Introduction

If you operate an e-commerce business and have even a little experience with Google Adwords and PPC marketing, you are probably familiar with the term mobile-first-world – the change in our offline/online behavior where "relying on our phones has become second nature". To no surprise, this shift has influenced digital marketers to focus their efforts more on capturing the attention of mobile device users. However, even though currently the main fuss is about the increasing dominance of smartphones (and rightfully so), I feel that several equally important observations are overshadowed, or at least not given the attention they deserve.

Feeling this way, I would like to examine and share my thoughts not just on mobile device dominance, but more on the changing online behavior of users because of this growing dominance. More specifically, on the increasing mobility, depth and complexity of online purchasing.

To get into this, let's delve deeper and look at some facts that help us understand what's behind the term mobile-first-world. For starters, cumulative searches on mobile devices have surpassed searches on desktops. Smartphones are also the first devices that people turn to in moments of need. These facts shine a light on two clear and extremely important things:

• more people use mobile devices to find information vs. tablets and PCs
• more people start their information search or initiate the shopping process on mobile devices

However, if you want to adapt and maintain a successful PPC program in e-commerce for the foreseeable future, it is not enough to just understand what mobile-first-world means. You must understand the crucial shift in user behavior–the increase in the number of touch-points, during which people engage with your business before making a purchase. Moreover, this rise in mobility also means that there are more engagement circumstances to think about, because until completing a purchase, people will engage with your business from different: days, times of day, physical locations, engagement intents (purposes), browsers, apps and so on. Finally, it's not that smartphones will quickly make other devices obsolete and become the single and only way of finding information online–for now, assessing how much your e-commerce is susceptible to mobile devices and the first two points is still very important.

Therefore, instead of talking and centering ourselves around mobile devices only, I would like to focus more on mobility–user mobility and the quick pace with which they interact with information in today's cross-device, cross-app, cross-browser, cross-X world.

Having all of this in mind, let's look at how online purchasing behavior has changed in the two periods–before the mobile-first-world and after it.  

‍

The Shift Before the mobile-first-world

I think you would agree that in the past years a typical approach to making a purchase online would seem like a specific task, where you dedicate a certain amount of time, usually via a PC, probably in the evening after work, doing some sort of research, choosing the actual product from the selection available and making said purchase. I call this the pre-defined purchase.

‍

After, in the mobile-first-world

Today, however, this process is breaking into parts and shifting towards a more flexible, on-the-go and mobile-friendly one. Therefore, now and in the foreseeable future a typical purchase will likely be along these lines–doing the initial product research on a smartphone in the evening while waiting for a friend in a car, browsing again through the selection and offers at the end of the day at home via a PC, going to bed and getting up thinking which product and offer best suits you, taking the subway to work while rechecking and finalizing your decision throughout the day on various devices, then finally making this purchase on a PC in the evening. I call this the mobile purchase.    

Example

Now let's see what these changes look like in a real-world example. With this example I want to accomplish two main goals:

1. to show you how online purchasing behavior and user-business engagement is changing and
2. to show you how to handle and adapt to this change within AdWords campaign management, specifically–how to apply smart bid management with the help of micro-conversion set up and tracking and cross-device conversion tracking    

Given that in recent years home furnishing along with clothes, cosmetics and electronics have been one of the most popular items sold online, let's say we own a simple furniture store that has a significant and increasing share of revenue from e-commerce. For the sake of simplicity, let's assume that our target CPA is $100.  

‍

1. Define and Understand Your Buyer Personas

Let's create ourselves a visual representation of our business's typical client–what are his or her most probable activities, device preferences and purchase intents during different hours of a typical workday. A good illustration of this would be the following table:

Example of a typical buyer's workday activities, device preferences and purchase intents by different times of the day

This type of person, which we assume represents our typical buyer online, is likely to act like this:

• decides to buy one of our couches
• does an initial research of this item in the morning on his phone
• checks the offers and selection on his way to work via his phone or tablet
• meets his wife for lunch and they together choose the best option
• makes a purchase while getting back from work or in the evening on his PC

The key element here is not the end point (the stand-alone time or device–the PC in the evening) at which the purchase was made, but rather the whole conversion path and the cross-X activity in it.  

