Is PPC a window to all visitor behavior?

[Matt Murray (blog)]

Niche market? Retail auto is one of the largest markets in the US.

Personas - There are many ways to look at the idea of a search persona. Adapt it to our market and you will understand why the average dealer website does little to address user experience. You have the service customer with their many needs, the finance customer who may have their own unique situation and, of course, the sales customer's wants and needs. Personas are about silos of experience that can be related to the keyword if you look closely enough. Bryan's stuff is good, but as I mentioned above, Vanessa Fox does a beautiful job of simplifying personas for any market in her book.

I don't doubt your conviction Larry. Once again I admire your passion for the industry. The point of this post is to show what we have learned from observing shopper behavior at a very high level. Are there more questions to be answered? Absolutely. Will they be answered in this post? Absolutely not. At the end of the day the content that I share is for the education of the dealer, not to spark theoretical arguments between vendors. If you would like to split hairs with me, let's take it off DealerRefresh.

 

[Alex Snyder (blog)]

I apologize for this tardy response. Thank you Matt for clarifying something that I want to follow-up with too: Yes, I work for Dealer.com, but on DealerRefresh I am speaking as I am on DealerRefresh. I may, on occasion, speak for my employer (as I did with Checkered Flag) maybe 3% of the time. The other 97% of the time it is just me. To suck my employer into something that I did not represent them as is hitting below the belt Larry. This is something you also tried to do when I was at Checkered Flag. By the way, you are the only person to ever do this. With that said, I want you to know I don't appreciate it and hope I've made myself perfectly clear.

Larry - I follow you on twitter, I have two of your blogs coming into my RSS reader, and I keep both eyes on quite a few automotive communities. I believe it is safe to say that I come across the majority of the things you place online. I say this for the 90+% of the folks reading who aren't commenting just so that they understand there is some history here.

I must admit that most of the time I do not agree with Larry when I can understand him. In fact, over an hour long phone call with Larry, roughly 3 months ago, I left the conversation even more confused than when it began. But I have to give Larry credit as his posts & responses are being articulated much better these days.

Larry - I am following you very well in this comment string - thank you!

With that said, and knowing that this is me representing me alone, let's have a real debate and get some things cleared up.

Amazon, eBay, and Edmunds have different agendas than dealership websites. Their conversion is much easier to track. However, I was not citing them as websites for comparison; I was citing them as websites who have used PPC to land customers on deep-linked pages within their main domain.

I must take contention with your labeling of dealership websites as Random Access Websites. I get where you're going with it, but I don't understand the need for the label. All websites, including most microsites, are "random access websites." Tell me different if I'm wrong - I could be getting the definition confused.

You call the automotive industry a niche industry and I don't think anyone can argue that generalization. But you call it that and then try to apply traditional website measurements and conversion rates from true eCommerce (sites with shopping carts) websites to automotive....do you recognize the contradiction you yourself are proposing here? Are our websites part of the niche that needs to be held accountable for things that don't participate in our little niche? I'm confused.

Yes, there are dealership websites doing double digit conversions and quite a few that are doing really strong double digit conversions in PPC. I am willing to agree with you that if we could accurately measure and tie all phone calls, floor ups, service drive appointments, parts counter visits, rental counter visits, and bodyshop estimate appointments then all dealership websites actually convert at HUGE rates - maybe even higher than what you say they should. In reality the only measurement that can accurately be measured is a form submission ("Internet Lead"). So, will you please set the record straight right now and tell us if your 30-40% conversion rate is based on total website performance, with all those things we can't measure, or is it based on the simple form conversion?

 

[LarryBruce]

Matt Automotive is a large market but it is a niche, just a big one.

The post makes some interesting points but it actually creates more questions than it answers.

As I stated in my earlier comments, I will reiterate:

1. Calls are preferred by customers far over a form submission. – Granted, but this is a type of conversion and it’s not news, anyone who has been selling on the web should know this. At my store Mortortrendsonline.com we have been selling completely over the web since 2001. We have never completed an auction on eBay, every eBay listing is a call generator and we get a lot of them. As I said in my earlier comment if you are wanting to convert to high sales ratio’s you should be optimizing conversion for a phone call those will convert to sales much higher than any other type of conversion.

2. Segmentation of New and Used customers was irrelevant because both had to follow the exact same behavior funnel on you sites; it shouldn’t be a shocker that the differences were so small. The page view difference is one of logistics not behavior, because there are less used cars to look at.

