Call-Only Ads: 3 Months In

Share with:

FacebookTwitterGoogleLinkedInRedditPinterest


Back in July, I wrote a post called Call-Only Ads Are Ruining Mobile Results. In that post, I shared data from a client who’d launched call-only ads, only to see mobile impressions, clicks, and conversions tank.

Based on my readers’ excellent feedback, we optimized and refined those campaigns. We turned mobile back on in the regular, non-call-only campaigns. We continued to work on bids. We were desperate.

Results did improve, but not to their former levels. Here’s an update on what’s happening with call-only campaigns and mobile in general.

Call-only campaigns have been live for about 10 weeks. So I looked at the last 10 weeks, and then 10 weeks before that, before call-only campaigns came into existence. Data here is for mobile traffic only.

Call-Only Campaign Impressions Are Low

impressions

Impressions ended up nearly equal post-call-only, but only because we added mobile back to the regular campaigns. Call-only campaign impression volume is less than half of what it was before. There’s no way we’d be able to even show up in the auction to generate clicks if we stuck with call-only campaigns, much less generate conversions from call-only.

Call-Only CPCs Are High

cpc

Yikes, check that out. CPCs for call-only are 58% higher than mobile CPCs were before. And this is with our bid management software attempting to keep call-only CPC down. It’s crazy to me that CPCs have gone up this much. Then again, this is Google we’re talking about….

CPAs More Than Doubled

cpa

This is where it gets really scary and, frankly, sad. Call-only CPAs are similar to what we saw before. And that makes sense, because we were using call-only extensions before. Users didn’t have the option to click through on a mobile device – we wanted the phone call instead. That’s how call-only campaigns work, too – at least in theory.

But look at mobile not-call-only. Yikes. Since we don’t have the option of forcing a phone call in the regular campaigns now, we’re stuck with letting users click through on mobile – which is pushing up our CPA dramatically. And no surprise – while the landing pages are responsive, the call to action is to fill out a form, which we know people don’t like to do on mobile. Sure, you can call from the landing page too, but it’s not obvious like it is in a call-only ad.

And before you suggest that we create call-only ads in the regular campaigns, we tried that too. I was thinking we could force mobile traffic to the call-only ads and desktop to the regular ads. But it didn’t work that way. The call-only ads got virtually no impressions while the regular ads got tons, even from mobile. There seems to be no way to get this thing to work the way it did before.

So, by all appearances we’re stuck. Mobile is hugely important to this client, so we can’t just shut it off. Using call-only campaigns killed our impressions, clicks, and conversions. And adding mobile back to regular campaigns killed our CPA. We’re damned if we do, damned if we don’t. It’s a no-win situation.

Have any of you been able to find success with call-only campaigns at a good CPA? Is this yet another conspiracy by Google to improve mobile CPC? Have you found any hacks that work? Share in the comments!

Related Posts:

Comments

  1. I have struggled with call only campaigns.

    I think I know why… I had initially misunderstood how they work (and that never helps!). I had wrongly thought that a person clicked on the ad and the click triggered a phone call. This is not what actually happens…

    When a person clicks on the ad the phone opens up the phone application with the number pre-entered. The user still has to actually press the “place call” button.

    Here’s what I think happens. If you’re looking for a locksmith, this is a great option. What you really want to do is get a locksmith to open your door as soon as possible. But if you offer a different kind of service – let’s say you`re a real estate agent – this is simply too aggressive a move. It is akin to the salesperson who lunges forwards the moment you cross the threshold of their store. The telephone sales rep. The ambulance chasing lawyer… in fact, all the techniques we tend to despise rolled into one.

    I am looking for a new car… so I search for the local dealership. I click on the ad. What I’m actually expecting is to go the website so I can take a look around. I want to get to know them on my own terms and in my own time. Instead this ad literally hijacks my visit and tries to get me to call the car dealer. I drop the phone like it’s on fire – Noooooooo – not yet, cancel, cancel, cancel.

    Until users get used to this type of behaviour from advertising, I don’t think we’re going to be able to get them to work. I for one will let other advertisers blaze the trail with their budget before I use call only ads again.

    • Melissa Mackey says:

      Hey Steve – In general, I agree with you. If the user wasn’t expecting or ready to make a call, it’s a bad experience for sure. In our case, though, we were using call-only extensions, so the user couldn’t access the site before. This client gets 85% of their conversions over the phone, so we WANT phone calls. It’s also a service that is somewhat urgent when the need comes up. My problem is, it isn’t performing like it was before, even though theoretically it’s the same. That’s why I’m so frustrated. Thanks for your comment!

  2. Aside from any weirdness on Google’s part with call-only campaigns, I think Steve is very correct that people are not expecting a call to be started from clicking on an ad that does not indicate this will happen. Duh, Google!

