July 8th, 2010 — Uncategorized
With my recent mobile fixation, I’ve quickly realized that some automotive mobile sites are nearly impossible to find. It’s hard enough to remember URLs such as http://suzukiauto.com, but good luck with http://mobile.volvocars.com/us/mobile/Pages/default.aspx or http://mobile.usablenet.com/mt/www.vehix.com.
Most people use search to find sites, but since the iPhone’s built-in Google search returns the main sites, finding the mobile site requires a redirect from the full URL. In most cases, this works nearly seamlessly: when I go to http://www.chevrolet.com I get sent to http://m.chevrolet.com automatically. Getting to the few sites (Acura, Jeep, Suzuki, Vehix, Volvo) that don’t redirect requires a substantial effort that few, if any, people would undertake (well, unless it’s their job. Lucky me). At least I’ve seen mobile ads for some, but that’s far from sufficient to drive traffic.
Cars.com has a unique approach on the iPhone. It presents the main page with additional links to the mobile site and its iPhone application, which allows users to choose the experience they want.
Edmunds offers iPhone visitors an upfront choice. It’s a nice method, but Fandango’s is slicker and simply looks better.
This doesn’t mean you should always redirect. Many smartphone users may want to view the full site and should be given that choice, even if it’s a suboptimal experience. Consequently, many mobile sites also offer links back to the main site. In a few cases, however, the experience is so poor (especially because of Flash-reliant navigation) that they would be better off NOT giving that option. Kia’s main site doesn’t even load on my iPhone while the Infiniti and Nissan sites basically tell you it’s not worth the effort. But they’re the ones that linked me there!
My testing was limited to my iPhone and it may be easier to find mobile sites on other devices. Still, it’s clear that our industry is still trying to figure out how to manage that interplay between multiple sites and multiple ways of accessing them.
June 16th, 2010 — Uncategorized
Earlier this week I looked at the relative SEO performance of OEM sites and this time I’m setting my sights on the third-party automotives. I used the same keywords as my 2009 SEO test, which allows me to make direct comparisons with how frequently third-party automotive sites appear in the default first page for both Google (10 search results) and Bing (10 to 20 search results, depending on the type of term).
The best performers across both search engines were Edmunds, Yahoo! Autos, and Automotive.com (a lead aggregation site). The first two have tremendous relevant content and lots of linked content, both of which are critical. But these leaders have clearly made SEO a priority.
The worst performers were Autotrader, kbb, MSN Autos, and NADA Guides. All four also feature great content and are among the most highly visited automotive websites, neither of which is apparent in their SEO performance. MSN Autos is almost invisible even on Microsoft’s own search engine, which points to the fairness of Bing’s search process even while it highlights the relative SEO weakness of MSN Autos.
Claiming a premium spot in search results is a zero-sum game – in order for one site to go up, another must go down. Compared with the 2009 results, this test saw Automotive, Edmunds, and MotorTrend appear less frequently while cars.com, usedcars.com, and Yahoo! Autos claimed more top spots.
One notable riser was U.S. News Ranking and Reviews, which consolidates model reviews and ratings from a variety of major online sources. It appears to be a relatively new entrant (at least, I haven’t noticed it before) and may be better engineered from the ground up to support SEO.
The full results are shown below.
June 14th, 2010 — Uncategorized
Every so often I like to compare the relative SEO performance of major automotive sites. In the past, I’ve found that OEM sites do well on a few obvious keywords, but relatively poorly on more detailed searches that should lead to internal pages.
For this test I used a variety of search terms, including:
- Brand name
- [brand] + “dealer”
- [model name]
- [model name] + “colors”, “specs”, etc.
I ran these keywords through Google and looked for hits in the first ten results. The targeted outcome for each search is a relevant OEM page with that specific information. For instance, [brand] + “dealer” should link to the OEM’s dealer locator, while [model name] + “specs” would ideally return a page with the vehicle specifications. Returning the home page or the model page is relevant, but not specific, and is therefore noted accordingly.
