Fighting bounce with a process, not a site<\/h2>\n
![\"Device](\"https:\/\/static.gosquared.com\/images\/liquidicity\/14_12_04_graph_traffic_sales_01.png\")
<\/p>\n
For the first time, visits from smartphones and tablets surpassed desktop visits altogether, 52.1% of visits (to ecommerce stores?) came from mobile devices on Black Friday. However, this trend does not translate to conversion rates. Mobile only accounts for 27.8% of all sales. The split between smartphones and tablets is skewed towards tablets which account for 42% of sales on mobile. Desktop reigns supreme with a 72% share of all sales.<\/p>\n
This situation may not be entirely unfamiliar: converting visits into users is one of the fundamental demands of marketing. What does growing mobile browsing tell us about the nature of buying things on a mobile device?<\/p>\n
The gradual adoption of responsive or dedicated-mobile sites is definitely one factor.<\/p>\n
The challenge now is no longer the implementation of a mobile site<\/em> but the implementation of a mobile process<\/em>. A beautiful site – that\u2019s step one. But if I as a consumer struggle to get the numbers from my bank card right, I’m much more likely to stop trying.<\/p>\n Keep pages light.<\/b> Responsive web design can be particularly vulnerable here. Poorly optimised responsive sites can look good but use heavy resources that slow down page loading<\/a> and as a result increase the likelihood of customer bounce.<\/p>\n<\/li>\n Set a purpose and design for it.<\/strong> Chuck the popups, the recommendations, the email subscription offers: the primary objective is to get the customer to buy and nothing else until that is done.<\/p>\n<\/li>\n<\/ul>\n Facebook’s reputation is a mixed bag when it comes to marketers. Whether it\u2019s the risks of brand damage on the news feed, the sheer difficulty of getting content to surface organically or constantly changing policies and algorithms, Facebook remains a battle ground.<\/p>\n The data from Black Friday makes the verdict clear: Facebook still matters to consumers. A lot.<\/p>\n This year, sales from Facebook drove a whopping average of $109.24 per order, double from $52.30 last year<\/a>. Facebook also converts web visits much more effectively than any other social network, converting four times as effectively in 2013<\/a> and twice as much in 2014. This suggests that despite the lower AOV in 2013, in terms of total order value Facebook is ideal for both the quantity and quality of visits.<\/p>\n The challenge for marketers goes beyond building a community – it’s in getting a community to act when needed. You need to know what gets customers excited, what they like to buy, what their hobbies and interests are. You need deep insights about your audience.<\/p>\n With this level of intense detail, you can develop targeted ads that will resonate with your target audience and in turn take full advantage of Facebook\u2019s converting power.<\/p>\n Facebook converts, simple as that.<\/strong> With a strong community built around a brand, the chances for Facebook to cause harm are much more limited.<\/p>\n Perhaps the most interesting observation from IBM\u2019s data lies in an unassuming paragraph: <\/p>\n Less Frequent, More Targeted Email Promos: Retailers sent an average of 5.3 emails on Black Friday 2014, decreasing more than 11 percent over the same period in 2013<\/strong>, as retail marketers continue to send more targeted — and less frequent — messages to shoppers. Open and click-through rates \u2013 when someone opens an email and clicks at least one link \u2013 were 12.9 percent and 2.4 percent, on Black Friday and Thanksgiving Day, compared to 15.4 percent and 2.8 percent respectively, last year<\/strong>. <\/p><\/blockquote>\n It\u2019s an undeniably bizarre situation. Everything has been done right – segmented the audience, targeted to their interests, called them by their first name, developed the best, most optimised layout…and then nothing. Carefully crafted email just sits in an inbox until the next \u2018delete all\u2019. Does that mean that targeting really doesn\u2019t mean as much, and less email is not necessarily good?<\/p>\n Before panicking over the invalidation of about a million and one blog posts calling for email segmentation<\/a>, keep in mind that the IBM data does not take into account the performance of specific segments, only an average.<\/p>\n That said, 5.3 million emails is quite a large number to ignore: let’s consider one fundamental problem that extends beyond email, audience immunity.<\/strong><\/p>\n Five or six years ago, seeing a highly personalised email might’ve made someone go ‘woah’<\/em>. Today, when my inbox looks like this I\u2019m likely not going to have the same reaction:<\/p>\n The problem with ‘popular’ tactics is that when everyone does something that’s supposedly special, everyone looks the same in the end. The solution isn’t a pretty or easy one: get creative<\/strong>.<\/p>\n When the next set of recommendations on email marketing appears, consider \u2018how would I be different if everyone did the same thing?\u2019 That\u2019s called innovation.<\/p>\n From GoSquared data<\/a> collected over Black Friday, an interesting trend that was not present in other sources of data: the erratic nature of site visits, especially in the early hours of the day. The graph below presents pageviews per minute over the course of Black Friday from over 50,000 sites that we track.<\/p>\n Note in particular the erratic nature of visits throughout the day: peaks and troughs occur by the minute. We saw top sites seeing surges as much as 10x their usual volume at various points throughout the day; correspondingly, it would be safe to prepare for 10x your expected traffic volume on major sales days.<\/strong><\/p>\n This analysis underlines the point we made last week<\/a> – you need to be prepared for traffic spikes.<\/p>\n Sites crash. Sales are lost. It generates negative sentiment and bad press. Consumers get frustrated very quickly. In the UK, major retailers like Argos and John Lewis suffered outages as they were unable to handle the demand they experienced on Black Friday, while Best Buy is now unfortunately known as the site that went down twice<\/a>.<\/p>\n Amazed to see some of the largest retail sites struggling to anticipate demand today and provision resource.<\/p>\n — GoSquared (@GoSquared) November 28, 2014<\/a><\/p><\/blockquote>\nSo what does this mean in practical terms?<\/h3>\n
\n
![\"RWD](\"https:\/\/static.gosquared.com\/images\/liquidicity\/14_12_04_processing_time_graph_01.png\")
<\/p>\n\n
Facebook matters: get the right kind of clickthroughs<\/h2>\n
![\"FB](\"https:\/\/static.gosquared.com\/images\/liquidicity\/14_12_04_AOV_Facebook_Pinterest_01.png\")
<\/p>\nThe email paradox: more targeted but less opens and clickthroughs<\/h2>\n
![\"inbox\"](\"https:\/\/static.gosquared.com\/images\/liquidicity\/14_12_04_inbox_on_black_friday_01.png\")
<\/p>\nBonus GoSquared insight: prepare for traffic spikes at all cost<\/h2>\n
![\"GoSquared](\"https:\/\/static.gosquared.com\/images\/liquidicity\/14_12_04_pageviews_over_time_01.png\")
<\/p>\n\n