Does SEM work? Not according to this stinging indictment.

[Eley Duke]

So, who's tested this out? I'd be interested in hearing from dealers and what results they found if they reduced their SEM and saw any change, or if they put it to other forms of online, such as a market place, and saw any change. Our SEM was turned off for 8 days in the end of Nov and we saw little traffic change.

If Dataium did a study on 20 million car choppers per month, I think that's a significant sample size. The info from Cars is impressive, if only 6% of dealer web site traffic is coming from SEM, and only 1% result in a lead, we might be wise to rethink and test this. I am in a extremely competitive market for Chevy, maybe I am wasting my new car digital money when I could be putting it somewhere with a better ROI.

I'd like to hear some dealers views on this and what they are experienced.

 

[ddavis]

Eley, my closest competitor, in Dallas, stopped all of his display advertising and put all of his money in SEM. He was spending 50% more than my entire internet budget. He had this idea that he could drive all of the traffic to his website and not be competitive with his pricing. His business went into the tank and he is no longer the GM of that store. I know there were other issues, but this didn't help.

 

[csabatka1]

So, who's tested this out? I'd be interested in hearing from dealers and what results they found if they reduced their SEM and saw any change, or if they put it to other forms of online, such as a market place, and saw any change. Our SEM was turned off for 8 days in the end of Nov and we saw little traffic change.

If Dataium did a study on 20 million car choppers per month, I think that's a significant sample size. The info from Cars is impressive, if only 6% of dealer web site traffic is coming from SEM, and only 1% result in a lead, we might be wise to rethink and test this. I am in a extremely competitive market for Chevy, maybe I am wasting my new car digital money when I could be putting it somewhere with a better ROI.

I'd like to hear some dealers views on this and what they are experienced.

In Aug-mid Jan I ran a decent manual SEM campaign. 3 markets, St. Joseph, MO, Lincoln, NE, and Grand Island, NE. We have Ford, Lincoln at all locations, Kia at St. Joe and GI, and Mazda and Pre-owned store in Lincoln. I set up campaigns for new and pre-owned in each market.

What I found was:
Lincoln market (our biggest) was cheapest PPC and users searched more new vehicle search terms and very generic (new ford dealership, new f-150, etc.)
St. Joseph market was searching more pre-owned and very specific sets of keywords (ex. used Ford F-150 for sale in St. Joe)
Grand Island market was nearly 1.5x cost PPC over Lincoln and users search a mixed of new and pre-owned.

Conclusions From Analytics:
Depends on you unique market if SEM approach is best. Our markets showed different levels results and I cannot track exact sales or leads on SEM traffic, I can only do a comparison traffic to sold or website lead chart and website solds and leads trended up with the the search engine increased traffic. Website conversion rate and website solds remained consistent (within 2% of previous month). With that data I can say or assume sales increased based on increased traffic, due to SEM - to what degree, it is not measurable.

Looking at the Total Search Traffic column you can say that it appears the traffic remained somewhat consistent with a small bump in July/Aug. But if you take out monthly trends and compare it previous year, we saw substantial increase in total search engine traffic over the previous year. SEM took away from some Organic visitors, but still show increased in total shoppers over previous year.

From what we experienced we decided to go to an all or nothing managed SEM approach with Haystak in March as an AutoTrader replacement. This will be able to give us better numbers and stats of if a SEM approach works for us. Does this really work? We will see within a few months. We didn't stop or tv, direct mail, or other marketing avenues, just took AutoTrader out of equation and inserted paid SEM. All-in approach is never good, we still wanted a good mix of marketing avenues.



Analytics (we increased budget in late Oct):

Month	Paid Visitor Total	Organic Search Traffic	Total Search Traffic	Search Traffic Increase Over Previous Year
May	0	9,300	9,300	2,661
June	0	11,864	11,864	4,913
July	0	13,189	13,189	4,924
Aug (28 days paid)	1,158	12,868	14,026	6,246
Sept	1,253	10,560	11,813	6,801
Oct	1,723	10,951	12,674	6,400
Nov	3,369	8,362	11,731	6,032
Dec	4,391	7,991	12,382	6,577
Jan (12 days paid)	2,600	9,761	12,361	4,352
Feb	0	10,079	10,079	3,587

 

[jscole86]

So, who's tested this out? I'd be interested in hearing from dealers and what results they found if they reduced their SEM and saw any change, or if they put it to other forms of online, such as a market place, and saw any change. Our SEM was turned off for 8 days in the end of Nov and we saw little traffic change.

If Dataium did a study on 20 million car choppers per month, I think that's a significant sample size. The info from Cars is impressive, if only 6% of dealer web site traffic is coming from SEM, and only 1% result in a lead, we might be wise to rethink and test this. I am in a extremely competitive market for Chevy, maybe I am wasting my new car digital money when I could be putting it somewhere with a better ROI.

