How Geico Wastes $90MM a Year on SEM

[tplDrew2]

The more and more people I talk to, the more often I find myself taking a militant stance against gigantic search engine marketing (SEM) spending. My position is simple: if you create relevant, frequently generated content that’s high-quality and of a significant volume, you don’t need AdWords. That’s it.

[omarchowdhury]

This person obviously does not understand the value in capturing BOTH organic search visitors and paid search visitors.

Let us say 30% of the searchers for "auto insurance" click on Geico's organic search listing. What about the other 70%? Some of them click on other organic search listings, some of them will click only on other paid search listings, some of them will click on a combination of listings (comparing).

It would only make sense for Geico to buy paid search listings as well - to capture SOME of that other 70% that do not click on their organic listing.

Simply, ranking organically and buying keywords will only increase the chance for a potential customer to find your site. And with an industry as competitive as car insurance, if Geico doesn't spend that $90MM to acquire new visitors someone else will.

Also I highly doubt the accuracy of the $90MM figure as Spyfu's database is only an estimate.

Finally let's not forget this article also doubles as a marketing material for Tipping Point Labs, I'm sure they agree they are "expert story-tellers" who completely "understand online content marketing".

[tplDrew2]

Thanks Omar. I think you make some great points. However, as much as google would like you to believe that AdWord buys are more successful than the top ten organic search results, that's just not the case. Organic search results have higher conversion rates than adword buys. Hands down. I'll give you the SpyFu inaccuracy, however, it's some insight into adword spending. Also, although the post helps market tippingpoint, we're much more eager to spark a debate and engage in an active dialogue. Thanks again for your insight.

[omarchowdhury]

How can you say that paid search traffic converts less than organic search traffic? The beauty of paid search traffic is being able to target the RIGHT keywords for the RIGHT customer and the RIGHT time.

It is inherently wrong to say paid search traffic converts worse than organic search traffic because different variables will change the outcome (conversion rate). Those variables can be keywords, negative keywords, type of keyword matching, ad copy, time of day, landing page, geography, industry, product. I can change these variables and make the paid traffic convert poorly compared to the organic visitors. I can also tweak these variables and have my paid traffic converting much better than organic traffic!

What is my point? The data that you used to draw your conclusion, that organic is better than paid, was simply data from a poorly converting paid search campaign. Not to worry though, most paid search campaigns are full of flaws because a lack of understanding of the Adwords system.

Furthermore, Google does not want me to "believe" anything. I have websites that rank organically AND which receive paid search traffic (Google). I have my own data to draw conclusions.

[profgubler]

I see some of his points. However, buying keywords on searches that you rank highly in is often not a negative. We buy brand terms for our clients all the time and it only helps them. They do very well in the organic listings for these same terms, but they often make more money through the PPC terms. The reason I feel that it often works that way, is that people clicking on PPC terms, are often more in a buying mode.

Because, people clicking on PPC ads want to be marketed and sold to. Is this always the case, no. You have to run the analysis on it to determine if it is worth it to buy keywords you do well with in the organic searches.

[breck]

Absolutely. Also, I question the assumption that a significant percentages of searches for "cavemen" would convert.

I would ballpark that conversions on "cavemen" versus "auto insurance quotes" would be much less, probably 2-3 orders of magnitude or more.

But the basic idea that Geico is wasting some money on search is probably right. However, it could also be that they are not spending ENOUGH. It's hard to tell without being on the inside.

[randallsquared]

It's hard to tell when you are on the inside. :)

[breck]

Good point!

[daniel-cussen]

Totally. When I'm looking to buy something, I often click on ads google places. Often to buy.

[JoelSutherland]

This article misses the much bigger picture on two points:

1) Geico is forgoing efficiency in favor of growth. In the past 20 years it has roughly quadrupled its market share. It can do so safely as it is owned by Berkshire Hathaway. As many other auto insurance companies have failed, Geico has climbed to a top 4 insurer.

2) Geico is owned by Berkshire Hathaway. Unlike some of the other companies, it does not have profitability as an exclusive motivation. Even at break-even, it generates significant float for Bershire to invest. This is why growth is favored to high margins.

[blurry]

Unlike some of the other companies, it does not have profitability as an exclusive motivation. Even at break-even, it generates significant float for Bershire to invest.

Can you explain?

[JoelSutherland]

An insurance money makes its money from the premiums people pay and spends money on claims. Once a claim is made, there is a period of time before the company must pay it.

Even at break-even, an insurance company must keep a large amount of cash on hand to pay future claims. This is called float. Even though the company does not 'own' the money, it is free to invest it while it holds it. Buffet has used Geico and other insurers primarily as vehicles to invest their float.

[byrneseyeview]

On the other hand, it's important to note that Berkshire's insurance companies have been unusually profitable. Their combined ratio (ratio of expenses to premiums collected) is often below 100% -- not especially common for the insurance business. However, that average incorporates more volatility, and more deviations from the norm.

[tapostrophemo]

Well, any insurance company that doesn't have a combined ratio (CR) less than 100 is not making money.