The worst performers in my test were Land Rover, Mitsubishi, Scion, Mercedes-Benz, and MINI. The first three barely appeared in the top results while the latter two usually returned only the homepage, which forces the user to search within the site. While all five sites are built entirely in Flash, we know that this doesn’t preclude effective SEO performance. After all, top-performing sites such as Hyundai, Infiniti, and Volkswagen are also Flash-heavy. Site developers need to put the work into making the content searchable (or at least associated with appropriate keywords) and perhaps provide a means for other sites to link directly to these pages.
Even with the laggards, I was encouraged to see that many searches returned links to internal pages. This represents a marked improvement over my 2009 SEO field test. For instance, in my latest test all but one Acura-based search returned a directly relevant page, whereas last year none of the specific searches did so. Lincoln, MINI, Nissan, and Volkswagen also did substantially better.
The full field test results are shown below.
March 14th, 2010 — Uncategorized
The Pontiac and Saturn brands may be dead men walking, but as long as there are vehicles for sale, GM must support them to some degree. Until recently, both brand Web sites with full vehicle research information were still active. Several weeks ago, these were replaced with more limited sites.
Pontiac’s scaled-down site seems more concerned with Buick and GMC, which are both generally co-located at the same dealerships as Pontiac. Many pages and links (e.g. “Pontiac Owner Offers”) suggest that shoppers consider these sibling brands. The “Service and Maintenance” link sends shoppers to www.goodwrench.com, which is disconcerting. There is some vehicle research and inventory search to find used Pontiacs, but Pontiac-related content can easily get lost in the clutter.
The site is clearly a placeholder prior to killing Pontiac’s online presence, so it’s curious that GM continues to buy some Pontiac keywords on search engines (Google, Bing, and Yahoo) AND direct shoppers to www.pontiac.com. The Google paid ad text for “pontiac” says “Build Yours on Official Site”, but that functionality no longer exists! It would be better to direct shoppers to www.gm.com or a competitive GM model, or just stop Pontiac-related SEM altogether.
Saturn’s current site feels more like a respectful salute to the brand’s legacy. Useful links for current owners are prominent and shoppers can quickly locate used vehicles. Other GM models are also suggested as alternatives, but it’s not the focus of the site. Saturn’s search keywords are not being purchased, either.
Winding down the brands is certainly not simple, but online it seems as though Pontiac is stumbling a bit while Saturn’s approach is more straightforward and, well, approachable.
October 20th, 2009 — Uncategorized
Moderator: John King, Saatchi & Saatchi
Panelists: Mike Margolin, RPA; Michelle Morris, Google; Payam Zamani, Reply.com; Kari Allen, Yahoo!; Skip Streets, Compete
- Current search volume is similar to Q1 of this year.
- Cash for Clunkers stimulated not only search in general, but search resulting on an OEM site subsequently.
- Following up on leads from Cash for Clunkers even now, is worthwhile.
- Query volume is largely dependent on deals to be had.
- Mentioning competitors in ads and search depends on strategy – it’s not for everyone, so strategy needs to be clear.
- The intersection of Search and Social Networks has huge potential for the future of search.
- Think beyond what’s typed in; the future of search will be recommendation engines providing solutions, not just a list.
Finding Cross-Tier Ad Efficiencies
Moderator: Jon Schulz, Specific Media
Panelists: Dorthy Miller Shore, The Miller Agency; Scott Kelly, Ford Motor Company; Dave Metter, MileOne
- Let the OEM build consideration and keep Tier 2 and 3 in the lowest of low places in the funnel.
- Mix Tier 1 messaging back in with the lower funnel.
- As a dealer, make certain your site is the absolute best it can be in order to capitalize on all efforts from Tier 1 and 2.
- Manage expectations of who’s paying for the online advertising – clear and regular communication among the Tiers is key.
August 27th, 2009 — Uncategorized
In order to assess the current state of SEO for third-party automotive Web sites, I conducted a field test which I applied a set of automotive search terms to Google and Bing. I noted which sites appeared in the default first page of results – ten for Google and ten to twenty (depending on the type of search) for Bing. My findings are included in the charts at the end of this posting.
This analysis is obviously far from comprehensive, but it does highlight relative standings. The strongest field-test performer was clearly Edmunds, which appeared in two-thirds of all the searches I performed. AOL Autos, Automotive.com, MotorTrend, and Yahoo Autos also did well. On the flip side, Autotrader, kbb.com, and MSN Autos are among those that did poorly.