I'd like to hear some dealers views on this and what they are experienced.

This past Tuesday, I sat in on a webinar that Brian Pasch did that was called "Becoming VDP Factories". We've known for quite a while that Vehicle Detail Page views are one of the most important metrics to measure for our 3rd party partners like Autotrader and Cars.com, and there is a direct correlation to VDP views and sales (assuming the sales process is good).

Next to a phone call / email / chat, the most important thing you want from your website is to maximize the number of vehicle detail pages that are looked at. What I learned from this webinar is how to look at the % of visits that resulted in a vehicle detail page view, and look at it from each source, including your SEM spend. I assume it's different for each dealership, but if you spend $1,000 on SEM and get 500 visitors, what percentage of them viewed a VDP- 30%, 60%? What was your cost to have that visitor view the VDP aka cost per VDP?

IMO this is how we need to start evaluating our SEM spend. Anyone else sit in on this webinar?

 

[ddavis]

In Aug-mid Jan I ran a decent manual SEM campaign. 3 markets, St. Joseph, MO, Lincoln, NE, and Grand Island, NE. We have Ford, Lincoln at all locations, Kia at St. Joe and GI, and Mazda and Pre-owned store in Lincoln. I set up campaigns for new and pre-owned in each market.

What I found was:
Lincoln market (our biggest) was cheapest PPC and users searched more new vehicle search terms and very generic (new ford dealership, new f-150, etc.)
St. Joseph market was searching more pre-owned and very specific sets of keywords (ex. used Ford F-150 for sale in St. Joe)
Grand Island market was nearly 1.5x cost PPC over Lincoln and users search a mixed of new and pre-owned.

Conclusions From Analytics:
Depends on you unique market if SEM approach is best. Our markets showed different levels results and I cannot track exact sales or leads on SEM traffic, I can only do a comparison traffic to sold or website lead chart and website solds and leads trended up with the the search engine increased traffic. Website conversion rate and website solds remained consistent (within 2% of previous month). With that data I can say or assume sales increased based on increased traffic, due to SEM - to what degree, it is not measurable.

Looking at the Total Search Traffic column you can say that it appears the traffic remained somewhat consistent with a small bump in July/Aug. But if you take out monthly trends and compare it previous year, we saw substantial increase in total search engine traffic over the previous year. SEM took away from some Organic visitors, but still show increased in total shoppers over previous year.

From what we experienced we decided to go to an all or nothing managed SEM approach with Haystak in March as an AutoTrader replacement. This will be able to give us better numbers and stats of if a SEM approach works for us. Does this really work? We will see within a few months. We didn't stop or tv, direct mail, or other marketing avenues, just took AutoTrader out of equation and inserted paid SEM. All-in approach is never good, we still wanted a good mix of marketing avenues.



Analytics (we increased budget in late Oct):

Month	Paid Visitor Total	Organic Search Traffic	Total Search Traffic	Search Traffic Increase Over Previous Year
May	0	9,300	9,300	2,661
June	0	11,864	11,864	4,913
July	0	13,189	13,189	4,924
Aug (28 days paid)	1,158	12,868	14,026	6,246
Sept	1,253	10,560	11,813	6,801
Oct	1,723	10,951	12,674	6,400
Nov	3,369	8,362	11,731	6,032
Dec	4,391	7,991	12,382	6,577
Jan (12 days paid)	2,600	9,761	12,361	4,352
Feb	0	10,079	10,079	3,587

I would be interested in how this effected your lead count. According to the Dataium study, traffic coming off of cars.com (assume it would be the same for AutoTrader) are much more likely to submit a lead.
What is your internet budget compared to the total advertising budget?

 

[joe.pistell]

Wow, Joe! So many obscure and unrelated references. Big Data. Landscaping Businesses. I'm assuming that you have started smoking SOMETHING since the move to Vermont. I look forward to your insights AFTER you read the article.

Geez Ed, the FIRST PARAGRAPH in the study, the CONTEXT is established:

"We find that new and infrequent users are positively influenced by ads
but that existing loyal users whose purchasing behavior is not influenced
by paid search account for most of the advertising expenses, resulting in average
returns that are negative. We discuss substitution to other channels and implications for
advertising decisions in large firms."

​

http://conference.nber.org/confer/2013/EoDs13/Tadelis.pdf


Ed,
This weed smokin' tree huggin Vermonter is a master at marketing. Marketing is strategy. It's a chess match. EVERYTHING in chess is context. World class marketing types are skilled at spinning data to create pretext*.

SUMMARY: Without reading it, looking at the business profiles alone, I knew moving the ebay evaluation was un-related to our industry (parts and service traffic not incl'd).

