Previous installments:Day 0: the ground rulesDay 1: idea number one
Todayās going to be on the shorter side as Iām still busy emailing folks. Also, Iād love to do a comprehensive post on my early validation attempts in a few days, after Iāve tried a few more channels.
The start of validation
I have three primary goals with this early stage of validation:
- Find which group of people have the best response rate
- Find which channel of sourcing these folks yields the best response rate
- Of course, find a real interest in my idea
#1 and #2 are to let me find one or two specific channels that work in terms of response rate, and double down in what works.
Yesterday, I mentioned there were 5 different groups of people I might sell this to. Today, Iāve focused my efforts on the two groups I assume will have the best response: UX leadership/seniors and Product Managers.
I hit my goal of reaching out to 10 targeted prospects, and even this small number turned out to be a great exercise in finding my audience. For the first few rounds, Iāll be testing two different cold audience channels:
-
Leveraging Twitter searchĀ (you can search for tweets containing images which is super handy in my case: filter:images āerror messageā)Ā Using this, I found companies who were having a recent influx of customers pinging support with errors Bystander.io could help fix.
-
Blog posts about error messagesĀ This helped in solidifying whoās thinking about these user-facing error messages the mostāāāauthors were exclusively UX folks. I followed this breadcrumb trail to see what business either the authors or interested commenters belonged to.
So for my cold outreach, I had a blog post to discuss or a real example use-case from their customers.
Hereās a pixelated view of my prospecting spreadsheet:
Thereās better ways to track this, but at this early scale a Google doc will do just fine!
Tomorrow, now that I have the beginnings of a method to prospect these two channels, Iāll push out messages to another group of folks. And Iāll message 10 people in a new channel, to begin opening up testing response rates from different sources in parallel to these two.
Scaffolding the actualĀ app.
Yesterday I also mentioned Iād like to get a seed going for the app itself. A this point we want the pipes that can leveraged in any idea we might pivot into, so we scrap as little development time as possible if we kill this first idea.
I started out as a front end dev, and Iām most comfortable with JavaScript. So weāll be going fullstack JS to make sure I can move fast, using technology Iām most familiar with (there is one exception, weāll get to that!)
The (MEAN+extras) stack:
- Angular 5+ on the front end (a frontend that requires Typescript will be a nice help while weāre building)
- Bootstrap: it looks like weāll go with the Nebular theme as a starting point.
- Node on the backend with Express
- MongoDB + mongoose (hosted on MLabs for the free development tier while we hack things together)
- Auth0 for authentication, this will save a ton of time, and their technical content team is amazing.
- Now, for the unfamiliar, something Iāve really wanted to put to use in a project is Redux. I know itās only recommended for larger apps as thereās a lot of boilerplate and a bit of a learning curve. But, this inclusion is to sprinkle in a fun challenge. And the debugging options with a single source of state are awesome. Weāll use NgRx.
Tomorrow, dayĀ 3!
More validation, of course.
I think Iāll pull the trigger on Product Huntās Ship. I donāt like putting up a full landing page until I have static mockups for screenshots, so Iād like to test Ship as a pre-official-landing-page solution.
Iāll set up my authentication service in Angular to wire up Auth0, using some of Kim Maidaās awesome posts to help speed this along.
And Iāll try to fix my woodstove, ācause itās cold in here.
As always if you have any feedback or want to chat my email is on my website or drop a comment hereĀ :-) Thanks for reading!