Ryan Glass

This is very different from the pre-launch roadmap which was heavily focused on development of the service.

For the next periid we'll be focusing on marketing, SEO and content creation

I've retired the waitlist for new sign-ups - those already on the waitlist still have a few days to get their accounts but I am moving to live sign-ups today.

Early-bird discount will be 70% off for life - not quite as much as the waitlist discount. When I get ten early-bird sign-ups this will go down again.

Our upstream provider replaced a network switch to give 40G networking. It went OK - service was paused for 8 mins which is pretty fast for this changeover but ruins our 100% uptime record for over a year.

One of the issues with running a monitoring site is that uptime is critical. So I use a managed VPS for the main server as they have someone looking after it 24/7...

...but getting them to do anything complicated is difficult. They were meant to pause crons while the network was down but somehow managed to f*** this up and therefore some false positive downtime alerts were sent out to customers.

Not ideal but these things happen and no complaints from customers yet.

I am sending out invites gradually as each new user takes a bit of work to set up and I don't want too long a delay if I sent them all out at once and get a lot of uptake.

Wait time is mostly due to getting the Twitter developer account OK'd by Twitter for each user. This can be just a few hours or occasionally up to a few days and several emails as Twitter is fairly thorough with their checking for new API users. Even though this is a pain I understand why they have to check.

I could cut this out by running everything through one Twitter App under twitMate but there are advantages to each user having their own app. Users aren't associated with each other in any way and each user is totally isolated from all others. In my opinion it's more work but higher quality.

There is a little manual work required to set up each new account: set up a Twitter App to connect with the Twitter API is the first job.

This involved a bit more setup than you might think as the codes for each product changes when you do this in Stripe so a few changes necessary in the DB.

Getting closer!

This was to test customer onboarding and although it wasn't as thorough as I would like (a couple of people I had lined up to test dropped out) overall it went OK and I found and fixed a small bug.

Next up - sending invites to the waitlist members. I know there are a few from here on the waitlist so not long now till you get your invite!

Still a couple of places on the waitlist if anyone likes the look of it 🚀

and we're on to public beta testing!

I made some progress today, working on the last onboarding section.

Here new users pass us details for posting their tweets and URLs of their landing pages (these will be linked from tweets to drive traffic to their website/blog posts etc)

Nearly done - just image uploads to go!

Today I had some productive time where I built the 'audience management' section of customer onboarding.

This is a series of short webforms where we get the information needed to target the correct audience for the customer.

I used 5 short webforms with simple choices (1 click) or a single text input field and 'next' / 'back' navigation. I prefer this to one more complex form as it walks th euser through the information.