Much has been said about splitting your campaigns per device. The comments tend to be opposites, from:

“Why would you do that? What a waste of time!” (seen on Reddit, where of course all the PPC experts hang out).

to

“Separating mobile from desktop is the most important thing you can do to increase performance.”

I hardly ever see actual advertisers and PPC practitioners explain how they structure their accounts. Having spoken with many different PPC managers, and often seen their work, I’ve come to the conclusion that most of us have similar approaches, which can be summarized in the following way:

  • Only split up your campaigns when it makes financial sense
  • Don’t overcomplicate your account by splitting up campaigns unnecessarily

If you’re pressed for time, then that’s it. It’s the gist of my post. Don’t overcomplicate it and only focus on splitting up your campaigns when it’s necessary.

The rest of the post will go more in-depth with when it’s necessary to split up campaigns and how to split them in practice.

If you want to take the shortcut, download this step-by-step guide for exactly when you should split your campaigns into mobile, and how.

The Case for Splitting Your Campaigns in Individual Mobile, Tablet, and Desktop Campaigns

The benefits of splitting your campaigns up can mainly be summarized in three main benefits:

  • Bid Adjustments: Location, Hour, Day. The three devices behave differently, so you should bid differently for these three areas.
  • Performance Segmentation: Out of sight, out of mind.
  • Change Begets Change: Every time you change a “desktop” bid you essentially need to re-evaluate your bid adjustment per device.

Bid Adjustments Per Location, Hour, and Day of the Week

You can only apply bid adjustments per location, hour of day, and day of week at the ad group and campaign level. You can’t specify a different bid adjustment for mobile based on the hour of day or similar traits.

User behavior on mobile and desktops (and even tablets) differs during various times of the day, as the graphic shows below:

Image Credit to SmartInsights.com

Our mobile use peaks during times when we’re on the road, desktop peaks when we’re at work, and tablet peaks at night. Makes sense.

That means you might want to set a different ad schedule for your mobile devices than for your tablet/desktop devices. For example, you might reduce your mobile spend during mornings and lunch breaks if you’re an ecommerce store and you see a negative ROI at these times.

Performance Segmentation: Out of Sight, Out of Mind

One of the biggest reasons that I like to separate mobile, tablet, and desktop is due to the performance differences I see from time to time.

If the performance on mobile is a lot worse, or better, than the other devices, then I like to create mobile-only campaigns. The main reason is because of the good old saying: Out of sight, out of mind.

Yes, there are strong segmentation options built into the AdWords interface and most third-party reporting tools. BUT it’s not right in my face. A poor performing mobile campaign can easily hide behind a successful desktop campaign.

Yes, I can go into the segmentation, but take a look at the two screenshots below and tell me, which one are you more likely to take action on?

 

 

Bid Modifiers Are Outdated Every Time You Update Your Bids

Have you ever stopped to consider when you change your keyword or ad group bids in AdWords, you actually need to revisit your bid modifiers?

If a keyword converts incredibly well on desktop and you increase its bid, but don’t take into account how the increased bid will affect its mobile performance, you’re essentially setting yourself up for failure.

Increasing a bid for mobile devices can have big traffic implications. Getting your ad to show “above the fold” on mobile can be the difference between a small traffic trickle and a flood of traffic.

Ideally, by having mobile, tablet, and desktop in the same campaign you would change your bid modifiers according to its performance every time you update your bids. I’m sure you don’t do that, though (I certainly don’t), which is why separating by device is so important.

Don’t Expect Mobile and Desktop to Perform The Same

A common misconception I see is that mobile campaigns need to perform at the same level as your desktop campaigns. Ideally, that’d be nice, but it can be a limitation to scaling your campaigns.

Let’s say you’re performing incredibly well on desktop (e.g. 10x ROI) due to years of running campaigns, hundreds of hours spent optimizing, and generally doing a great job staying ahead of your competition.

At the same time as your desktop campaigns are crushing it, your mobile campaigns are “only” getting a 6x ROI. Many advertisers will automatically aim to get the 6x ROI up to the same 10x ROI as they’re used to getting from desktop. Don’t.

You need to treat mobile different from desktop.

You need to go back and revisit what your min. ROI level is for you to break even. If your lower limit for breaking even is 4x, then a 6x ROI is pretty good for mobile. You should, of course, always strive for the highest possible ROI, but don’t go for a higher ROI at the cost of volume. If you’re profitable at 6x ROI, then scale your campaigns at 6x.

