Google Places Search – Before and After Dashboard Analytics Comparison

specific listing changes in actions for those business showing increasesGoogle Places search was released on October 27 with a great deal of fanfare and commentary. It has been available for over a month and the Places Dashboard analytics data can now provide some insight into the impact of the new display.

I have assembled an overview of the impressions and actions based of the Dashboard analytics for 45 listings for the 4 weeks prior to the rollout and the 4 weeks after the rollout.

I recognize that these numbers are, by design, superficial and are on occasion buggy (showing fewer impressions than actions). We also don’t know if the analytics kept up with the main search results page. However in aggregate, they do provide some general insights about the switch. The business listings represent a range of industries, locales and business size so the sample should represent a reasonable cross section of reality.

Overall
Change in Aggregate Actions Up 28%
Change in Aggregate Impressions Down 25%
Action Impression Ration
Action/Impression Ratio Before 7.17%
Action/Impression Ratio After 12.26%
Businesses with Increases
Number of Listings with Increased Actions 20
Average % increase in Actions for Businesses
with Increases
328%
Average % increase in Impressions for Businesses
with Increases
7%
Businesses with Decreases
Number of Listings with Decreased Actions 22
Average % decrease in Actions Business
with Decreases
-55%
Average % decrease in Impressions for Businesses
with Decreases
-23%
Number of Listing with No Change in Actions 3

Discussion

The number of listings with increases is roughly equal to those with decreased actions although the average increase was significantly higher than the average loss. I have not done a granular enough analysis to fulling understand but it appears that strong (ie better ranking websites) got stronger and the weak lost visibility.

In one particular case, it is obvious that a business in the burbs that was doing well organically but not locally on the major city searches has been rewarded with a blended placement on the major geo phrase. The bears further study but gave me the distinct impression that the geo area that defines a given search has in some cases been expanded and that businesses that were previously outside of that range now have a shot at first page blended placement.

It is interesting that the actions increased while the impressions were down. The average of actions to impressions increased from 7.17% to 12.26%. This would imply that the UI changes with Places search significantly increases Places related actions (directions, website clicks, more info clicks) and is more efficient at leading to actions that Google wants. This makes sense in that previously some local search traffic was “lost” to organic results. These results obviously reflect the new blended reality.

One area that would be of use to explore would be to examine the relative changes in specific behaviors within the Actions and compare that to changes in website traffic overall. It would be interesting to know if Google or the business is capturing more of the activity.

As in many changes, there were winners and losers. But in the end if more people are finding what they want with fewer searches then those businesses that understand the rules, the users AND Google will all benefit.

Please consider leaving a comment as your input will help me (& everyone else) better understand and learn about local.
Google Places Search - Before and After Dashboard Analytics Comparison by

11 thoughts on “Google Places Search – Before and After Dashboard Analytics Comparison”

  1. Nice, Mike: I love deep analytics. I find them to be critical.

    Where I found deeply interesting information was in the use of Google Analytics. I took the accounts and looked at the geographical overlays. Then I dug deep into country/region/ and finally town. That is where the deep insights began to emerge.

    Upon running info by towns and/or by keywords I then went deeper to find keyword traffic by town. I looked at that for all keywords and I also looked at that for ppc campaigns.

    Additionally I searched for the critical conversion/keyword phrases in google.com; adjusting the location of the IP via the left hand field that allows for searching from different locations.

    The insights seem to confirm the actual visits with the visual rankings seen by searching via different towns near the subject businesses.

    Some of the businesses seem to be clear winners as a result of the merged rankings of organic/local and some are losers. The businesses that are winners are showing the following:

    A) Clearly ranking higher for critical keyword terms visually when searching on those terms and adjusting for different town locations
    B) Less obviously, because the sample size is too small to make strong judgements, there are a higher number of visits/day for critical keyword/conversion phrases from towns where rankings improved.

    In all cases, the businesses I checked had merged local/organic records with organic info, places page link, # of reviews, etc. The powerful presentation in UI seemed to assist in encouraging more visits.

    I haven’t yet analyzed the information within the dashboard. Thanks for your work. The fact that there were more “actions” seems to indicate more visitor interaction with the rich merged record information in the UI.

    As the actions increase, which segments of the “actions” increase the most; visits to the website, directions, or visits to the places page? Have you taken a look at that?

    Also as actions increase are you looking at overall traffic to see that kind of impact?

    I’m analyzing and acting hard based on the changes in the Organic Google presentation. Its a significantly different phenomena.

  2. Mike: 2nd quick set of insights. After reading the above piece and “mouthing off” I quickly looked at a couple of sets of dashboard statistics as you did above. I compared 35 day periods directly before and after the google.com change. Upon review I realized I hadn’t reviewed dashboard stats since the conversion. (oh my) A couple of quick observations:

    1. One account hasn’t had any impressions since 11/12…but it still shows a nice number of actions. Weird huh? Sort of par for the course….my experience is that Google does not do a great job in delivering dashboard stats. What is your experience?

