2K members online now
2K members online now
Learn how to reach mobile device users, drive phone calls, and increase app installs
Guide Me

Are Call Extensions Always A Good Idea?

Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

This is my first post, hi and thanks for any thoughts on this subject!


The account I run is for a 6 store regional brick and mortar car electronics chain. My campaigns were originally structured (I had inherited the account) in such a way that all of our product categories were in just one campaign with several ad groups that covered the whole metro area that we were targeting. In March I was advised by both my Google AdWords account manager person, and another reputable PPC agency whom I trust, to create a campaign for each location so that I could use call extensions. We have a different phone number for each store which is why this was necessary. Anyway, so I made the switch in mid-March and I have seen a few things happen:


1. Zero increase in phone call leads despite huge spend on call extensions, one store even reported a decrease in overall call volume.

2. Significant decrease in overall conversions and website form fill leads

3. An increase in time it took me to manage the account

4. Increased difficulty managing our budget because so much of our spend was going to calls


One thing that I noticed was happening was that a ton of the clicks were on our branded ads. Almost all of our customers require installation services (think of the customer flow like a mechanic) so I suspected existing customers were using the call extensions on accident to check on the status of their car etc. and so I removed the call extensions from those ad groups. That did help but not enough to reverse the 4 issues above.


I can't find anyone else online or in articles who has had this experience. The general consensus seems to be that call extensions always work awesome and you must use them. Is it just our market? Is there something else I need to check to make sure I'm not doing something wrong? Do you think switching back to the original campaign setup as an experiment makes sense to try is a good idea? Thanks in advance for the help!

1 Expert replyverified_user
Marked as Best Answer.
Accepted by Rakesh Kumar (Top Contributor)

Are Call Extensions Always A Good Idea?

Badged Google Partner
# 2
Badged Google Partner

"The general consensus seems to be that call extensions always work awesome and you must use them."


Opinions are like you know what, everyone has one. Call extensions obviously don't work if the business doesn't have a call center, and doesn't do any business over the phone. When it comes to multiple locations in the same market, call extensions and account structure can be tricky.


"Is there something else I need to check to make sure I'm not doing something wrong?"


What kind of call extensions are you using? Are you using call only ads?


"Do you think switching back to the original campaign setup as an experiment makes sense to try is a good idea?"


The account structure should reflect the structure of the website. If you have one website for the company, and a locations page, then anyone who lands anywhere on the site - Regardless of their location or search term - Can find the specific location details of the store they wish to patronize. Location targeting isn't perfect, and if a user wants to call or visit a particular location, get them to the site and let them make the choice. Rather than assuming you can always tell what a user wants, and trying to force them into a particular path.


As to the issues you are experiencing with his new structure:


1. Call extensions don't "cost" anything. Mobile click to call clicks are just that - Clicks. The cpc is what it is, whether that's a click to call click, or a click through to the website. When you are viewing the call extensions data in the ad extensions tab, those costs are for all clicks which occurred when a call extension was shown. The more impressions you accrue for call extensions, the more costs will accrue in that tab. That does not mean those are click to call clicks, or that you are accruing any costs for call extension impressions. You have to segment by click type to see how many actual click to call clicks you are accruing, and what the costs of those clicks are. A click to call click is not necessarily a phone call. Especially if you are using call only ads. A user may not intend to call the business when they click, and if their phone automatically initiates a call, they may abandon the process all together, and return to the SERPs to click on your competition. Without some kind of comprehensive call tracking in place, you really don't know what kind of actual call volume the advertising is generating or not.


2. Again, where are you landing the ads? If, as part of this account re-structuring you are landing all the ads on a specific location page for each campaign, that page may not be the best converting landing page. If you have a value based conversion strategy in place, and analytics tracking, then look into your behavior reports and find the highest value / highest volume page. If that's the homepage, then that's the best landing page to use for the control ad. Rotate indefinitely, and compare your best performing CTR ad using the highest value landing page, against any of your test ads. Don't assume that "consensus" means anything, or that what works for others will work for you.


3. AdWords is always an ongoing process. There are times when you have to put more effort/time in, and others when you have to be patient, and analyze data before you know what changes, if any, to make. If you have completely re-structured the account, you will have to put in a lot more time in the beginning to get the new structure dialed in. The current or previous structure may not be the best way to go. It can take years to find a structure that really works, and even then you still have to put in work over time. If the current structure is too much to manage effectively, then obviously some changes need to be made. However, that doesn't mean the account structure you inherited was perfect either. It's just part of the process.


