Friday, October 21, 2011

7 tips for turning Facebook likes into buys - Part 2

This is part 2 of a 2 part post outlining 7 tips for converting Facebook likes into buys. You can read part 1 here.

4. Measure which posts engage customers best

We experiment with a range of different posts on our Shoes of Prey Facebook page. 2 of the posts that we find work best are:

1. A simple post briefly discussing a fashion trend, for example, "Loving vintage inspired trans-seasonal Mary Janes..." then displaying two shoes we've made that match that trend and asking which shoes people prefer, the ones on the left or right.

2. Whenever Jodie is attending an event she'll choose an outfit and Susie will photograph her wearing 3 different pairs of shoes to go with that outfit. We'll then ask our Facebook page which shoes she should wear to the event.

Both of these posts are simple to put together, get great engagement and most importantly lead to sales of the shoes in the photos.

5. Use targeting options to your advantage

We recently held a friends and family sale at our offices. Because this was an offline sale held online at our Surry Hills offices, we wanted to only target our advertising to customers who lived locally. We created a Facebook ad targeting people who like the Shoes of Prey page and live in Sydney. This was only 2,060 people however the ad got an excellent click through rate and we had about 200 clicks for $70 of spend.

At the sale we asked customers who made a purchase how they'd heard about the sale and we recorded this information. Nearly $10,000 or revenue could be attributed to customers who heard about the sale via these Facebook ads, a fantastic return on our investment.

6. Time of day to post

This one is getting quite detailed. We created a custom report in Google Analytics that tracked sales referred by Facebook by the time of day those sales occurred. We then mapped that data to the time of day we were posting to Facebook. What we found was the times that converted into sales best for us were 3-4pm and 6-7pm. That makes sense, we're a fun shopping experience so customers are shopping with us in their afternoons at work or when they get home in the evenings.

The times of day might be different for your business, so creating a custom report like we did to measure this for your own business could be worthwhile.

7. Multi-Channel Funnels

Google Analytics recently released a feature called Multi-Channel Funnels. Say a customer first visits your website after clicking a link on Facebook. A few days later they search for 'design your own shoes' on Google and visit the site again. A week later they type directly into their browser and make a purchase. Most analytics software would track this sale to the last visit, in this case a direct visit and no attribution would be made to the Facebook or search visits even though these contributed to the sale.

Multi-Channel Funnels changes this and allows you to see which sites contributed to the sale as your customers moved through your various sales funnels. It's very useful for seeing which sites, like Facebook, introduce customers to your brand who later go on to convert.

Using this report we've found that Facebook contributes to 8% of our sales while driving 4% of the traffic to our site, so it's a marketing channel that's well worth us continuing to invest in.

Are there any other tips you have for converting Facebook likes into buys?

Slides from my presentation below:


  1. Awesome post - thanks Michael. I am especially interested in the Multi-channel funnels as this has always been on my wish list of things to measure.

    Time of day is great too and I am assuming this is for Australia only? Did you see a change for international customers? This is the one we are grappling with as our customers are primarily in the US and Germany.

  2. Glad you liked the post Ruth. We did the time of day report for Australian customers only. You could run the same report to only show US customers then a separate report only for German customers. Time zones will be reported in Australian time if that's the time zone setting on your Google Analytics account but you could just manually adjust the time zones manually yourself.