Comparing directly competitive sites can highlight best practices. For instance, while kbb and Edmunds both offer vehicle overviews, the latter has far more content, e.g. 1,400+ words for the Jeep Wrangler vs. 199 at kbb.com. More content on the pages results in more content to index and a greater range of possible keyword hits.
SEO is just one of many ways that sites can drive traffic. Some automotive independent sites have clearly emphasized their search optimization efforts, while others seem to have given SEO short shrift. But it seems likely that those at the top will have an ongoing advantage against their rivals.
SEO Field Test Results for Google
SEO Field Test Results for Bing
August 20th, 2009 — Uncategorized
The need to build one’s Web site for search engine optimization (SEO) has become a critical component to the marketing of almost any Web site. This posting provides advice on how dealers can improve their Web sites to make them more likely to show up on search engines. Even if you can’t implement all of these techniques yourself, this article can help you understand how an SEO vendor could assist you and what questions to ask them.
- Good text for consumers equals good text for search engines. If your site was built from a standard dealer template, you probably have generic text on most pages. Change it! Write text that’s specific to your brands, vehicles and dealership(s).
- Make sure that relevant keywords, the words Internet users would actually type into search engines, appear somewhere in your site. Some of this is common sense—if you’re a Hyundai dealer, the terms Hyundai, car, Elantra, Genesis, etc. are clearly important. So are the town/location and name of the dealership.
- Also consider keyword research to determine terms users are actually using on search engines. Google, Wordtracker
- Use unique and relevant titles for each page on your site. For instance, the accessories page can have the simple title “Accessories.
- Fix all broken links. Search engines analyze a site by jumping from link to link—broken links may mean that some pages on your site don’t get visited and therefore that content doesn’t get categorized.
- Have a unique and relevant file name for each page on the site; these may be the same as the titles, e.g., “Accessories.
- Include a site map that links to the most important pages, e.g., new vehicles, used vehicles, sales, service, parts, appointment, address/directions, contact information, among others. This helps the search engines find all your pages and makes your site more useful to site visitors.
Avoid abusing or gaming the system. Search engines may penalize sites caught engaging in these practices by reducing their ranking or removing their listing altogether. Rule of thumb: if it seems questionable, don’t do it.
- Add misleading and irrelevant text (e.g. “sex”) in a page or in tags
- Use hidden text or links, e.g., white text against a white background. Include only text and links that are visible to site visitors.
- Have multiple versions of the dealer site just to increase your search ranking.
- Participate in linking schemes that create links from irrelevant sites to your site. This scheme became more common because Google’s ranking scheme takes into account the number of links pointing to a site.
Of all these suggestions, the most important by far is to have good content. Put simply, a search engine is supposed to find the content that site visitors will find useful. Make sure that useful content is on your site. If you don’t have good and relevant information, your Web site won’t be useful to your site visitors and won’t rank prominently in search engines, no matter how many tricks you use.
Granted, anyone who is searching for your specific dealership online will find it. At the same time, you want to make sure that you capture the more casual searcher and also give yourself some ammunition against rival dealers who may be buying up your search terms.
August 4th, 2009 — Uncategorized
Search Engine Optimization (SEO) has become a core part of the web strategy for almost every major company and automotive manufacturers are no exception. I recently did a field test to see how well the U.S. brand sites for OEMs perform on Google and Bing. Google is an obvious choice given its dominance of the market. I chose Bing for two reasons: 1) Microsoft is trying to do search differently than Google, so the results are more likely to be more different 2) The recent deal with Yahoo will make Bing the clear number two player in search
The best performer [in this admittedly limited test] was Toyota, which returned brand site results even for specific searches such as “camry colors” and “tundra accessories. The poorest performer was Infiniti, which didn’t have any OEM results on the first page of Bing. Hummer’s results were also problematic: every search defaulted to the global page, which requires one or two more clicks and a popup in order to access the U.S. site.
There were also notable differences between Google and Bing. Specifically, Bing more often returned OEM brand pages for detailed searches such as “tiguan mpg.” It’s possible that Microsoft’s emphasis on relevance over popularity would give OEMs a greater opportunity on their own automotive terms.