Now... Off to Stowe!
Grab a snow board, my growler of VT craft beer and where's that bag of organic weed?


Ed, Meet you at the base of Stowe for a brew at the http://www.matterhornbar.com/



*pretext[ˈpriːtɛkst]n1. a fictitious reason given in order to conceal the real one

http://www.thefreedictionary.com/pretext

 

[ed.brooks]

Geez Ed, the FIRST PARAGRAPH in the study, the CONTEXT is established:

"We find that new and infrequent users are positively influenced by ads
but that existing loyal users whose purchasing behavior is not influenced
by paid search account for most of the advertising expenses, resulting in average
returns that are negative. We discuss substitution to other channels and implications for
advertising decisions in large firms."

​

http://conference.nber.org/confer/2013/EoDs13/Tadelis.pdf

SUMMARY: Without reading it, looking at the business profiles alone, I knew moving the ebay evaluation was un-related to our industry (parts and service traffic not incl'd).

Now... Off to Stowe!
Grab a snow board, my growler of VT craft beer and where's that bag of organic weed?
View attachment 1557

Ed, Meet you at the base of Stowe for a brew at the Stowe Vermont Matterhorn Bar and Sushi Restaurant > Home



*pretext[ˈpriːtɛkst]n1. a fictitious reason given in order to conceal the real one

pretext - definition of pretext by the Free Online Dictionary, Thesaurus and Encyclopedia.

If I read this correctly, your PRETEXT for not getting getting involved in a national discussion on revolving around the CONCEPT of whether branded-keyword SEM does nothing more than cannibalize your own SEO results seems to be the CONCEIT* that the automotive vertical is completely different from, and unrelated to, the larger marketing community.

This weed smokin' tree huggin Vermonter is a master at marketing. Marketing is strategy. It's a chess match. EVERYTHING in chess is context. World class marketing types are skilled at spinning data to create pretext*.

I don't dispute your position as the Master of Marketing - I even voted for you to be Pope of Interactive at last week's conclave. I thought you had more knowledge than Francis. Enjoy the High-Quality Vermont Hydroponics, dude.

*conceit [kən-ˈsēt]n (1) : a result of mental activity : thought (2) : individual opinion

 

[joe.pistell]

If I read this correctly, your PRETEXT for not getting getting involved in a national discussion on revolving around the CONCEPT of whether branded-keyword SEM does nothing more than cannibalize your own SEO results seems to be the CONCEIT* that the automotive vertical is completely different from, and unrelated to, the larger marketing community.


I don't dispute your position as the Master of Marketing - I even voted for you to be Pope of Interactive at last week's conclave. I thought you had more knowledge than Francis. Enjoy the High-Quality Vermont Hydroponics, dude.

*conceit [kən-ˈsēt]n (1) : a result of mental activity : thought (2) : individual opinion

haha, ok, all wordy hosersh*t aside, heres my $0.02

The study's abstract clearly states

1. 
   [*=1|left]"...We find that new and infrequent users are positively influenced by ads"
   [*=1|left]"...but that existing loyal users whose purchasing behavior is not influenced by paid search account for most of the advertising expenses, resulting in average returns that are negative."
   
   

Ebays business is not our business, Car dealers and ebay have little incommon.

​

1. 
   [*=left]Ebay has a few million products for sale. You can buy ski passes, Shoe laces, car manuals, a TV, carpets, dry goods, real estate, antique cars, a vacation, etc... eBay loyalists know that ebay is it's own marketplace with no equal in the world.... its a destination all unto itself.
   [*=left] Car dealers website shoppers come by once every 3-6 years & they only want to look at cars.

The Shopper dynamics could not be further apart. Applying lessons learned from ebay and applying it to car dealers has the potential to create incorrect conclusions.

​

 

[joe.pistell]

Ed,

Repeat visitors are a great thing for car dealers. From my 20,000+ surveys at delivery, one of the hallmarks of a buyer is multiple visits to the dealers website.

Funny how this metric was the where the low ROI came from. It just goes to show how different the 2 business models are.

 

[csabatka1]

I assume it's different for each dealership, but if you spend $1,000 on SEM and get 500 visitors, what percentage of them viewed a VDP- 30%, 60%? What was your cost to have that visitor view the VDP aka cost per VDP?

IMO this is how we need to start evaluating our SEM spend. Anyone else sit in on this webinar?

Pretty easy to do this within Google Analytics. The link below gives you the steps needed to do this sort of tracking within Google Analytics. Set up your Google Adwords or Bing Ads to use Campaign URL tracking. Then within your reporting in Google Analytics you can track PPC visitors and view the content/pages they visited by using the Advanced Segment filter in the reports. Looking at the Content Report you can search for a portion of your VDP URL and view visitors of VDP pages only based off your PPC campaigns.

Google Analytics PPC Conversion Tracking