    We also tried CO, but set up a new account apart from our other campaigns. Test new crap from Google that way. For one thing it may avoid any drop in impressions. We ran it for a while and the end results were poor. I don’t have the details, but our main criteria is getting conversions at lower cost. It failed. And we tested sending traffic to about 6 different sites, using our former best mobile ads.

    Along with what Steve said, there’s simple the fact that these ads are CRIPPLED. The only thing different is you have removed a possible conversion path for the user. The only way they can work is if a very high percentage of your potential clicks are already motivated to call.

    Now, if you want to run a call-only campaign I suggest you experiment with the following:

    Have one campaign like you would normally. Once you have gone through the refinement period all new campaigns have, duplicate the entire campaign. In the new campaign (which you name as being a mobile campaign) you adjust your bids. Keeping in mind what your mobile CPCs are, you jack up your mobile bid adjustment to +300%. Then you lower your max bids so that 3x the amount is about where you want to be for mobile. This will cut your desktop bids off at the knees and allow your mobile ads to run free.

    In our tests we STILL get some clicks from desktop/tablet despite the bids being so low, and in our case the few D/T conversions that we see have a really low CPconv. Weird, huh?

    So what about the first campaign running normally for all devices? You may have already lowered your mobile bids by 100% to shut them off. That would be logical, right? Remember, this is GOOGLE we are talking about. I recommend leaving them on as is and keep an eye on everything. Make adjustments only if needed. You may find that more is indeed more.

  3. Melissa, Just read your response to Steve’s post. As we all know, it can be really hard to know what is causing some results that we see in the accounts. I was impressed by your stats, but given you last comment I would like to see stats going back 12 months. And then understand if there were other changes not mentioned in your OP. I don’t need to see the information, I am just saying that is what I would want to look at before I conclude anything. It could be a result of potential traffic and competitor bidding. If there was less available traffic then it would follow that bid competition would increase.

    Sometimes I think we “over-portray” Google as the villain. Not that anyone can blame is for doing that. 🙂

    • Melissa Mackey says:

      Good call, Chris – I’ll go back and look at more data. Probably not 12 months since we completely revamped the client’s strategy less than a year ago, but I agree that looking at a bigger data set makes sense. Stay tuned!

  4. I tested out Call only ads as well.
    I had lack-luster results and mentioned it to one of the google reps that we work with.
    The response was that they were developed for people who had a bad mobile experience on the website.

    I found that the call extensions worked much better (lower CPA) for generating phone calls.
    The website was responsive and had an easy mobile conversion path/experience.

    Just my 2 cents.

  5. What does this Internet marketing pioneer think about Google call only campaigns?
    Don’t waste your money!
    This program would be better labeled BETA
    People “accidentally calling” on CPC’s approaching $50.00.
    This reminds me of shady internet marketing programs offered by criminals.
    Does Google really need money this badly?!?
    Now you know why the DO NO EVIL mantra has been omitted from their parent company Alphabet.

  6. Kurt Henninger says:

    Melissa,

    Thanks for the your update and perspective on this issue. In a previous post, I had given my own take on the issue of these campaign types.

    I still have the same problems with them and have pretty much given up on them except for a “test” campaign or two and converted back to “call extension” heavy campaigns for clients who want calls.

    I have had multiple issues with these campaigns, low impressions, high cpcs, low quality scores.

    Overall even though the “click to call only” functionality has been depricated on more traditional campaigns and call extensions, the “older” version still out performs call only ones by a mile.

    Also, I’ve noticed that the “quality” of the calls from the campaigns using call extensions has increased. While this is a very squishy number itself, I notice fewer dropped calls and a higher “call length”.

    • Melissa Mackey says:

      Yep, call-only campaigns are definitely a failure in my experience. It’s too bad, because call-only extensions used to work great.

  7. I know that most of this post focuses on the distinction between performance of call-only ad extensions vs. call only campaigns, but I have a somewhat related question to add to the discussion.

    Has anyone noticed a drop in overall call volume in their regular campaigns since CO campaigns launched? I have mentioned it in passing to my reps in relation to specific accounts, but received no comment on the issue. I have seen it happen regardless of having a CO campaign (or not) in the same account as my regular campaign, and I don’t believe it’s due to an influx of competition for these specific, B2B niche clients. Just curious.

    I’m with you, Melissa. I don’t see the value in CO campaigns, and the cannibalism of ad serving when you might have overlapping keywords (regular campaign with CO campaign) makes them a pain to manage the user experience of the ads effectively.

  8. I would only use call only for campaigns where you are soliciting a response for people who are in the informational stages of purchase only with an offer to answer their questions about the topic on the spot and tell the client to put his or sales hat on. This is what I often do anyway….

Speak Your Mind

*

This site uses Akismet to reduce spam. Learn how your comment data is processed.