaningful inbound. Failed assumption: reducing build time reduces time-to-users. -
Hub 8 – Signal quality versus volume
Raw engagement can rise while conversations stay flat. The piece covers the shift from tracking likes/comments to tracking how many engaged people were captured and replied, with one example where a higher-engagement format produced fewer usable leads than a narrower one. Failed assumption: breadth vs depth, breadth is key.
These are all the topics that I'd like to share for now, will update this list as more pieces come online. If you've a question on my journey, can drop me an email at alan@sprkly.app. I read all the emails.