<p>My side projects always failBuilding stuff has always been the fun part for me.It’s why my side projects generally follow this pattern:Have an ideaBuild somethingTry (unsuccessfully) to get usersGive up and move onHow I’ll acquire users was always an afterthought &#8211; something I did after there was a product for them to use.This time, I’m trying a different approach.If a product ships in the woods, and no users are around to hear it, does it make money? 🤔💸 From idea to revenueI’m a solo, bootstrapped founder building a SaaS startup in public.In Step One, I chose a problem to work on.In Step Two, I chose my niche and my target users.In Step Three, I validated the problem for my target users.Last week, I created a distribution strategy for HelloHailey &#8211; before I wrote any code.I’m sharing all my product decisions, metrics, successes, and failures in public.Want to get next Monday’s update?Get my real-time case study📣 Distribution before productDistribution is how a company gets their product in front of potential users.You might think I’m crazy for thinking about how I’m going to sell a thing that doesn’t even exist yet. But when it comes to launching a software product, distribution is at least half the battle.If a product ships in the woods, and no users are around to hear it, does it make money? 🤔 🚀 Silicon Valley’s biggest success stories are living proof✉️ Hotmail (now Outlook)Hotmail launched in 1996 and sold to Microsoft in 1998 for $400 million. They achieved viral growth by piggybacking on their users’ outgoing emails.By appending “Get your free email at Hotmail” to every outgoing message, they quickly grew from 0 to 12 million users.🥳 EventbriteEvents were a natural distribution channel. Attendees became event organizers, who invited more attendees, and so on.☁️ DropboxA viral referral program catapulted Dropbox from 100k users in 2008 to 33.9M in 2010.📝 Crafting my distribution strategyFirst, let’s revisit the problem I’m solving:📺 Anatomy of the perfect channelAs I described in Step Two when I chose my niche, I want channels that are free with low barriers to entry.Free &#8211; I don’t plan on paying to acquire users for a long time. “Cheap” isn’t good enough.Low barriers to entry &#8211; A long-term growth strategy that builds barriers is a wonderful thing. But I’m still in the validation phase and need users now.My goal is to get in front of product and engineer leaders in tech.Here’s where I think I can reach them:Outbound salesWith an estimated Annual Contract Value (ACV) of around $1k, outbound sales won’t be profitable in the long-term.For now, though, it’s a great way to hustle my way into the first 10, 20, or 30 customers.The MVP Sprint mailing listIf you’re a product or engineering manager, this isn’t an article. It’s a distribution channel. 🙂With almost 250 subscribers in its first three weeks giving me feedback and encouragement, it’s made the pre-product process fun for me.Redditr/ProductManagement has ~40k readers. It’s an engaged community with lots of smart, motivated product people.My early successes with posts for Step Two and Step Three show that the community appreciates my content.TwitterTwitter is full of interesting and influential people sharing thoughts and having public conversations. And they’re all accessible &#8211; just 280 characters away.LinkedInLinkedIn is a content-deficient platform &#8211; there’s more demand for quality content than supply.And pretty much all product and engineering leaders spend time on the platform.Feel free to connect and shoot me a message! 🙂SEOIn the early days, I can hustle my way to 20-30% month-over-month growth.But when 20% growth requires adding 20+ new companies, I need to be able to rely on more scalable channels.I’m optimistic that SEO will be my best sustainable growth driver in the long-term.The productEventbrite’s product was inherently viral. A product where employees get value by inviting their coworkers has the same potential for (intra-company) virality.🚗 Every channel needs a strategyA channel without a strategy is like a car without gas &#8211; it won’t take you anywhere.I have two initial growth hypotheses:1. Product managers will engage with “meta” content about HelloHailey metrics and strategyIf you’ve made it this far, and you’re a product manager, then you’re supporting my hypothesis.I’ll share all my product decisions, metrics, successes, and failures in public. This product meta content will reel PM’s into my funnel. Some of them will convert into users.The HelloHailey growth funnel2. HelloHailey will expand within an organization through product-led growth (PLG)After acquiring the first user inside a new company, the product itself will drive growth within that company.Products in my category have strong network effects &#8211; users get more value when more of their coworkers join.I can take advantage of this with a pricing structure that increases with user count, usage, or some similar value metric.How can I improve my distribution strategy?I like my strategy, but I know there are holes in it. Have any tips for me? Let me know in my Twitter thread:Or just show show me some love with a like or retweet. 😁A little engagement goes a long way with the Twitter gods. 🙏🤔 Reducing uncertainty one week at a timeSo now I have a good idea how I’ll reach users. But how will I get them to pay me money?This week, I’ll learn about the purchasing process &#8211; how companies purchase low-ticket SaaS products like HelloHailey.Here’s some of the questions I’ll try to answer:Who can approve purchases?What does the approval process look like?From what budget(s) will companies pay for HelloHailey?Want to get notified when it’s published? I’ll tell you what I learn next Monday, whether I succeed or fail.Send me next week&#8217;s updateI’ll be documenting my startup journey from idea to paying users over the coming weeks and months. I’d love to have you along for the ride.Icons made by Freepik, Icongeek26, and Pixel perfect from FlaticonLet&#8217;s block ads! (Why?)</p>

Breakdown

My side projects always fail

Building stuff has always been the fun part for me.

It’s why my side projects generally follow this pattern:

Have an idea

Build something

Try (unsuccessfully) to get users

Give up and move on

How I’ll acquire users was always an afterthought