Search Engine Watch
SEO News

Go Back   Search Engine Watch Forums > Search Engine Marketing Strategies > Vertical Markets & Specialty Search > Local Search
FAQ Members List Calendar Forum Search Today's Posts Mark Forums Read

Reply
 
Thread Tools
Old 06-13-2006   #1
Online Survey Diva
Member
 
Join Date: Sep 2004
Posts: 26
Online Survey Diva is on a distinguished road
Google Local Is Not a Converter.

Running Ads in Google Local vs. Main Google Search -- Disappointing (So Far)

I've been working on this case study for a while now and it was published today in Marketing Sherpa. We tested sales conversions for a mattress company, Dormia, on Google Local ads versus Google.com ads and found that while Google Local ads received significant clicks, they just aren't delivering the actual sales conversions many thought it would versus Google.com ads.

Quote:
Our theory is at Google.com, people are buyers and they are looking for a local store that sells mattresses. However at Google Local, the most searched for term tends to be store name, not category. It's possible people are only going to Local if they are actually searching for the store itself."
Anyone have different results?
Online Survey Diva is offline   Reply With Quote
Old 06-14-2006   #2
GAustralia
Member
 
Join Date: Oct 2005
Location: Sydney, Australia
Posts: 194
GAustralia is on a distinguished road
Reply

Online Survey Diva -

Google poses a conundrum for local advertisers. Locals have lots of tools to limit the extent to which their ads show up: limits by country, state, city, kilometers etc. However, other advertisers are not so limited as their broad, phrase, exact, geographic, etc match rains on the parade of locals. The result is that locals have to bid a lot more against others more geographically dispersed to get to the front page.

Online Survey Diva have you analysed geographic words with your keywords? Such as "Mattress North Pole." If you purchased this adword and you are the only one in the world who has done so, when someone types in "Mattress North Pole" you must compete with everyone who broad, phrase, exact, etc. matches any and all combinations of each of the words in "Mattress Noth Pole."

Maybe you have some advice for local campaigns? My business is based in Australia so we have not yet had the option of Google Maps Local.

Regards,

GAustralia
GAustralia is offline   Reply With Quote
Old 06-14-2006   #3
Chris Boggs
 
Chris Boggs's Avatar
 
Join Date: Aug 2004
Location: Near Cleveland, OH
Posts: 1,722
Chris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud of
Hi Diva...thanks for the interesting topic. I am curious about the last statement made in the case study regarding the effect some listings within Local have on some searches within regular Google results. Could you elaborate with examples (even in an IM?)?

I was also wondering how many coupons (roughly) were actually printed from Local, or if the coupon printing was not measured? Lastly, are you positive that the tracking code on the coupons worked properly in relation to the source of the traffic?

I am not trying to sound skeptical, just digging for more. Thanks again!
Chris Boggs is offline   Reply With Quote
Old 06-14-2006   #4
Online Survey Diva
Member
 
Join Date: Sep 2004
Posts: 26
Online Survey Diva is on a distinguished road
Smile

Chris,

I'm not sure if you've done searches like this on Google.com before, but if you try searching on a keyword that is geographic in nature, such as "mattresses albany new york", you will see that Google places the top three natural results ABOVE the natural website ranked results, truly making the local results the top rankings.

Working for this local-based retailer, we've been working on trying to get the stores higher in the natural Google Local listings, but most of the rankings there are a mystery. I came to the forums some time back to see what I could find, and no one answered my question -- that's how few people know much about Google Local's natural rankings!

I called my Google rep to discuss it, too, which you know didn't get me very far! An interesting search I did with her in Google Local was "mattresses seattle washington". The top three results included an appliance company -- and it outranked Select Comfort, a known mattress brand. The appliance company didn't have "mattress" in its title or url -- NOTHING. And it's link went to a chamber of commerce website. So how does one location or brand outrank another in Google Local natural listings? It's somewhat of a mystery.

I'm not sure there's a way to measure the printing of the coupon vs. use of the coupon. In other words, how many people printed it vs. how many people brought it into the store. Do you know of a way? I couldn't find a log analyzer that might tell me that either.

I'm definitely positive that the tracking code worked properly. My conclusion is this... if you also look at the info from Hitwise (link at the bottom of the article), it appears that brands are heavily searched on Local, not generic terms. So it appears that people have a store in mind when they go there and are really looking for the nearest location. Therefore, Local isn't acting like the Yellow Pages, per se. It's really a store locator tool at this time; whereas Google.com is more of the yellow pages in the sense that people browsing to decide on a brand use more generic searches there, such as "mattress".

