I'm Ben Gruber

2014 / 16 September

Boarding Past Transparent Launch Week 1

Week 1: Landing Page Launch, Initial Data, & Next Week Plans

I have always wanted to chronicle one of my start ups from start to success (or failure). So now I am doing it in complete transparency. The goal is to hold myself accountable, force myself to document data that might yield insights, and for others to learn from my mistakes and successes. The following is week 1 of that process.


The Idea

I am launching a new website called Boarding Past (www.boardingpast.com). With a simple email authentication the service finds all of your past travel receipts, catalogues the data and creates an interactive map of your entire travel history. It then automatically updates your map with your future travel as you receive new travel confirmation emails.


The Plan

It’s a simple concept, but tough to execute, because it requires writing a ton of code to collect the data from the hundreds of airlines, travel services (the Orbitz and Kayaks of the world), and hotel email receipts dating back 10+ years. So as I build my initial prototype, the plan is to launch a landing page campaign to test the concept, marketing channels, and important copy/messaging.


The Landing Page

BoardingPast.com September 9-15

After taking an afternoon to design a basic one-page site [inspiration from land-book.com], I built the basic layout. The site needed to be simple, completely understandable, and have a clear call to action. I also wanted to add some potential ‘virality’ to the landing page, so I decided to use a third party system to try and stoke the flame. I decided on Untorch, which prompts users who sign up to invite 5 of their friends in order to jump the beta line and get early access to the service. The design and implementation took me about 5 hours and I had my landing page up – awesome. Now to drive traffic.


Driving Traffic

I am operating on an extremely tight budget, especially while my site is in beta and just collecting emails, so I will need to use very low cost CPC (cost per click) adversing and social media to get people to my landing page. For week one, my plan was to run Faceboook ads and try and get my Twitter presence started. Hopefully my efforts on Twitter will drive traffic, but if not, it will at least allow me to try Twitter Ads, which require you to have an active Twitter presence before advertising.


The Results

Top-line Analytics: 99 user sessions to the site from 90 unique users (almost all were driven from Facebook ads). Average time on site is 0:52 seconds which is weird cause it’s a one page site (I guess thats good). I had 7 Sign ups with 0 of them coming from unTorch referrals. This gives a solid but not outstanding conversion rate of 7.7%.

Boarding Past Twitter

Twitter: I followed 12 people, got 3 followers, favorited 17 tweets, wrote 17 tweets, 1 of which was about Boarding Past. I wrote directly to 3 users about their tweets and spent roughly 5 hours on the site total. This resulted in 0 BoardingPast page views at this point.

Boarding Past Facebook Ad

Facebook: My ads targeted users age 22-60 that were “interested” in travel. I spent a total of $54.34 and reached 16,285 people (18,583 impressions) with the ads. That yielded 136 “Website Actions” (124 ‘clicks’ – 117 of those were unique). This gave me an average cost per unique click of $0.46.

This translated to 90 actual clicks to the website and 1 page like. I cannot figure out exactly where the rest of the 33 clicks were – presumably comments and post likes. I also could not figure what Facebook deemed a “website action” since 12 of them did not involve clicking anywhere?

Notable Takeaways: The majority of my Facebook ads were served mobility (on a phone) leading to 119 mobile clicks of the total 124 (better check my media queries). This was backed up by the Google analytics which showed 76 of the 99 website sessions were mobile (13 tablet and 10 desktop).


Changes and Next Week Plans

1. Split Facebook Ads into Desktop & Mobile Ad Sets

2. Create a More Mobile Friendly Landing Page

3. Make Copy More Clear and Obvious

4. Install Proper Tracking to Analyze Channels Separately

The major takeaway for me was that the vast majority of my traffic was coming to the site through a phone, so I wanted to take that head on. Should I limit my traffic to only desktop users since people are much more likely to enter an email with a keyboard in front of them or should I try to improve my mobile landing page? I decided to do a little of both.

First, I separated my Facebook ads into two different ad sets. One that targeted mobile only, and one that targeted desktop only. This allows me to split the budget in half equally, so that the mobile clicks did not eat up my entire budget this time. It also allows me to track the mobile and desktop ads separately to see which converts to signups better (I unfortunately did not have this in place on this first weeks test).

Next, I examined my landing page on a phone, which I had neglected to check in my haste to launch. What I found maybe the reason I did not have better conversion rates. I had forgot to add one line of code (the viewpoint for developers out there – which meant resizing the browser did not reflect the actual mobile view – oops), which basically meant that users saw the site as if it was a regular website (images below). Tiny text, tiny email field, tiny buttons. Users probably had to zoom in just to see what was going on. Not a good experience. So I made the necessary changes – bigger text, bigger email signup, bigger button, centered and above the fold (images below). Here a good piece I found to help me make the right mobile landing page changes.





Lastly, I changed the copy to be less cool, creative, and mysterious to more direct, obvious and clear. I noted that ads with more transparent copy performed better on Facebook, so I decided to match the ad and the landing page for good continuity.


Well, one week is in the books. Let’s hope for an extremely productive week 2!

No comments so far.