What One Million Dollars In Spend Has Taught Us About Enhanced Campaigns

By , President at Hanapin Marketing

470 SHARES

Enhanced campaigns are as bad as we feared. Hanapin has migrated many of the accounts under their management and have found that performance is typically slipping; and sometimes it is even cliff diving.

This post analyzes ~$1 million dollars of that ad spend pre and post enhanced campaigns (ECs) and uncovers where performance has dipped, how spend allocation has changed (due to less control) and a few takeaways we’ve learned for how to stay ahead of the competition on this.

A quick note about this data is that while CPA is up across the board this is partly due to seasonality for a few major accounts. It’s obvious though, that CPA is up on computers for Search and Display but that mobile and tablet CPA is up at a much, much larger percentages and has started to make up larger percentages of budget allocation. This is also macro data so one account with high CPA goals that has increased budget over the past few months could drastically skew these numbers.

 

The cliff notes are that since launching enhanced campaigns:

  • Mobile CPA is up 40% (this is with negative 20-50% bid multipliers in most campaigns)
  • Mobile spend is up 80%
  • Tablet CPA is up 13%
  • Tablet impressions, clicks and cost are up drastically
  • Total CPA is up 17%
  • Total conversions are down 7%

Some accounts, I will say, have seen increases in CPA but they were so far below goal that the extra sales were worth it even at the extra cost. All in all though, ECs have been detrimental to most accounts. In many accounts we have had to opt out of mobile completely (bid -100%) and are seeing inflated CPA due to a sudden increase in Tablet traffic.

 

Now, here’s the data for all those super-PPC nerds who want to know what they are in for. This first section is looking at ~$1 million dollars in ad spend across ~50 accounts pre enhanced compared to post enhanced WITHOUT segmenting by network (Search Vs. Display):

 

Change in Primary PPC Metrics

Table showing enhanced campaigns has spiked CPA by 17%

The above table shows the change in primary metrics by device for before and after migration. In other words, impressions on computers has increased 5% since the migration, clicks are down 13%, conversions are down 12% and CPA is up 12%.

These numbers all lead to more budget being soaked up by mobile and tablet devices and thus while there are still impressions out there to be had for desktop, I am not able to allocate my budget to that better performing device. For example, computer clicks are down 13% but mobile and tablet clicks are both up 45% (and mobile CPC is up 24%)!!! With CPAs that have increased by 40% and 13% respectively! Yes, a 40% increase in mobile CPA and that is with mobile bid multipliers that average greater than -30%.

 

 

Change in Allocation of Spend/Clicks/Conversion After Enhanced Campaigns

This shows the changes in percentage of spend coming from mobile devices and tablet devices since the migration. The reason this is being pointed out is that you can no longer control budget allocation by device (which you used to be able to do by targeting devices in different campaigns and using campaign daily budget to control spend).

Mobile Devices

If we look at the percentage of impressions, clicks, conversions and cost coming from mobile devices (compared to desktop and tablet) we see the following results:

Enhanced Campaigns has resulted in mobile spending a greater percentage of budgets

So this means that mobile has gone from making up 10.2% of total spend to 16.8%. And remember, CPA has gone up 40% to go with that extra spend. Sweet.

Tablets

For mobile, we can at least adjust the bid multipliers to virtually opt out of it. No such luck with tablets. Let’s take a look at what’s been happening there:

Tablet spend allocation has not spiked as much with Enhanced Campaigns

 

Much less alarming numbers here. Cost is up from 10.2% of total spend to 11.6% and as the very top chart showed, CPA is up 13% which is pretty close to an increase in computer CPA we’ve seen. The only real issue is that tablet CPA is 42.5% higher than computer CPA to start with. So while it hasn’t performed much worse since the migration, it is performing much worse in general and now we have no shot and stopping the damage it is doing.

 

Enhanced Campaigns by Search and Display Networks

 The bulk of performance change came from display. Interestingly enough, impressions are up significantly but clicks are down. Spend is up in both networks and CPA is up a staggering 38% in Display.

Table showing that display is performing much worse on enhanced campaigns

More specifically, Display CPA is up 67% for mobile devices! Search CPA on mobile devices is up 40%. Computer Display and Search CPA is up 28% and 8% respectively. So you can see that while CPA is up across the board, it is certainly up even more drastically for mobile devices. Lastly, tablet CPA on search remained stable but is up 60% on Display.

 

A Handful of Enhanced Campaigns Takeaways

It’s a bit ugly in the ECs space right now. Optimizations are difficult (you can’t pull conversion data and spend data in the same place as bid multipliers so you have to merge spreadsheets to actual measure impact of bid multipliers and make optimizations based on it), you are best off bidding -100% on mobile until you can slowly turn it up and get it running at acceptable CPA’s, Display is the hardest to optimize with the changes and the impact is very clear.