Just my two cents.
Online Survey Diva is offline   Reply With Quote
Old 06-14-2006   #5
Online Survey Diva
Member
 
Join Date: Sep 2004
Posts: 26
Online Survey Diva is on a distinguished road
Smile In response to GAustralia...

GAustralia:

Thanks for the comments. I sure did use the geographic terms in the keywords too, but no go. What I found out later was the Hitwise information, which I mention in my reply to Chris. Very interesting stuff.

It at least helps me know where I should focus the bulk of my time, and Local isn't really doing much these days.

That's interesting on the international angle with Local. I'm not experienced with it internationally, but check out the Hitwise data for the US. Unless things are radically different in Australia, I'd spend more time with Google.com -- Google Local just isn't perfected yet, and frankly, it needs a lot of work!

Diva
Online Survey Diva is offline   Reply With Quote
Old 06-14-2006   #6
Jonathan Mendez
Member
 
Join Date: May 2005
Location: NYC
Posts: 42
Jonathan Mendez will become famous soon enough
No offense, but what I read was more a story than a case study.

Without relevant metrics it's impossible to validate or even reply to the assumptions. Frankly, I'm surprised Sherpa would run this without more data.

As far as user's goals with queries, it's difficult to ascertain if the users wanted detailed information on the store or were they just further down funnel in purchase mode.

It's a very interesting subject. Especially as local gets pushed more and more by G and opens up a whole new world with mobile search. I'd love to see the numbers if you can share them.
Jonathan Mendez is offline   Reply With Quote
Old 06-14-2006   #7
Online Survey Diva
Member
 
Join Date: Sep 2004
Posts: 26
Online Survey Diva is on a distinguished road
Results

Jonathan,

I'm not responsible for what Sherpa writes, but since you brought it up, the data is as follows:
  • We did the test for two weeks.
  • The offer was the same for both Google Local and Google.com ads.
  • The ads were relatively the same (with small exception on ad title based on Google Local requirements)
  • We found a higher clickthrough rate on Google Local than Google.com
  • However, we found that Google.com had much higher in store coupon usage. In fact, we received 0 coupons from Google Local.

We can't release actual data (only percentages) because of proprietary info, and Sherpa did not want to use anything but percentages either because they also do not want to publish data proprietary to a particular company. That's their policy.

Unfortunately, we can't track how many people PRINTED the coupon, which is the other metric I'm dying to know. So if they do click through, how many print the coupon, then compare that with how many USE the coupon in the store.

I think there are lessons to be learned here about Google Local ads vs. Google.com ads, and the Hitwise information was very interesting to me as well.

Diva
Online Survey Diva is offline   Reply With Quote
Old 06-14-2006   #8
Jonathan Mendez
Member
 
Join Date: May 2005
Location: NYC
Posts: 42
Jonathan Mendez will become famous soon enough
Can you share what percentage of you overall impressions and clicks were from Local?
Jonathan Mendez is offline   Reply With Quote
Old 06-14-2006   #9
Online Survey Diva
Member
 
Join Date: Sep 2004
Posts: 26
Online Survey Diva is on a distinguished road
Impressions

Jonathan,

Sure thing. Here's the breakdown:

Google.com Total Impressions:
462224

Google Local Total Impressions:
3630

Click through rates on Local was just over double what it was on Google.com ads.

If you read the Hitwise info, you'll see that Local only accounts for something like .1% of all Google domain traffic, whereas Google.com is over 80%, so that helps put the numbers in perspective. There's obviously a lot more traffic on Google.com than Local, although we don't know the true total traffic numbers from Google.

Diva
Online Survey Diva is offline   Reply With Quote
Old 06-14-2006   #10
Jonathan Mendez
Member
 
Join Date: May 2005
Location: NYC
Posts: 42
Jonathan Mendez will become famous soon enough
Well without knowing your CTR I guess my issue is that based on 3,000 impressions you don't have a large enough comparative sample size to state that "Google Local is Not a Converter."

It maybe so, but without knowing how many clicks you recieved and what your conversion rates were for each it's impossible to validate how meaningful your story is for marketers.
Jonathan Mendez is offline   Reply With Quote
Old 06-14-2006   #11
eWhisper
Member
 
Join Date: Dec 2004
Location: Chicago
Posts: 42
eWhisper is on a distinguished road
The first question that comes to mind is are you sure of the impression numbers as it relates to Google Local?