I also noticed some great SEO practices worth mentioning. Hyundai has built its pages to produce marketing messages within its organic search results. For instance, searches on “azera” and “hyundai sedan” bring up the following results.
GM has created separate microsites for its dealer locators. Even though the appropriate brand site pages often show up in the top results, this ensures greater prominence for an important lower funnel shopping tool.
The full field test results are shown below.
SEO Field Test Results for Google
SEO Field Test Results for Bing
June 12th, 2009 — Uncategorized
After years of losing ground in the search wars, Microsoft recently launched the latest iteration of its search service. Microsoft is positioning Bing as a “decision engine” that returns a limited set of highly relevant answers rather than a typical “search engine. The intent is to help users find what they want faster.
For instance, searching for “ups” on Bing returns the United Parcel Service with seven key sublinks, a customer service number, and “track a package” functionality right on the results page. On some sites, Bing will similarly provide results page access to site search functionality. The left menu presents a quick means of redefining the search parameters. Overall, the page is uncluttered and addresses most needs quickly.
Bing’s approach is different than the now-standard search paradigm, which orders results by popularity while mixing in sublinks, additional search options (e.g. “similar pages”, “filter”), and news results. With this particular search, it also creates much more text to parse.
Bing’s ability to provide direct access to site functionality is handy on some automotive sites. For instance, searching on “Acura” bring up the Acura brand site, sublinks, and acura.com’s site search functionality.
Other automotive-related searches show some differences from the standard search results. For instance, model-level searches (e.g. “Ford Fusion” or “jetta”) often return MSN Autos data high in the results, sometimes even above the manufacturer site.
The search parameter redefinition can also be helpful for automotive searches. For many vehicles, the first listed option is “Reviews”, which reflects the importance of online reviews in the vehicle shopping process. The menu also changes based on vehicle type. For instance, the first listed option for “ford f150” is “Accessories” while for “prius” it is “Problems.
Bing offers some nice evolutionary steps related to search, but some can probably be easily co-opted by competitors. With automotive searches, Bing may help consumers get to information a little faster, but I don’t see anything particularly game-changing.
The initial returns are encouraging for Microsoft – in the first week after Bing’s launch, Microsoft’s average daily penetration among searchers rose 1.7 percentage points to 15.5%. (Note: people may use multiple search engines in a day.) The challenge for Microsoft is getting the casual and curious Bing trialers to switch permanently.
April 11th, 2009 — Uncategorized
As everyone knows (or is quickly figuring out), Twitter enables real-time microblogging, with information and news being disseminated almost instantaneously. Some automotive companies have chosen to be a part of the conversations, but Twitter still offers tremendous value even for those that don’t join. Companies can learn a lot just by listening.
Back in September, we posted Do-It-Yourself Blog Monitoring to show how anyone can quickly set up a free social monitoring solution. The three primary elements are:
- Staying on Top of the Conversation
- Tracking Volume Trends
- Alerting, e.g. if you anticipate a negative perception / PR issue
These same three elements can be implemented for Twitter.
Staying on Top of the Conversation
Even if you’re active on Twitter, you can’t possibly follow everyone who might be talking about you. Fortunately, Twitter allows anyone to search across all tweets and also delivers updated results via RSS. The search results for “GM” on April 3 are shown below.
Tracking Volume Trends
Twist makes it easy to track Twitter trends, as shown below with the terms “GM” and “Ford”.
Chatter for both terms was consistently low until March 29, when news of Rick Wagoner’s resignation from GM first leaked out. GM-related chatter spiked the next day and then rapidly fell off. There’s no straightforward mechanism to save search terms, but one possible workaround is to embed dynamic charts into a non-public page that you can periodically check.
Set up search results (as described earlier) via RSS with appropriate and specific terms, e.g. “<my brand /> hate”.
Twitter is still new enough to the scene that the available tools are rudimentary compared to the wealth of mechanisms that can be used to analyze blogs and message boards. But, as shown above, tracking the pulse of conversations related to your company is not difficult. And there are plenty of other great ideas out there on how to monitor Twitter – here’s a couple of additional resources:
Monitoring Dashboards: Why every company should have one
How to Monitor Online Conversations