3. 12% conversion across the board to some dealers I am sure that sounds great but, to be as you say “realistic” you have to dissect that number by branded vs. non branded terms, then look at sales conversions vs. service conversions and you’ll start to get the real picture as to how to grow you internet commerce and your overall dealership using the web.

4. 18% average on branded terms just proves by statement that a 30% + conversion rate from a main random access website is achievable; because you have to have some in that number that are above the mid 20% range.

5. A 4x conversion rate on a branded paid link vs. an organic link is an interesting stat. With that said I would need to see all the data and the parameters of the test to see if numbers are clean. That number doesn’t make sense to me unless there was something different about the sponsored link vs. the paid link because links don't convert landing do.

hese are my observations about this post. They are not bad and they are not good… they just are.

Really none of them have anything to do with search personas except to maybe say that the call volume is indicative that the site isn’t addressing the search persona so the visitor is just calling, that is a stretch indeed. Here is the best way I can explain my view of search personas. Now I am going to show my more geeky side here, no wise cracks…I can get that anywhere.

Search personas are broken into 2 groups. To illustrate these personas I am going to use characters from Star Trek:

Logical:
Fast Paced / Competitive – Kirk
Slow Paced / Methodical – Spock
Emotional:
Fast Paced / Spontaneous – Scotty
Slow Paced / Humanistic – Bones (Dr. McCoy)

I really can’t see how anything in this post addresses these types of search personas. Ultimately you persona will affect how you interact with a site, what you are looking for when you do, and why you click on what you click on. This is a very high level thing and something you really shouldn’t be looking to address until you site is converting waay above where it is now.

 

[LarryBruce]

Ok guys after this I have call the time limit rule on this post. We seem to be beating a dead horse and we are so far off the subject it unreal.

Alex I’m not trying to suck anyone into anything, again you should know me better than that. I say what is on my mind some can follow it some people can’t… it is what it is. To be clear it wasn’t a shot just what I was thinking.

Alex you have been in business long enough to know your comment and actions always reflect on your employer. It was not a stretch for anyone to assume your comments that my conversion benchmark is causing problems and that I should back off would come from interaction with Dealer.com clients, come on guys…really?

I appreciate you following me on twitter as I do you and reading my blog posts even when they don't make sense to you sometimes, but Alex this is a conversation and it’s not about winning anything.

I believe an off page strategy is better for car dealers than a deep link strategy it’s just that simple. You brought up 3 companies and asked why they don't use landing pages and I answered your question to the best of my ability.

Yes all main websites are “random access” you may not like the terminology but that doesn’t change the fact that random visitors wonder in to your main website and that you can randomly go anywhere in your main website, you just have to live with that and it is one of the main reasons the most successful companies in PPC use off site microsites and landing pages.

Alex traditional web measurements? I don't know what you mean by that. There are measurements for shopping cart conversions and for lead generation. The definitive site for this is fireclick go to My Social Following - Buy Real Social Advertising for Instagram, Youtube, Twitter X, Tiktok, Facebook, and more and you will see measurements across a great many online conversions shopping cart and otherwise, what you won’t see is anything close to my 30% number, because they only track online conversion. So no Alex I don't see the contradiction, my numbers come from what I have experienced in my own store on site and off.

A conversion = the following:

Any form submission, phone call, email or chat session where good re-contact information is captured for the dealership to follow-up with the customer, and I will stand behind that a dealership should be able to get 30% plus like this from its online marketing properties and you can absolutely track every one, we do that in PPC right now.

I hope Dealer.com is not just tracking internet form submissions? Not to drag your employer into it again but I assume you know how and what they track, so I have to ask when you make a statement like “In reality the only measurement that can accurately be measured is a form submission (“Internet Lead”).” That is absolutely not right.

Again guys we can take this offline on a call if you like or at NADA or some other meeting where we may all be but I can’t put anymore time into this post, sorry.

Kevin I hope I was helpful to you as you were originally the one wanting my comments. If there is anything you want to discuss offline around what I have said here or that I can help you with feel free to email or call me.

If you don't have my number see below:

281.455.3811

[email protected]

 

[Ed Brooks]

Wow – The best laugh I’ve had all week! I don’t mean the passionate, well thought out arguments being made above. I read down to Alex saying, “We bring Larry in here and we’re going to have one very long comment string – lol.” And then I scrolled down. And scrolled down. And then scrolled down some more. I don’t care who you are, that’s funny.