4. Again, the only costs associated with phone calls are mobile click to call clicks. If your CTC volume and costs have increased, then you are getting more CTC clicks. You still need call tracking to know if those clicks are resulting in intended calls of any value or not.


Here's a post I wrote about AdWords phone calls and call tracking. It's been a while, but most of the information still applies. Take a look, open up the account, and compare your tabs and segments to try and find out how much phone calls are, or are not, actually costing you.


Can you cite the website please? I spent my first ten years as an adult in the automotive industry. I worked for a company with twelve locations and I had keys to five of them. The only thing I have done longer in my life is managing AdWords accounts, so this topic is of particular interest to me. I would love to continue the conversation, but without being able to view the website and do some homework, I (we) can't really offer any specific advice. If "consensus" is what has caused the issues, then specifics are the only path to resolve them. 


Are Call Extensions Always A Good Idea?

Top Contributor
# 3
Top Contributor

I started to type my reply and then noticed that @tomhalejr gave a full and  comprehensive one. I erased mine. I fully agree with Tom

Moshe, AdWords Top Contributor , Twitter | Linkedin | Community Profile | Ad-Globe
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer’

Are Call Extensions Always A Good Idea?

Badged Google Partner
# 4
Badged Google Partner

Thanks @MosheTLV, that's very kind of you to say. Smiley Happy 


Are Call Extensions Always A Good Idea?

Visitor ✭ ✭ ✭
# 5
Visitor ✭ ✭ ✭

Hi Tom, 

Thank you for such a thoughtful and thorough answer!


I'm going to try to answer some of the questions you asked:

-We are not using call only ads

-The website is: (we are just rolling out some ecommerce but that is a different thing altogether)


This: "The account structure should reflect the structure of the website." really resonated with me. We have one site for all of our locations and when we made the switch to separate campaigns by location I had a feeling we were making some weird geotargeting choices since our markets overlap. Once I switched it over to the separate campaigns by location I also started using store location specific landing pages with address/phone which as you say, may not be the most optimized since that person may have wanted a different location. 


"When you are viewing the call extensions data in the ad extensions tab, those costs are for all clicks which occurred when a call extension was shown." = I didn't realize that, so thank you! 


Thank you again for your help on this. I am the only person at my company who works on this so I am sometimes in need of a healthy discussion but there's nobody here to bounce ideas off of. This was SO helpful.





Are Call Extensions Always A Good Idea?

Badged Google Partner
# 6
Badged Google Partner

Since this is a single company with multiple locations, and not a franchise, it makes everything a lot easier. Assuming, for the time being at least, that you are focusing on service over online product sales, phone calls are still a valuable metric to focus on and optimize for. The difficulty here of course is trying to figure out an account structure which will allow you to use call extensions most effectively, due to the number of locations. There are several ways to approach account structure, and there are a lot of variables that go into the equation. You're going to have to experiment with things to find something that works for you. However, here's an example of how I might structure an account in this situation:


1. Brand: Since it sounds like you have some brand awareness for the company, I would identify and isolate the brand terms into their own campaign. Since you will always get the best bang for the buck when it comes to brand terms, you can't really lose with a brand campaign. I would dedicate as much budget and bids as necessary to achieve 100% of the potential search impression share. You can also be a little broader with your targeting - Consider the Charlotte Nielsen DMA region, or perhaps a 10+ mile region around the stores. 


You can just start with a single brand sorter adgroup, and then, if necessary, add additional adgroups for brand + location - If/when each location has enough volume to warrant the additional structure. Start with the broadest versions of those terms, but you will definitely have to keep an eye on your search terms report in the beginning. Exclude any obvious negatives, and if you do have a high volume of irrelevant queries coming through, then knock down the match type of the most costly terms with the lowest conversion rate.


If you already have some historical data, then use your best performing all time ad as a control, choose rotate indefinitely, and create a test ad based as a variation of your control. If you know through analytics what the most valuable page is, then use that as the landing page for the control, and then you can use the locations page (for example) as the final URL for the test ad. If you don't have analytics data, then the homepage is a pretty safe bet for your control ad. As long as you have some kind of conversion tracking in place, and you are only making subtle variations of the control ad text, you can be pretty certain that whichever ad has the higher conversion rate is because of the final URL. Now you can test variations of the ad message, using the best converting landing page.


That should help to have a positive effect on your conversion rates, and perhaps help to resolve the conversion issue.


