6 Min. Read 382Reads

2017 Holiday Season Prep: Data Learnings from Q4 2016

Shani Rosenfelder Oct 19, 2017

Are you ready for the peak season? You better be as US retail e-commerce sales are expected to jump 16.6% during the 2017 holiday season, according to eMarketer. Most of this growth will come from, not surprisingly, mobile, and particularly smartphone commerce which will jump almost 60%. So, yes, big money is at stake and it is therefore absolutely vital to get ready. Starting right now, if you haven’t already.

To help you prepare, we’ve pulled some data benchmarks from the 2016 holiday season. There’s no better way to plan for this year’s season than based on last year’s data. 

Overall, we’ve looked at 95 million installs of 130 shopping apps that measured 22 million in-app purchases generating $450 million.

We’ve got insights on:

  • User acquisition
  • Shopping activity
  • Average revenue per purchase
  • Share of buying users

Let’s dive in.


1) Non-Organic App Installs & In-App Purchases

What does the Android graph tell us?

  • The peak week of the Q4 holiday season was at the end of November. During the weeks of Black Friday and Cyber Monday, we found a 46% higher rate of shopping activity than October (based on a count of purchase events), and 19% higher than December (to find out which day – Black Friday or Cyber Monday reigned supreme – see the day breakdown ahead).
  • Marketers started to scale their campaigns for Halloween as a lead-up to the peak period and mostly kept up the heat until the week before Christmas. Overall, if we compare pre-Halloween and post-Halloween install activity (until the week before Christmas), the latter outperforms the former by 18%.
  • A post-Christmas install rush is more than evident with many users installing new apps on new phones they just got for the holiday. In fact, the first week of January saw 31% more installs, even in comparison to the strongest two weeks in November (Black Friday and Cyber Monday).
  • New Year’s and holiday season clearance sales lead to a significant pick up in shopping activity, with the first week of January almost as high as the week of Cyber Monday (only 3% lower).

And now for iOS.

What can we say about iOS trends?

  • On iOS, the trends were more pronounced than Android with a steady increase in shopping activity until it peaks during the week of Cyber Monday. Overall, the weeks of Black Friday and Cyber Monday saw 70% more purchases than October and 31% higher than December. 
  • UA activity picks up significantly two weeks before Black Friday, rising 44% compared to the previous weeks in Q4.
  • We can also clearly see a rush of installs after Christmas due to many new phone purchases. The first week of January saw 33% more installs than the pre-Black Friday period.
  • Purchase activity rises after Christmas with New Year’s and holiday season clearance sales up 26% vs. the week of Christmas and 30% vs. the month of October. 
2) Peak Period App Installs & In-App Purchases – A Daily Comparison

A look at daily totals during the end of November rush on Android shows Black Friday had 3% more purchases and 16% more installs than Cyber Monday. Thanksgiving was not a significant day for shopping.

On iOS, we see Cyber Monday leading the pack with 9% more purchases and 12% fewer installs compared to Black Friday. Similarly, Thanksgiving is dedicated to eating Turkey, not to shopping.

3) Organic Installs During Q4

Organic traffic shows less fluctuation than non-organic traffic, especially on Android. Organic shopping app installs peak during the week of Black Friday for iOS and Android: 25% and 16% higher than the Q4 average, respectively

4) Average Revenue Per Purchase

When it comes to buying power, iOS dominates Android with an overall average purchase value of $24.3, which was 41% higher than Android’s $17.3.

The end of November rush exists not only in terms of volume of shopping activity but also actual money being spent. During this period, iOS and Android users spent 58% and 66% more than other weeks in Q4, respectively.

We can also see that the iOS basket size is relatively unchanged throughout the period (excluding the weeks of Black Friday and Cyber Monday), whereas the average spend on Android varies – from low spend early in Q4 (46% lower than November) an increase starting November and a relatively high rate in December before Christmas.

5) Share of Buying Users

We wanted to examine the relationship between the period when users were acquired to their quality in peak period – specifically whether they became buying users or not. So we looked at five cohorts of users: the first half of July, the first half of November, peak period – the 2nd half of November, Black Friday & Cyber Monday, and first half of January.

July’s share of buying users is presented as a neutral period, to compare to November and January peaks.

We can see that users who installed just before Black Friday and Cyber Monday were the most likely to purchase during peak period (91% more than users who installed in July). Those who installed shopping apps on Black Friday and Cyber Monday, were 42% less likely to purchase on that day. Early January installers also showed a high probability of purchase during January.

The mid-November peak had 140% higher rate of buyers than Black Friday and Cyber Monday.

On iOS, users who installed just before Black Friday and Cyber Monday were the most likely to purchase during peak period (77% more than users who installed in July). Those who installed shopping apps on Black Friday and Cyber Monday were 59% less likely to purchase on that day.

As in Android, the peak period had a 147% higher rate of buyers than Black Friday and Cyber Monday.

So what it does it all mean for shopping app marketers running campaigns in the US?

1. End of November shopping frenzy. This period with Black Friday and Cyber Monday saw the highest volume and purchases and highest average purchase. Efforts to maximize the potential of this shopping frenzy should include:

  • Getting users early ahead of the peak period. The more data you have the better you can segment. This will pay off when you run re-engagement campaigns during the peak – for retargeting, email, and push notifications.
  • Getting users just before the peak season. Although you won’t have much data for sharper segmentation, our findings show that you’ll get users who are far more likely to actually become buying users come peak period.
  • Campaigns on Black Friday and Cyber Monday. It definitely makes sense to continue pushing UA activity on the days in question, but know that those users are far less likely to buy on these days than the users you’ve acquired earlier in November.

2. Post-Xmas is a real [shopping] holiday. An increase in installs coming from new phones plus New Year’s and clearance sales means you should continue pushing UA and re-engagement after the holiday. Having said that, the average purchase in this period is much lower than the end of November.

3. Constantly monitor cost. Since media cost for both UA and retargeting is likely to mirror high demand during peak periods, remember that if your revenue is high, a higher cost is not such a concern. In any case, you should closely follow media cost and be as agile as possible in your campaigns to ensure you maximize your ROI.

4. Use multiple media sources for optimal flexibility. This leaves far more room for ongoing, data-driven optimization so you pay the lowest possible cost and gain the highest possible return. By maximizing your reach to a larger user base, you also maximize the chances of addressing new, relevant users, which can help in keeping your conversion rates up (make sure your media partners can offer sharp targeting otherwise your conversation rate will suffer).

5. Enjoy the holiday season. Although for you it’s probably best to vacation sometime in January… 😉