Of course, don’t settle if you can improve your campaigns, but don’t let a super optimized desktop approach limit your mobile potential.

The Challenges of Mobile-Only Campaigns

Let’s face it. The biggest reason why running mobile-only campaigns isn’t standard is because it’s time-consuming. Having to create double, or triple if you include tablets, the number of ads, search term mining, ad extensions, bid management, etc. is simply time-consuming.

Creating the campaigns is the easy part. You basically just create one campaign, duplicate it, and make some minor modifications to your ads, ad extensions, and bid adjustments. Easy peasy.

At scale though, device-specific campaigns becomes a lot more time-consuming to manage.

Too Little Data to be Useful in Ad Testing

The second challenge of separating mobile, tablet, and desktop is related to the data shrinkage (as Aaron Levy named it) that happens.

Depending on your industry you’re essentially reducing your traffic per campaign by between 33% and 66% with a device-only approach.

However, the argument goes for whether you should be lumping these traffic sources together in the first place. Having it all together might just be giving you a false sense of statistically significant data, when you’re truly putting three data sources together that should be reviewed separately.

Your Traffic Volume Determines When To Split Campaigns

The reason why you often see conflicting advice in PPC is because PPC managers work with a wide variety of clients.

For example, if you see someone run PPC campaigns for Roto-Rooter, you can be sure they’ll pitch mobile-only campaigns. Mobile for local companies is crucial to their success and a massive lead source.

An ecommerce store that sells iPhone accessories, has 10,000 products, and gets hundreds of thousands of visitors from AdWords will benefit substantially from splitting their biggest campaigns per device. Even just a 10% performance increase will at that level means tens of thousands of dollars every month.

An ecommerce store selling a niche product with a limited amount of traffic will see little benefit from splitting up their campaigns in most circumstances. Getting a 10% performance increase might just mean a couple of extra sales per month. No necessarily worth the hassle of the added complexity.

So, depending on the clients that the author you’re reading is working with, you’ll see varying advice and best practices.

Situations Where You Should Separate Mobile, Tablet, and Desktop

My main advice to ecommerce advertisers with at least a decent traffic volume (~$10,000 per month) is to only split when it makes sense. Below follows my personal blueprint for when to split campaigns:

  • Split if your performance difference is substantial
  • Split your top 20% campaigns
  • Split your “Single Keyword Campaigns”
  • Split your Shopping campaigns

Split If Your Performance Difference Is Substantial

Depending on the product you’re selling and the state of your website, you will have varying conversion rates for mobile devices.

In general, high-value and complex items convert poorly on mobile (beds, grills) and lower-valued products convert quite well (toys, iPhone accessories, gadgets).

With this in mind, if you have a substantial difference in your conversion rates (or ROI) for mobile and desktop, then I highly recommend you to split the majority of your campaigns.

I’ve often taken over ecommerce campaigns where the overall desktop performance was satisfactory, but because of a lagging mobile performance the advertiser was losing money on their AdWords investment.

With that being said, then the opposite is also true.

Avoid splitting your campaigns if you’re seeing the exact same performance on desktop and mobile.

Start by just testing small and see if a mobile-only campaign can help improve your performance. If it does, split out some more of your campaigns, but if you don’t see a performance increase, then just stick with the combined campaigns.

Again, no need to increase complexity if you don’t have to.

Split your top 20% campaigns

A bit contrary to the advice above, you should always separate your top ~20% campaigns. These are the 20% of your campaigns that produce 80% of your revenue. The exact ratio might be 30/70 or 10/90, but whatever it is, make sure you split up your highest-revenue generating campaigns.

I recommend this because it is where you’ll see the biggest return on the added complexity. Increasing the complexity of 20% of your campaigns to get a 10-50% performance increase for mobile traffic is a great decision.

However, increasing your complexity for 80% of your campaigns that only produce 20% of your revenue is not a good deal.

Split Your “Single Keyword Campaigns”

I’m a fan of “single keyword campaigns” in the sense that if I run a major keyword for a client (keyword examples being: beds, toys, iPhone accessories, pizza, pull out couch) then I want to have the data right in front of me. The best way to do that is to create a campaign solely for that keyword in the various match types.

These campaigns often generate a large amount of traffic and conversions. They’ll therefore frequently fall in the top 20% rule above, but if they don’t, I still recommend splitting them up into device-specific campaigns. It’ll give you more control and the complexity increases only slightly because you have ~3 ad groups per campaign.