With the Local Business Ads (I'm assuming this was the ad used?), they can show up on Google.com and other properties as a regular ad.

Therefore, when you list impressions and visitors from Google Local, are you using the stats from your AdWords campaign, or from your tracking system?

With the local business ad, was the logo used? Was the offer local in nature? Did you track phone calls from the ad (with the Google Local ad, you can put a phone number directly into the ad as another line), etc.
eWhisper is offline   Reply With Quote
Old 06-14-2006   #12
Chris Boggs
 
Chris Boggs's Avatar
 
Join Date: Aug 2004
Location: Near Cleveland, OH
Posts: 1,722
Chris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud of
Quote:
Originally Posted by Jonathan Mendez
Well without knowing your CTR I guess my issue is that based on 3,000 impressions you don't have a large enough comparative sample size to state that "Google Local is Not a Converter."
I have heard that statistically 50,000 impressions and 500 click-throughs are the minimum required to achieve a sample worth trusting.
Chris Boggs is offline   Reply With Quote
Old 06-15-2006   #13
Online Survey Diva
Member
 
Join Date: Sep 2004
Posts: 26
Online Survey Diva is on a distinguished road
In response to eWhisper

eWhisper,

Well, I look at it this way... if that is true that the ads were also appearing on Google.com, then that just means that the Google Local true numbers would be even LESS! It is my understanding that they do not show up on the content network, and I can tell you that I do not activate my ads for this geotargeting on the content network.

The numbers are from both Google and from our web stats.

The logo was used.

The offer was local (featured the mall store, ex: get 10% off at our Crossgates Mall store - print the coupon). We do also track calls on Google Local with a special 800 number for each store. But calls for us (mattress company) do not equal sales -- store visits do, so we had no conversion based on calls. We also tried to incentivize people to print the coupon, hoping that would lead to better tracking.

Bottom line... offline tracking is challenging, but also, we're not sure if the fact that we're dealing with a high ticket item may also have lead to low conversion.

Our take on it is that Google.com is the place for us to focus, not Google Local. In my mind, it's not quite ready for prime time. Just the traffic stats for Google Local show that it's really not a strong player right now. I'd rather focus my efforts where I get the most ROI -- google.com.

Diva
Online Survey Diva is offline   Reply With Quote
Old 06-15-2006   #14
Online Survey Diva
Member
 
Join Date: Sep 2004
Posts: 26
Online Survey Diva is on a distinguished road
Response on sample size...

Frankly, I think we're close enough to determine this, and based on my past experience with surveys and sample size, it is my understanding that this sample size is large enough.

The issue here, of course, is not the clicks. It's that we had lots of clicks and no conversions. Shouldn't we expect at least 1 conversion out of over 3000 clicks? That is what was shocking to me.

Normally I might even say that the offer needed work, but the fact that not even one coupon appeared in our stores for Google Local was really surprising.

I would have anticipated that those visiting Google Local were closer to purchase than those in Google.com because they are looking for a local store. But to find that was possibly not the case was what is interesting to me.

Again, it just shows me where to focus my efforts. There's lots of effort in preparing for Google Local -- getting the store location entered, waiting for it to be added, then checking back so that when it IS up you can create ads, etc. Plus, being a retailer, we change out our specials often. Is it worth my time to change all those ads, including Local? Maybe not for the Local stuff.

Diva
Online Survey Diva is offline   Reply With Quote
Old 06-22-2006   #15
tailfins
 
Posts: n/a
Diva - did you have 3,000 clicks or 3,000 impressions from your Local campaign? Above you said 3,000 impressions, but your last response indicated 3,000 clicks.
  Reply With Quote
Old 07-18-2006   #16
earlpearl
earlpearl
 
Join Date: Feb 2005
Location: Washington DC region
Posts: 72
earlpearl is on a distinguished road
Diva:

I've been away from here for a long time and found your study interesting.

I've been working on finding what gets higher rankings at G Local (G maps). My business tends to outrank its distance from a location

I PM'd you and referenced a discussion about this at seorefugee: http://www.seorefugee.com/forums/showthread.php?t=1963

that you may want to look at. Incidentally, the Bill in the thread is SEW's very own Bill Slawski who publishes on SEW's blog.

Dave
earlpearl is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off