We have also found that the lower you bid on mobile the worse mobile performs. Meaning you are better opting out then going in at -50% because you only get bottom of the barrel clicks at a mobile bid that is -50% of your desktop bid.

The better structured your accounts the better results are. Those accounts that are very segmented so making ad group level bid modifiers affects fewer keywords and those keywords are more tightly related perform better from enhanced campaigns.

So what have you guys seen? Or is everyone else still holding off until the night before the migration, I mean “upgrade”, deadline?

 

The Five Laws of Social Advertising

Twitter Facebook LinkedIn Google+ Email Print More
  • Simon Corbett

    Awesome post confirming what we were all fearing and what Google whole reason for making this massive change – more money in Google’s pockets!

    -100% mobile bids all the way! And fingers crossed that either tablet traffic and costs calm down or Google backtrack a little and realise computers and tablets ARE NOT THE SAME and give us the ability to make tablet bid adjustments.

    • JeffreyAllen

      I do think we will get more control (bid multipliers for tablets) and reporting will get better (conversion data next to the bid multiplier settings). We have had some wins with EC’s too, which are glossed over since I was only reporting on a macro level. But yeah, it’s been rough!

  • http://www.facebook.com/mr.larry.kim Larry Kim

    nice article title!

    • JeffreyAllen

      Thanks, Larry!

  • Chris Oliver

    Thanks for a great post which confirms what we all anticipated. Google were previously encouraging us to set up separate campaigns for desktop, tablet and mobile. It was a good idea however it increased the workload.

    We all know (and so does Google) that different devices perform totally differently and bids need to be adjusted accordingly. A very simple solution would be to allow advertisers to have three bid multipliers within the campaign settings – desktop, tablet & mobile. How easy would that be !!!

    CPA’s are climbing and we’re starting to see clients shifting AdWords budget to fund other activities and channels such as Facebook. We need to get control back!!

    • JeffreyAllen

      I think that what you are talking about here is exactly what will happen. People will shift budgets, then CPC will come down in Google and things will settle down.

  • http://twitter.com/RoboticSoul Russ Denman

    en·hance

    tr.v. en·hanced, en·hanc·ing, en·hanc·es

    1. To make greater, as in value, beauty, or effectiveness; augment.
    2. To provide with improved, advanced, or sophisticated features: computer software enhanced with cutting-edge functionalities.

    • JeffreyAllen

      Yup.

  • http://www.webranking.com/ James Svoboda

    >> Meaning you are better opting out then going in at -50% **because you only
    get bottom of the barrel clicks** at a mobile bid that is -50% of your
    desktop bid.

    Yep. I came to the conclusion a few years back that using bid modifiers for day parting was one of the worst ways to manage a campaign. Using them for device is not in anyone’s best interest. *cough* Except for Google’s. #solidpost

    • JeffreyAllen

      Thanks, man! I noticed the same thing about day parting but I honestly hadn’t made the connection until now!

    • http://twitter.com/GnosisArts Gnosis Media Group

      I like how you put that all together, James. Great actionable analysis – I don’t think I’d have ever gotten to that conclusion. Adjusting mobile bids to -100% now for a GDN campaign to gauge effects.

    • Stefan van Vliet

      Hi James, could you explain a bit why bid modifiers for dayparting are bad news? Im very curious since Im using it for my own campaigns. Thnx!

      • http://www.webranking.com/ James Svoboda

        Hi Stefan,

        Jeff summed it up fairly well. When you reduce your bids through campaign or ad group level bud modifiers it applies to all keywords. You really are losing bod control. Just because you want to pay less during off-hours does not mean that your competitors are as well. If they are bidding the same, when you are bidding less, they will get the impressions and clicks for valuable keywords and you will usually be left with broad, BMM or phrase matched queries… which will typically be lower converting keywords.

        • Stefan van Vliet

          Clear, thanks!

    • http://www.jeffalytics.com/ Jeffsauer

      You mean taking away granular bidding opportunities that we use to find efficiency in campaigns and replacing it with less data + a percentage based bid means we spend more? This is why I hated “enhanced” campaigns from the beginning. Takes the power out of our hands and makes the system a black box.

  • http://twitter.com/GnosisArts Gnosis Media Group

    This was the day I feared …

    • http://righteousmarketing.com robertbrady

      The day you fear is July 22nd.

  • http://twitter.com/johnucciferri John Ucciferri

    thanks for posting Jeff…very needed to see some early results…hope they improve as it evolves though i am skeptical.

    • JeffreyAllen

      We have definitely got better as time has gone on, I bet if I only looked at the last week the performance would be much better than the first week we had accounts in enhanced campaigns…

  • http://twitter.com/PPC_Shinobi Alex Guest

    Really great post and some interesting stats from Hanapin’s accounts.