    OTOH: when comparing year to year stats in adwords and analytics I see some very weird numbers that don’t make sense and don’t seem to represent real traffic. Oh well. Can’t totally depend on this information is the lesson I have learned over time.

    2. On the one business that I know has been negatively impacted by the change in appearance….some scary information when comparing 35 day periods before and after:

    A) Total impressions after the change in UI were less than 1/2 of the total impressions in the immediate 35 day period before the change.

    YIKES. That is scary.

    Dug a little deeper and looked at some phrases per dashboard stats. One key phrase went from over 500 impressions to less than 70 in the two different periods. YIKES….scary. Dashboard, in that case, includes geographical information (heh heh…a bit of a flaw– ;) ). Dashboard impressions were a teeny percentage of impressions via adwords. Addtionally analytics showed more visits for the phrase via both organic and ppc clicks.

    Ah well….in at least those two cases dashboard stats left a lot to be desired.

    I tend to try and review a lot of different sources for statistics before making hard judgements. The various forms of analytics need to be confirmed by a variety of sources before I make judgements.

    Nevertheless, thanks for your insights and analyses on the above. 45 businesses is a far healthier sample than the 2 I just glanced at. I suspect that with 45 businesses it tends to level out the weak reporting in any one or several accounts.

    I’ll report more on my visits into the dashboard later. Can’t wait to hear what others say on this post 10/27 change in Google.com visibility.

  3. Mike:

    I hate to be the only person commenting here. In fact I believe this is especially rich territory in a very general sense: analytics and activity on smb sites since the Google.com UI conversion to a blend of organic/local results.

    What is a more rich topic? If your interested in high rankings and how your site is doing one better be looking at various forms of website analytics.

    …and with that…one more quick caveat about Google’s dashboard stats:

    I had reason to quickly glance at the dashboard stats for another acct. The acct has a tag.

    Oops No Tag data None at all. There were two interesting totally unrelated elements connected with the account:

    A) Google shows billing info connected with the tag.
    B) Besides not showing tag information there is a Google promo suggesting we get a tag for the account.

    OKAY GOOGLE….you are charging us for the tag. We are getting ZERO FEEDBACK In fact you are urging us to buy a tag

    I really have little faith in the Dashboard. Look there are all sorts of what seem to be relatively small problems with Adwords. We spend a lot on it. If you are going to charge us for TAGs…and you are going to want us to try Boost. At least create the semblance that you know what you are doing with our money.

  4. Okay I’ll comment here too. You put a lot of work into this. You deserve more comments! I’ve seen some very weird results in the dashboard analytics – certainly enough to make me trust them less. The only metric I really trust is the feedback I hear from small business owners about conversions. Are they getting new bookings, new customers? Is there phone ringing?

    Mike, could you tell if conversions have increased for the 45 listings? Do these improvements in analytics translate to Places being an improved product for these business owners? Of course seasonality and local economies play a role for each business, but what are your thoughts on an actual increase in conversions? Thanks.

  5. @earlpearl

    I don’t think the Places Dashboard Analytics are even worth looking at right now. For us, the Impressions line

    went flat in early November while Actions are still showing, lot’s of Actions.

    Our account is Active and we rank very well in Places Search. I’ve tested impressions by searching from

    different locations and Google is delivering the impressions. Because we’re in Hawaii and most of our business

    comes from California, I’ve been checking the California results very closely.

    The New Places Search is still evolving and we get many different views when Search Results are returned. For

    example, on one of our top keywords I got the following results — all in the same week which was last week! –

    1) 3rd in a 7 Pack which is sort of a hybrid betwen the old Onebox and New Places Search

    2) A competitor showed as the only listing in a Onebox

    3) 1st in the New Places Search view

    I feel like I’m on Let’s Make a Deal. What do you suppose will be behind Curtain One, Two or Three next week,

    LOL!

    I think all of the flux in the system is causing Conversions to drop. In Advertising and Sales, a consistent

    message is critical. Whenever prospects have questions or are unsure about something, they do nothing. I believe

    all this flux and even the New Places Search presentation creates this uncertainty — they wonder what this is

    because it didn’t look like that before. I believe that even PPC CTR and Conversions are affected by it.

    If you’re in one of Mike’s groups for whom Impressions and Actions decreased, it becomes especially difficult

    when you add in poor conversions.

    What do you think?

  6. @Maui: I mostly don’t depend on dashboard stats. IMHO they are a bs form of stats used as a way to rope in “customers” and then big G hopes to convert them into paying customers.

    The dashboard doesn’t tell much. I prefer 2 forms of analytics. Then I can compare them. Also I get better detailed data.