I have to commend Matt for giving a high level introduction of the concept of personas to dealer marketers. While this conversation devolved into arguments about definitions, about Bryan & Jeffery Eisnenberg verses Vanessa Fox, landing pages verses “Random Access Websites”, the fact remains that Matt did an exemplary job of giving a high-level overview of the concept.

Not that I’m eager to add gasoline to a string already wandering off-topic, I did want to address a few things:

“…of all the examples you have given Amazon is the closest to our business” - There is a huge difference between Amazon and the car business – an even bigger difference than the difference between selling a $20 item and a $20,000 item. That difference being that Amazon MUST convert online because there is not a brick and mortar alternative. Every customer visiting Amazon knows the only way to make a purchase is online. Conversely, every customer visiting a dealer site knows they have the option of showing up at the dealer. That “showing up” option is made even more attractive by the size of the purchase (who would buy a house by filling out a web form?).

My main takeaway – The higher the sense of urgency, the higher the percentage of calls verses web forms. The closer the customer is the sale, the further down the funnel they are, the more likely they are to pick up the phone. Larry makes the point that phone calls convert to a sale better than form submissions. He further makes the point that he makes he pushes a call over a web submission for this reason. I think there may be a real cause and effect problem here. My guess is customers aren’t closer to the sale because they phone, they phone because they are closer to the sale.

Larry states, “Given a niche market and low traffic volumes I think any focus on personas is stepping over dollars to pick up dimes.” I would certainly disagree – In the interview with Larry links to with Bryan Eisenberg, Bryan discusses putting links for the fast-paced Competitive and Spontaneous users high on the page and links for the slow-paced Humanistics and Methodicals lower. The fast-paced users need to get to their content NOW and the slow-paced will read through much more. A great example is Autotrader.com - calm down Larry : The big “Find Cars” button and form is highlighted and at the top while the “Car Research” and “Reviews and Top Picks” tabs are smaller and lower. The Competives get the instant gratification they crave, while the Methodicals get the full and complete information they need to make a decision. This is an example of how to order information for folks on different buying tracks.

As for whether we are a “niche”, I’d propose we use the term “vertical” as it has the connotation of “focused” without the connotation of “small”.’

“YOU NEVER, EVER WANT TO SEND PAID TRAFFIC BACK YOU YOUR RANDOM ACCESS MAIN WEBSITE…EVER!” – Larry, you say “never, ever” but do you mean on a dealer name link as well as a longer tail phrase? The reason I ask is because if we bought “Billy Bob’s Subaru” and generated a click, we know nothing more about this customer then that they are interested in us: We don’t want to send them to a 2011 Forrester microsite. We can’t tell if they want new, used, or service. What makes sense is to send them to our “Random Access Website”. It also makes sense to track them via cookie. To use Amazon as an example, when I Googled “Amazon” moments ago, they came up first in the organic listings as well as having an AdWords placement on top. Clicking the SEM link took me to the “Random Access Website”. I’m sure I was cookied and tracked, but I was directed to the only logical place; Amazon.com

Again Matt – Great post!

 

[LarryBruce]

Ed I said it was the closest to our business not exactly like our business! No Ed I would not buy a house by filling out a web form, there are few that would, but some have. I would however make and appointment via web form to see a house, I would call and agent to get more information from a website that is conversion via lead gen vs. conversion via purchase / ecommerce. The principals that make someone convert to buy or to engage via lead gen are the same.

The point was Ed that out of Edmunds that sells information, eBay that is an auction and Amazon that is an online retail store, Amazon is the closest of those 3 businesses to a car dealership. The question Alex had is why don't these companies use off site landing pages and I gave my opinion on Edmunds and eBay. Amazon (the closest to our business) actually does use off site landing pages and microsites, very successfully I might add.

“My guess is customers aren’t closer to the sale because they phone, they phone because they are closer to the sale.” Actually Ed your guess is wrong. The customers are just as close to a sale web form or call we have done tests on both and found virtually no difference in the time frame to buy form vs. phone. However we did find that if the visitor calls the dealership, they are 7 times more likely to buy from that dealership than if they just submit a web form through the dealerships website. So the reason we push the call is to get a higher sale ratio for our client.