    My area of concern is tablets on Display campaigns and these have really taken a beating, killing off what was once a great remarketing CPA for one client. How have your remarketing CPAs been on tablets for your e-commerce clients?

    Agree with Simon here that tablet multipliers would be greatly appreciated and definitely make life a lot better for us account managers.

    • JeffreyAllen

      I do think we’ll get tablet multipliers. Tablet CPA has been okay in remarking, as a general rule. But very poor in other display campaigns.

  • LukeAlley

    One of the best PPC Hero articles I’ve ever read. Great info guys. A lot of benefit in this post.

    One question, I’ve heard that some advertisers are still creating “desktop/tablet focused campaigns”(-100% bid for mobile) and “mobile focused campaigns”(+300% bid for mobile) into separate campaigns (pre-EC style) and then trying to use bid multipliers to direct the traffic accordingly. Have you done this and is it effective? Or do we need to just bite the bullet and combine and manage them all from one place?

    • CassieAllinger

      +++ to Luke’s question about “hacking” EC to desktop & mobile focused campaigns instead of merging. It seems like the system was designed to force this not to work. But who’s tried it? And results???

    • Jacob W

      I would also love to hear about any results anyone has had doing this.

    • Shane Jennings

      Sounds like we could use a follow up post! I’m also very interested to see if any of your accounts used workarounds and if the mobile impact was significant on those accounts. If you chose not to use workarounds was it for ease of transition or business preference?

    • LukeAlley

      Any PPC Heros out there willing to comment?? :-)

      • JeffreyAllen

        We haven’t tested it yet, that’s why the silence. Sorry. It’s actually next up on my list of things to experiment with, though. My main concern is that we end up getting a lot of cross contamination, though. Desktop clicks in the mobile campaign that because the bids are so low are probably bad clicks,

        • LukeAlley

          That’s my thinking also. Seems counter-intuitive considering that Google is trying to make some of the management easier on us.

          Thanks for the insights!

  • http://ppccampaigngenerator.com/ PPC Campaign Generator

    I too put implemented bid multipliers in the -30% avg. range like you. I think the comments below are correct in saying that Google will make tablets also have bid multipliers. Let’s summarize bid multipliers: bid multipliers suck.

    There’s no doubt introducing a massive new change like Enhanced Campaigns will cause volatility to the Adwords system as many advertisers take an educated guess on the best way to upgrade (and those advertisers that just upgrade and go home :) ). This volatility definitely causes inefficiencies which Google loves because it makes them more money. I’m certain the creator of Enhanced Campaigns gladly took this into account when pitching everyone on the notion that “Context is everything, people are on many different devices on the go, blah blah blah…”

    I’m curious though, why would you are see worse performance for a -50% or greater bid modifier? If CPCs are lower, I’d imagine you are in a lower position but you’re saying click quality is also lower? Is this true on your best-performing mobile keywords?

    Great article to keep the convo flowing.

    • JeffreyAllen

      If you think about it, a -50% bid modifier is going to often mean you are below first page bid for the most competitive (and so most likely best performing) keywords. So the only thing you end up showing for are the keywords that are less profitable so others aren’t bidding on them.

    • Dragan Pozder

      Broad match (and phrase to some extent) usually gets less efficient with the lower bids.
      I will wait until the deadline. Since many advertisers will also wait, everyone’s Adwords stats will change significantly after the July 22nd.

  • David Riedmiller

    Great post, guys. The thing I hate most about enhanced campaigns is that it is impossible to have different mobile target CPA’s when you’re using CPA bidding. Multipliers do NOT work when you’re using CPA bidding. Seriously, Google, are you kidding me? Demographic multipliers and all, but NO way to account for different mobile CPA’s? So far, the only solution I have is to duplicate campaigns, set the mobile multiplier to -100% in one campaign and have the other campaign running on adjusted (lower) CPA’s w/ mobile opted in. Any other ideas?

    • JeffreyAllen

      That’s about the best work around we’ve come up with.

      • Moritz Helmstreit

        I think so too, but you’ve got a lot of dublicates in your accounts then. So to identify those it is getting much more complicated. If the dublicated campaign is a product-campaign, you have to care about two campaigns, if a product changes (URL or whatever).
        Also you’ll waste money in the second “mobile campaign” on computers… >> However, I have no better idea than that…

  • fixmedinner

    I like the idea of setting bid modifiers at much higher (negative) percentages for mobile devices. However, I would be weary about its affect on quality scores. Have you seen any negative impact in this regard?

    • JeffreyAllen

      I haven’t seen any impact on QS and not sure why it would affect it? Would love to hear your thoughts on that!

      • James Scaggs

        Wouldn’t it hurt your QS if you incur boatloads of impressions at a terrible CTR? That essentially makes it much harder to build a healthy CTR by starting out so low.

