GitLab Inbound Marketing, 6 Nov 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: about.gitlab.com global nav test summary

Description

Dashboard — https://datastudio.google.com/u/0/reporting/1c0992b7-9073-41b1-a18b-8d18408d0d89/page/FCsh

A

Hey there shane rice here with the inbound marketing team. uh Looking at a global nav test, we've been running for the last two weeks, a little over two weeks on about.getlab.com. um So when you look at the nav, you have a get free, trial, uh cta and then uh search option.

A

Then next to that you have register sign in and so what we want to do is we want to consolidate those ctas a little bit. So we move register on the home page down below the fold and then we move the get free trial cta all the way to the right. So now, when you click on sign in it says actually says on the nav you'll have search explore, sign in and then the get free trial cta on the far right. So it's kind of standalone and gets a lot more attention.

A

That way was the thought. So um what you're looking at here is a um dating studio, dashboard that I put together to look at the clicks on the control uh versus click on the test version, and so um you know uh what we see here is. You know out of almost 1.4 million uh page views, uh and this is across about.gitlab.com for this time period uh there were 3781 clicks on the cta in the control and there were 12 440 clicks on the test.

A

So you know when you look at this, it clearly shows that the test uh is doing a better job of getting attention on the free trial from the navigation and so from a site-wide uh consideration. It makes a lot of sense for us to consolidate the navigation and add in that cta on the far right as a standalone call out. um So what does this mean, though? For you know the actual.

A

Registrations, gitlab.com registrations, uh trials for gitlab.com enterprise trials, etc. So what you're looking at now is a spreadsheet that I put together with some data from sisense and so looking at the period from you know, october 20th, through november 4th same period uh september, 29th october 14th that controls for days of the week. So we're not like you know, seeing a spike because we're not on the weekend or something like that, so um gitlab.com registrations were actually up uh about five percent.

A

uh Getting trials for gitlab.com were up around 10 during the same time and then the enterprise trial. Interestingly enough was only up about half a percentage point, um and so what I, what I kind of am thinking might be happening here is so when you go to the install page.

A

um You know we have a um a cta on page there that you know sends people directly to the self-managed trial, and so with this uh test running, what I'm wondering is, if maybe you know we're cannibalizing some of those clicks and it's just going to the free trial, and then people are choosing. You know whatever option they're presented with there as far as more likely people are going to pick the the the com trial at a higher rate, just because it's on the left and we've done testing around that in the past.

A

I can pull up some videos if you're interested in that, but um it's fine, the way to we tested it and the way it is is the way it should be. When we, when we present two options, I mean um definitely some things. We have uh ideas for tests going forward there.

A

um So my recommendation at this point is looking at this data, seeing that we have such a high um conversion rate whenever we our click-through rate, when we, you know, focus on that cta in the navigation um that we move forward with this test, actually move forward with this in production at 100, and then we keep an eye on this over the next two weeks, just to see if we see these same trends where the conversion rate stays as high uh gitlab.com registrationsgitlab.com trials, enterprise trials see what those numbers do when it's 100 to make sure this isn't just uh you know uh some.

A

You know some reason why this was creating a unusual result and then once we've got that data with 100 with this in production, um then we can move on to additional tests from here. um You have any questions. Let me know and uh look forward to talking to you thanks.