    Hard for me to form lasting impressions on final conversions/sales since the G presentation change on 10/26. We have 2 forms of conversions: contacts is A and what gives us a shot at analyzing the web effectiveness.

    B is sales. Since our sales involve a sales team in all cases we get different results….and it really depends on the sales force.

    We have smbs that are “winners” in the visibility race since the Google UI change and “losers” with less visibility. The losers saw a reduction in leads. We are working on that. I think we have some ranking improvement and improved visibility for one of them as of today (at least for the time being).

    Our contact conversions are roughly 1/2 as of a reported 1 time visit (not necessarily so as visitors can be using a work and home computer) and 1/2 with more than 1 visit. (guess that means that more than 1/2 visit more than once before they convert/contact).

    That situation you described with 3 different looks for a keyword search is frustrating. My best experience in all of that is to simply always be first. (LOL–so much easier to say than accomplish)

    My ppc results and ctr have been pretty consistent since before and after the Google UI change. Haven’t noticed any changes there.

    Thanks for commenting.

    Dave

  7. I checked Google Places Analytics today and the analysis of Search Queries displays once again. Impressions and Actions still make no sense but, based on the following, it looks like that will be coming soon.

    I was pleasantly surprised to see that there are new layers of analysis. The whole design has changed. The Dashboard and Analytics are merged and there is new analysis related to the source of Impressions and Actions. The Actions analysis is much more detailed.

    You will now be able to tell whether Impressions came from Google and Google Maps or Google Maps Mobile.

    Actions are broken down as “On Google & Google Maps” and “On Google Maps Mobile”. Google & Google Maps is broken down into three categories — Website clicks, Drving directions, and Views. Google Maps Mobile is broken down into eight categories — Place Page, Place Bubble (GMM only), Reservation (GMM only), Directions, Call (GMM only), Add as contact (GMM only), Street View (GMM only) and Website.

    They continue to provide “Terms used in searches for your business” and they’ve added “Locations used in searches for your business”.

    Wow! I just noticed that there are drop down menus that allow you to sort out specific groupings of Impressions and Actions. For Impressions your options are — All impressions, Impressions on Google.com and Google Maps, or Impressions on Google Maps Mobile (GMM data is for selected phones).

    For Actions there are two drop down menus you can select from that will give you an amazing level of granularity. The first menu provides the following options — All sources, Google and Google Maps, and Google Maps Mobile (GMM data is for selected phones). The second drop down menu lets you select from the following — All sources, Directions, Place Page, Add as contact (GMM only), Call (GMM only), Place Bubble (GMM only), Place Page, Reservation (GMM only), Street View (GMM only), and Website.

    From the above, it’s pretty obvious that G is serious about going mobile. They are also very serious about capturing the millions of businesses that do not currently advertise with Google and this is a way to do it along with new advertising products like Google Boost that simplify the process for small businesses that normally have limited resources to learn the ropes for complicated forms of advertising like AdWords PPC.

    If that’s not proof enough, look to Google senior personnel moves and you’ll have all the info you need to see the future. As Aaron Wall points out in his SEO Book Blog, Marissa Meyer is changing positions and it is a big tip of the hand as to G’s intentions. He points out, “Marissa has been responsible for much of what you see in Google in terms of how it is organized. Marissa has just moved to head of Geographic and Location Services. Google Earth. Google Maps. Google Local. Google Street View. Mobile location data and targeting. Expect more data to be organized around locality.” —> SEOs Should Focus On Where Google Is Heading

    Nice work G! So that’s what you guys have been doing. Keep going, this is getting good. Oh, but G, while I have your attention, can you please make the New Places Search Results work too, LOL?!!!

    Whoaaaa!!! I may be living in Bizarro World . I took a couple of hours off, did my bike ride for the day and came back to check this for accuracy. When I opened up Google Places to check out the New Places Analytics, poof! — not there. Back to the old, tired, non-working version….. Maybe I took a nap and was dreaming? I wish I would have taken screen shots.

    I suspect that I was exposed to one of those quick New Places Search flashes that have been happening so often. Well, anyway, it’s coming. Ain’t it exciting!

  8. Thanks for writing an article on this, Mike. It is difficult to find much material on the stats side of Maps / Places. The SERP value of having a verified listing seems obvious, but the value of the actual listings and Places pages is vague.

    I had a client listing with more actions than impressions recently. According to one answer I found on Google’s support forums, they count a request to the address of the business as an action for the Google listing itself. So even if the business’s tag and business info never display to the customer, it is still counted as an action. Seems to be a bit of a disconnect there.

    Again, thanks for the post!

  9. Hi Michael
    My thoughts are that their analytics are not very robust and it is nothing but a bug… and their explanation could not be the whole story.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Comments links could be nofollow free.