        • fixmedinner

          Well put. This is exactly what I would be concerned about – low CTR hurting QS.

  • Ryan Campbell

    Great insights and analysis. We’ve been seeing the same thing on accounts we’ve “upgraded”. :(

    • JeffreyAllen

      Sort of glad to hear we aren’t the only ones! We’ve been getting a handle on it. Just far from simple.

  • http://www.jordonmeyer.com Jordon Meyer

    Yikes! Thanks for digging in and sharing this great insight with us! We have all feared these outcomes and it’s nice to see numbers next to the theories. Now let’s all go buy some GOOG before July 22.

    • JeffreyAllen

      Don’t think I haven’t thought about this/made a small investment in my future…

  • FloydZest

    You guys crushed this post with that massive amount of data! Thank you sharing. I’ve experimented, but I’m dreading July 22. Does anyone know if Google has created updated videos in the learning center?

  • http://www.jeffalytics.com/ Jeffsauer

    Thanks for providing the data behind what we intuitively thought would happen. It’s easy to say it, but you proved it. Thanks Jeff. Great post.

  • James Scaggs

    Hmmm. It sounds like mobile clicks and CPA are up not bc enhanced campaigns are inherently bad but bc there are suddenly more and more advertisers targeting mobile automatically bc of enhanced campaigns. Also if you are launching a new EC then your quality score has to build up all over again. By entering the mobile auctions with such reduced bids you are fighting an uphill battle on building CTR on those campaigns. I’ve seen this in some accounts but for the ones I’ve switched to EC and come out aggressively with my mobile bid multipliers I have been able to establish strong CTR and ten optimized to the point that my ECs are now outperforming the old campaigns. This is especially true for clients with local intent where mobile is so important. Just food for thought.

    • JeffreyAllen

      I don’t think enough advertisers are in EC to make the increased CPA be a result of that. I do not expect to see that spike until late July. A good chunk of the issue is the garbage traffic the lower bid multipliers drive. I’m happy to hear you have seen some good results though!

  • Charl

    A very interesting read, thanks for it.

    I’ve seen a mixed bag of results across our markets. Some of them have really benefited from Enhanced Campaigns, while others are still finding their feet.

    One of the clearest signs I’ve seen is a steady flow of paid traffic through the week. Pre-Enhanced the peak traffic and conversions were always on a Monday and then slowly dies off during the week, Friday being the lowest and then a slight spike on Saturday. Now the amount of traffic coming in stays about the same all week. Mainly due the daily budget allocations now run across all devices, so as the PC traffic dies down during the week, Mobile is there to pick it up and keep going with it. Tablets seem to cover everything inbetween.

  • Ben

    For some accounts, bidding down by -50% would be alright for some accounts though right? Assuming you already have a high non-mobile bid?

    • JeffreyAllen

      For some accounts it will be okay. But on average it will not. I am currently working on a new account structure idea revolving around keyword relevance + Average position. If you group keywords by average position this will help to better understand what the impact of using bid multipliers will actually be, versus just paying more for the same clicks.

  • Antonia

    Thanks for the great post! That’s exactly what we were afraid of, so we were actually putting of the “upgrade” until now… Especially since we followed Google’s best practice for the last couple of years and had separate campaigns for each device – all with very different performance and bids on all bid-levels. This makes the migration even more painful, since all campaigns were nicely set to meet the target CPAs…
    At least Google figured out that bid modifiers on adgroup-level might be something useful… Still not very happy with the “upgrade”…

  • Robin Havervall

    Nice post! First I thought enhanced campaigns made it hard to control the mobile-specific campaigns that were set up. Working with mobile-specific ad text and keywords that were similar to “desktop-campaigns” etc. This was solved with a -100% bid on the other campaigns.

    Although, I like that the ad extensions can be done on an ad group-level. Some drawbacks and some good things came out of this I guess.

  • gravymatt

    One thought that I have had in the back of my mind is say you make the jump in EC campaigns, and bump along optimizing. Then, July 22 comes along and everyone is thrown into the mix and the playing field is altered again. Will all the optimizing be washed away or just 50% of your effort?

    And the second thought is suppose you again make the jump in EC and give up control over your tablets and reside to Bid Modifiers, but then July 21 Google says “Ok ok we will allow you to turn off the tablets., would all that work be for nothing?

    • JeffreyAllen

      I believe that all the optimizations won’t be washed away when everyone else migrates. We’ll better understand what is baseline and then can adjust as everyone else is trying to find theirs. We will likely have campaigns were we bid to -100% for mobile during this period but then will go back to our optimized settings once the marketplace settles down.

      I really hope they give us control over Tablets back! I don’t think it changes anything if it’s a new feature. In fact, figuring everything else out first will allow us to focus solely on Tablet performance…