Note, that a single-keyword campaign doesn’t mean that you literally only have one keyword in the campaign. I’ll normally have several keyword variations split up in different ad groups per match type:

  • [sofa bed]
  • [sofa beds]
  • [bed sofa]
  • “sofa bed”
  • “sofa beds”
  • +sofa +bed
  • +sofa +beds

Split Your Shopping Campaigns

An easy choice, but frequently overlooked, is to split up your Shopping campaigns.

With Shopping campaigns, there is an added layer of complexity that you can’t control in campaigns that combine devices: Products.

In regular search campaigns you have control over your keywords and ads, but with Shopping campaigns, you don’t have the same luxury.

It might be beneficial for you to showcase different products or apply different bidding. As highlighted by Crealytics, then there is a sharp curve when it comes to bidding efficiency:

Image credit: Crealytics – Guide to Bid Management in Google Shopping

Managing this curve becomes a lot trickier if you don’t separate your campaigns per device.

Situations Where You Shouldn’t Split by Device

With that being said, there are definitely situations where you shouldn’t split your campaigns by device. This will help you avoid increasing the complexity of your account with no significant performance improvement.

No Difference in Conversion Rates

If you don’t have any meaningful difference in conversion rates for mobile and desktop, then there might not be much to gain by splitting up your campaigns.

Don’t overcomplicate your account for no reason. It will have an impact on your day-to-day efficiency, which can potentially eliminate any performance gains you’ll get from using device-specific campaigns.

Your Optimization Schedule Doesn’t Include Mobile

Let’s be honest. Not everyone who reads this blog post will act on the tips they read. Some just like the good stories and the awesome GIFs you find in blog posts these days.

If you’re one of those who think all this advice sounds great but you don’t have a good optimization schedule already, then splitting up your campaigns will just add unnecessary complexity to your account structure and destroy any gains you might get from the split campaigns.

How to Work With Mixed Campaigns

The recommendations in this post outlines that most ecommerce stores shouldn’t split every single campaign they have. This means you’ll end up with a certain number of “mixed” campaigns that target both mobile, tablet, and desktop.

To work with mixed campaigns efficiently you luckily just need to take advantage of a few key features.

Use Mobile-Only Ad Extensions

Most ad extensions have the ability to create mobile-only extensions. This goes for callouts, sitelinks, and structured snippets:

The ability to write just for mobile devices can help improve your performance.

Use the IF(Device) for Mobile Ad Text to Make Them Slightly More Specific

Earlier in 2017, Google came out with IF(xyz) ad customizers, which means you can write a specific message for mobile users. Taking advantage of this will enable you to write ads that are more tailored to mobile users on the go without having to create brand new campaigns.

Routinely Segment Your Data So You Know How It’s Performing

This might be the most important piece of advice in this entire post. You need to routinely review your mobile and tablet performance. You can’t just look at your general campaign performance and optimize accordingly. You need to segment your data at every step of the way.

Is a keyword converting, but not hitting your ROI goals? Review device segments.

Is one ad performing better than another? Review device segments.

Do you want to increase bids? Review device segments.

Whatever you do – review device segments routinely and prior to making any significant changes.

Get into a Habit of Reviewing Your Bid Adjustments After Each Major Bid Adjustment

If you’re still managing your bids manually, then you need to build more frequent mobile bid adjustment reviews into your day-to-day optimization schedule.

Personally, I’m still not perfect at this, as it’s rather time-consuming, but for your biggest campaigns/ad groups this is a must.

My Recommendations

To summarize the tips in today’s blog post:

  • Separating mobile, tablet, and desktop is a must for your biggest revenue drivers.
  • Shopping campaigns should always be separated.
  • Don’t separate if you don’t have any real performance differences between devices.
  • Not all campaigns are worth separating – don’t over complicate things.

As you may know, I like to end my blog posts with a single recommendation. Here it is: If you’re not sure whether separating your campaigns per device will work, then try it for a few campaigns and see what happens.

Find your top three revenue-driving campaigns (+Shopping) and spin mobile into its own campaigns. You’ll get to see first hand whether it’s working out, and you’ll get experience without stuffing your account with complexity.

Download the cheat sheet for identifying when you should separate your campaigns by device.

[optin-monster-shortcode id=”touelqbxsojg8ha4j9qu”]

Leave a Reply

Your email address will not be published.