‍

2. Define Micro Conversions

Let's now consider the touch-points users can take in their engagement journey with our business that have valuable information for us. Remember that we can decide/define what type of conversions we count in our main 'Conversions' column as the remaining ones will be counted in the 'All Conversions' column.

For the sake of simplicity let's say we know three distinct actions a person could take on our website:

1. browsed our website for a short duration (least important action)–we do not count these as conversions
2. browsed our website for a longer duration (action of medium importance)–a user browsed for more than three minutes and visited more than five different pages on our website–we count these in 'All Conversions' only, because it's an important step towards making a purchase
3. purchase (most important action)–a person actually makes a payment for a product–we count these in 'Conversions' only  

‍

3. Track Cross-Device Activity

Here we need to identify and clearly understand important user activity that spreads across different devices. Consider these primary metrics:

• Cross-Device Conversions–conversions that involved more than one device, but are counted for the device, where the ad click occurred. This will give us an understanding of how often different devices participate at the start (first interaction) of every conversion path. These results can be accessed through the usual AdWords result interface and are included in 'All Conversions'.
• Assisting Devices–the number of last-click conversions and click-assisted conversions assigned to different devices. This will tell us how often every device participated in conversions somewhere during a conversion path–participation can be either click-based or impression-based. These results can be accessed through Tools > Attribution > Cross-Device Activity > Assisting Devices.

‍

4. Apply a Smart Bid Management Strategy

Now we are ready to apply a smart bid management strategy. We look at the actual results–the traffic, penetration, micro and cross-device conversions. We run the reports in AdWords and generate ourselves a comfortable view of the main metrics:

Example of AdWords engagement and conversion results during different hours of day

Combining our typical buyer's portrait with the actual results at hand, we can extract useful information from this and apply smart bid adjustments:  

10:00 AM - 12:00 PM

While having high traffic and CTR levels, we receive very low 'Conversions' and 'All Conversions' amounts here. Therefore, we conclude that people here are mainly from our browsing segment–we exclude them.

12:00 PM - 7:00 PM  

While also having good traffic and CTR levels here, we receive low 'Conversions', but high 'All Conversions' numbers here. Therefore, we conclude that people here are mainly from our medium importance group–we keep them with around baseline bids (despite high CPAs) as we want to stay competitive here and be somewhere in the middle of search results.

07:00 AM - 10:00 AM and 7:00 PM - 1:00 AM  

Seeing highest numbers in 'Conversions' and 'All Conversions' columns and smallest numbers in CPAs–we increase bids, because we want to be in the highest positions at the times when people show their purchase intents the most.

Now let's look at the results across different devices:  

Example of AdWords engagement and conversion results by different devices

Let's go over the results and the applied bid adjustments for every device separately:

• Mobile Devices. First, we see that mobile devices delivered 592 'All conversions' (44% of total), 61 cross-device conversions and 531 medium importance conversions. This data combined with the facts that they were also the starting device on 61 conversions and assisted conversions on other devices 187 times (based on clicks), we want to keep smartphones at baseline bid level for their high participation in overall conversions (despite high CPAs). In other words, we would not have had high desktop results without high mobile device assistance.
• Tablets. Following the same reasoning and identifying that tablets do not participate as much in overall conversions and have CPAs above target–we decrease the bids here by 20%.
• Desktops. Seeing that desktops are the device where most people complete their purchase (of course after receiving a lot of assistance from mobile devices), we want to show our ads here higher and, thus, we raise the bids here by 15%.    

Having this strategy in place, we can feel safe that we have covered the primary manageable bid management aspects in AdWords, which can be scaled and applied to different accounts (businesses), campaigns and ad groups.  

‍

Take-A-Ways

• Identify and create the buyer persona(s) for your e-commerce business
• Define, set up, and track several different conversion actions (micro conversions) that are uniquely valuable to your e-commerce
• Identify and understand the different parts of typical conversion paths–the roles of different times of day, physical locations, devices, and engagement purposes/intents
• Track cross-device activity and conversions
• Apply a smart bidding strategy based on these points

‍

Conclusion

To wrap up, I would like to emphasize that uncovering and having a clear picture of how your user-business online engagement looks in today's mobility and cross-X based world, will provide you with a strong foundation for maintaining highly performing PPC programs.

‍