When it comes to phone calls, in this case, I wouldn't add call extensions at the campaign level, or to the generic brand sorter adgroup. However, if you do start to see volume for brand + location search terms, then build out an adgroup for those (or each) location. Then, you can add a call extension with the correct number for each location at the adgroup level. If the user has qualified their brand query with a location, you know which location/phone number to show. If not, then they can still find a specific locations number on site. If you wait to accrue some actual search data (or you have it already), then you can avoid building out a huge structure of terms that may not show ads due to low search volume. It makes it easier to manage, and you know the time and effort are worth it. You may want to restrict the match types of these terms to modified broad, phrase match, or even exact, but you still have the broad brand adgroup as a catch all.


In the future, if you do want to expand into ecommerce, your brand campaign can easily be expanded across the country. If you don't have any brand recognition outside of the area, you won't accrue any out of market clicks, and it won't cost you any more money. If you do start to build some national brand recognition, you will be able to identify that interest sooner, and can easily adjust your budgeting, because you have a dedicated brand campaign.


2. Secondary KW Search: Assuming you still have some budget left, now you can move on to your next campaign. I follow the same approach to the brand campaign, with one big difference - Review your search terms report all time, for all campaigns, and all KW's. Sort by conversions, and identify the top five to ten exact match (non-brand) search queries. Again, you know these terms convert, so you want to isolate and optimize those terms to generate more conversions, and you know the time and structure is worth the effort.


I would still just start out with a single sorter adgroup. I would select match type based on budget. If you have money to spend outside of the brand campaign you can go broader to try and find more queries of value in the search terms report. If your budget is restricted, or these terms have too high a cost/conversion, then you can further restrict match type. As time goes on and you start to find additional queries/KW's of value, you can add those to your sorter adgroup. Then, once your KW volume grows to a point where it makes sense to break out additional adgroups you can. Since you know these terms convert, you will be reducing wasted spend, and hopefully increase conversion volume as well. Once you reach a point where you are limited by budget, then you will start having to make decisions as to what to cut back on. If you start at the top of the pyramid and work your way down, those decisions will be obvious, and easy to execute.


You're going to be in the same situation with phone call extensions for your secondary campaign. If/when you have enough conversion and click volume for product + location terms, then break those out into their own adgroups, with the correct call extension for each adgroup.


Where you can really get into trouble with your structure is if you assume that building out a huge product + location structure will work, if you don't have any data to support that assumption. It also becomes a problem if/when you have to cut back due to budget. You are going to be more hesitant to pause adgroups or reduce bids, because you put all the time into building out that structure. You also won't know exactly where your conversions are coming from, and you run the risk of pausing your best converting terms, because they are buried somewhere in a massive structure.


At this point, you only have two highly optimized campaigns. Which makes your structure much easier to manage, and should help improve your conversion volume issue. You know exactly where very cent is going out and where every conversion is coming in. Between your impression share columns and your campaign budget recommendations, you can not only make better budget decisions, but you can also more easily explain and justify budget increases to the owner.


If you have enough budget, eventually you will reach a point of diminishing return for your secondary search campaign. At that point, you can experiment with different campaign types by adding additional campaigns, but still maintain a manageable account structure. Because this is a single company, with a single site, in the same metro area, you don't necessarily need a campaign per location. You can break out additional adgroups per location - If/when there is enough location specific query volume to warrant the effort. Not every user qualifies their search with a specific city/store location intent. Even if they do, and you do have location specific ad copy for those location specific adgroups - Because of the website structure - You can do any kind of final URL testing you want, to find out what actually generates more conversions and phone calls.


Phone Call Tracking: Enough ranting about account structure. Smiley Happy In this situation, if you really want to track call volume through the the website, and from the advertising, the best solution is some kind of third party call tracking, which integrates with analytics and AdWords. Something like Mongoose, Call Rail, or Call Tracking Metrics as examples. These programs all have costs, and they require some amount of website coding on the back end. The advantage is that these programs will pull all the data through from click (entry to the site) to phone call, and report in analytics as goals (or imported adwords conversions), so you can get the data to further optimize the AdWords account and website to generate more phone calls.


There is the option to install AdWords on site call forwarding code, but the call data will only report for AdWords users, as conversions in the AdWords UI. It also will be a bit more complicated to setup, given your specific situation. However, the AdWords option is free, so the only cost is time.


Again, this is all just my opinion. Take anything of value to you, and discard the rest. Smiley Happy You will have to find something that works for you, but there are plenty of ways you can accomplish your advertising goals, with a more manageable, optimized account structure.