That’s the problem with guessing and one reasons we don't do it.

Ed stepping over dollars to pick up dimes doesn’t mean, that if you focus on the dollars first you don't go back and get the dimes. My point was that you should take the low hanging fruit first and there is plenty of it there for the taking. Personas are complicated high level thing and not something you do until you have a well established testing methodology and culture.

“In the interview with Larry links to with Bryan Eisenberg, Bryan discusses putting links for the fast-paced Competitive and Spontaneous users high on the page and links for the slow-paced Humanistics and Methodicals lower. The fast-paced users need to get to their content NOW and the slow-paced will read through much more. A great example is Autotrader.com” Ed this will be a great case study, can you get with the AT web marketing team and get the test data and methodology used to determine the link placement for competitive’s and mythodicals. I would think that would be good data for AT to share.

Ed, Niche, vertical whatever!

Ed I am not a big fan of buying the dealerships name when it comes to search marketing so I generally stay away from branded terms. There are rare occasions where branded terms do make sense and even in that case It still doesn’t make sense to send them to a “Random Access” website because by doing so you miss the opportunity to understand what driving visitor intent behind those terms and you make the respondent navigate your maze of information.

Anymore? Might just a well keep going now.

 

[Ed Brooks]

Larry,

Since you don't argue with any of the data showing that dealership web visitors are much more likely to phone than fill out a web form (of those that do anything) and your own data shows that those that phone are 7 times more likely to buy, it's not hard to assume (and I know you hate assumptions)that those that don't phone are furthest from the sale. Those that do submit a web form are a minority and I would submit to you they are the folks that want to stay in an "introspective persona" longer in the process.

I can understand the problem you have with this: from our previous discussions, I know you have a very hard time believing that some people (and I believe it is a majority), just show up without a preliminary email or call, without a "web conversion" for you to track. I'd further submit that while a form submission is "Good" and a phone call is "Better", then a walk-in is "Best" (to use the old Sears Catalog terminology). I believe that a walk-in is the absolute closest-to-the-buy, furthest down the funnel customer. They have switched to full-extroverted mode and walked into a dealership. I know you feel this is all "internet as research" crap. But I think it goes far beyond research: the website helped them to a decision.

And the when the decision is made, because they have the option of visiting a brick and mortar location, many do just that. Their personality is such that they don't want to trade emails or have a phone conversation, they want to sit down and buy a car.

It's precisely this difference that means Amazon, Edmunds and eBay are simply NOT analogous to a car dealership. None of them. BestBuy, Staples, etc. are much closer because they have an established brick and mortar component, but we'd still have to deal with the $20 item verses $20,000 item problem.

As for ATC or any other huge consumer site sharing their testing data with you, well, I wouldn't hold your breath. I know during the last major redesign of ATC extensive testing occurred, as you'd expect.

But any dealer site could implement the principles laid out by Alan Cooper, Jakob Nielsen or Bryan Eisenberg of making the information that the impatient need high and prominent while putting the information needed by the more methodical lower. Even if they don't test, the underlying concept has been tested extensively. I know not testing is a problem for you, but you have to admit that any site would perform better following these time-tested principles.

Lastly, I always enjoy our discussions Larry. Perhaps it's because my Myers-Briggs Type Indicator is ENTJ or on a Keirsey Temperament Sorter I'm a Fieldmarshall or on a DISC assessment I'm a very high D (with almost no C at all). In other words, I'm a Kirk

 

[LarryBruce]

"Those that do submit a web form are a minority and I would submit to you they are the folks that want to stay in an "introspective persona" longer in the process.” Ed 6 months ago I would have agreed with you but our test data shows there is only 1 day difference in the buy cycle those on the phone vs. those on the web.

What was even more interesting is that 83% of the phone in customers when asked gave their email address to the salesperson. This data leads me to believe that customer just prefer the phone interaction to the web form nothing more than that.

“I can understand the problem you have with this: from our previous discussions, I know you have a very hard time believing that some people (and I believe it is a majority), just show up without a preliminary email or call, without a "web conversion" for you to track.” Actually Ed quite the opposite I believe more are showing up without any previous engagement with the store they end up at. Actually that is my main issue. I don't believe people are just showing up because they want to, I think they are because they are so frustrated with the online process they are just taking what little data they can from it and going to the local store. The dealers that work hard to give customer easy access to the data they want, give the visitor a reason to engage and make easy to engage prior to the visit are getting the higher conversion rates in the numbers that Matt is showing in this post.

I also believe that if you aren’t converting well you are getting as many walk-ins from your web properties either. These things are connected.

Ed I think you mistake my questioning of the data. I absolutely believe that a walk-in is the closest to a sale; anyone would be a fool to think otherwise. But I also subscribe to the theory “Hope is not a strategy” with that said I want my web properties to convert as much as possible so I have the optimum chance at getting the most customers I can. I am not going to just throw up a site and not work to find ways to make it convert higher, just crossing my fingers and hoping that people will just walk-in. There lies the crux of my issue with AT, with no testing no work AT has just decided that what you have now is a good as it gets and we need to convince the dealer community to live with it and just keep payin us. I’m sorry can’t buy that one. Among other recent issues we will not get into in this post.

Are there people that just want to sit down and buy a car, sure there are they minority by far. Ed people just don't have that kinda time anymore. How many people do you know will give up their entire Saturday to go buy a car now?

Ed I know you’ve been brainwashed into thinking people are just going online looking at what available and showing up, I am sure AT would love for us to believe that the facts are that’s just not true. Are some yes… the dealerships that do the best job with AT on this will tell you it’s about $600 a car. Ed that’s not that great!

Ed my little store is an AT customer and we are a buying center in Houston. I can tell you exactly what comes from AT and what percentage of my business that is. I can also tell you we sell cars like Amazon does 50 to 60 times a month, we sell

cars like Staples and Best (clicks and bricks) 30 to 40 times a month and we sell like a traditional dealership someone coming to the store 30 to 40 times a month. But I can also tell you that out of the 150 cars we sell on average every month 145 of them convert, by web form, email, chat or phone. Take those numbers and times them by a factor of 15X and you will have the dealership I sold my dealership prior to this one to, Texas Direct Auto. I have watched TX Direct sell and deliver 1600 vehicles in a single month all over the country and 1000 of the sold like Amazon. This is why I am so passionate about this, I have seen it first had and I live it every day.

Sure we apply good web design principals all the time without testing, something is better than nothing. But to truly ring the bell and get all you can you have to test. I don't see why producing the data would be a problem for AT it would make dealers feel better that the money they are spending is going to improve the product they are buying somewhat, but that is a business decision AT has to make, don't bitch when Reynolds doesn’t share nicely either then, that door swings both ways.

In the end Ed I know AT is doing well, but are they doing good? In my mind right now they are not.

I too enjoy our discussions no one ever walks away sold but you do walk away thinking and that’s a good thing.

Take care,

 

[Ed Brooks]

Larry,

I know you've dismissed it in the past but I think it's worth linking to again: the study released in January of 2009, conducted by ATC and Northwood University is certainly on-point, conducted with a large sample size and transparent in it's methodology. http://www.slideshare.net/AutomotiveSocial/web-or...
While I'm linking to data, the 2006 ComScore study is worth noting as well http://www.comscore.com/Press_Events/Press_Releas... showing that anywhere from 2 to 10 TIMES more conversions occur offline than online after online information gathering.

Additionally, you are the only person I've ever seen define a conversion as: Any form submission, phone call, email or chat session where good re-contact information is captured for the dealership to follow-up with the customer.

The standard seems to be much simpler: The percentage of visitors who take a desired action.

The desired action for car dealer is usually selling a car, but it might be setting a service appointment or selling an accessory. Your definition is much more narrow, and while it plays to your goals, I'd submit it doesn't really define a conversion well.

And lastly, a place where I agree with you completely. "I also believe that if you aren’t converting well you are getting as many walk-ins from your web properties either. These things are connected."

I'm glad we agree on something. I disagree that they are connecting offline because "people are('nt) just showing up because they want to, I think they are because they are so frustrated with the online process", I believe there are huge numbers of people that don't want to spend time on the phone or on email after they have made a decision. They walk into a dealer armed with an enormous amount of data, ready to make a deal.

I have no idea where your $600 figure comes from. I assume you're talking about front-end gross. If I look at my best-performing dealers (by volume), it is way, way too low. Not by a few bucks, but by multiples.

 

[Craig Belowski]

Wait.....I'm confused, so does this mean Matt Murray is cool or doesn't it?!?

All I know is I'm not picking a fight with Larry....way too much to read.