Example: a CR of 92 means that for every dollar of premium, the company has to spend 92 cents, leaving 8 cents of profit. A CR of 102 means that the company has to spend $1.02 for every dollar taken in.

How much spent depends, in general, on operating expenses and claims paid. Limiting how much you spend can only get you so far; hence the rest of an insurers profit comes from investments.

(And where does the money for investing come from? Collected premiums!)

[byrneseyeview]

An insurance company with a combined ratio of 99 has a cost of capital lower than the US treasury. So yes, it is probably making money -- even if it has its assets in T-bills.

[antidaily]

It's Berkshire Hathaway with a k.

[coglethorpe]

If someone at Geico can track that the $90 million a year gets them more than that back in business, why is it a bad decision?

Sure, they could create some insurance related content and have some articles show up higher in the rankings, but I doubt they would hit first page for all they keywords they want.

[tplDrew2]

Thanks so much for your comment. I too hope they’re doing some real testing and tracking to make sure they’re seeing a big ROI. Too often I’ve seen agencies hand deliver reports to clients that are ’stretching’ at best. All the client cares about is that they’re performing ‘better than the benchmark.’ As long as that’s the case, they don’t dive much deeper. It's entirely possible to garner every result on the first page of search results, if the content is REAL content of value it can be done. Thanks for the comment. Really appreciate it.

[ssharp]

This guy completely discredits his company with a nonsense post like this. The reason Geico wouldn't PAY for the Caveman keyword is because it's not going to have much benefit. When someone types in "caveman" into Google, they aren't looking to buy car insurance. When someone types "car insurance quotes" into Google, they are. Geico wouldn't be dropping 90MM a year into that keyword if it didn't have a positive ROI. Althought "Caveman" might be cheaper to buy, I would bet any amount of money that it still have a negative ROI.

[tplDrew2]

Thanks for your comment. You don't need to buy the Caveman keyword... that's my point. What you need to do is convert people who are looking for caveman content at the height of it's popularity, and turn them into Geico loving insurance consumers. That IS the goal of the ad campaign, I would assume. I'd be surprised if their keyword campaign is THAT effective. If they don't buy the keyword, I'd imagine they see little to no drop in conversions. Thanks again for the dialog.

[ssharp]

So all Geico needs to do is convert people who are looking for cavemen into paying customers and they can stop paying $90 million a year to AdWords?

Honestly, your article doesn't make any sense. You're title is that Geico "wastes $90MM" a year in paid search. Yet you offer no evidence that Geico has a negative ROI on this spending.

If your point is that Geico isn't concerned with organic traffic, they are ranked #1 for "car insurance" and #2 for "car insurance quote". I think they seem to be doing alright.

What do you think the point of the Caveman commercials were? The appeal was "it's so easy, even a caveman could do it". The cavemen were an added benefit of a recognizable character. While Geico probably could have spent some time and money getting higher ranking for "cavemen", you don't provide any evidence that doing so would actually covert sales. Missing this is huge because common sense tells me that folks looking for the Geico Cavemen are probably more interested in some quick entertainment than car insurance.

[nikblack]

It is always tempting to point to a big company and to try and argue how stupid they are and how they waste money etc. but as most of the comments here have already pointed out, there are fatal flaws in their argument.

I will add one more to the list. The difference in click-through rate for the top result from the second result is 3.5x. ie. Top result gets 43%, second place gets 12% CTR. Ranking first means a lot.

Also the spikes in trends for the caveman keyword being linked to Geiko is very speculative at best.

[tplDrew2]

Thanks for your comment. Really enjoy the dialog. The caveman spike is correlation, not speculation. It does correlate very closely, the news stories for the time period even other platforms like YouTube, trend very highly for Caveman around the same months... but you're right - I don't know for sure. Where did you get your CTR stats? Those are intriguing. Our experience is very different. Thanks again!

[mattmaroon]

He's making the assumption that Geico hasn't done any sort of testing to verify that this is optimal. You can use urls in adsense to track which of your traffic comes from there and which organic.

My guess is that Geico isn't spending $90m blindly. But, like everyone in the tech industry, the author assumes that everyone not in the tech industry is stupid.

[transburgh]

Just a personal note, my experience with SpyFu's numbers are horrible. They have always been way off the mark.

[earl]

I've worked for a company spending hundreds of thousands of dollars per month on SEM. SpyFu's numbers are really poor.

[tomkinsc]

This is largely unrelated to the content of the article, but as a community of intelligent individuals, can we make an effort to avoid TLAs (three-letter acronyms?)

It would be helpful for the uninitiated to have acronyms spelled out. For those of us in the physical sciences, "SEM" is generally understood to mean "scanning electron microscopy." While this is obviously not the case of Geico's search engine marketing (surprising that they have a daily search advertising budget of $244k!), disambiguation would be helpful.

[speek]

That's ballsy. I like it.

[tplDrew2]

Appreciate it. Ballsy is fun.

[earl]

So, he thinks nobody at Geico has ever looked at that $90MM and tested what happens if they spend less? Really? The idea that nobody has ever wondered what would happen to his or her bonus that year if they saved even 5% of the SEM spending is kind of far fetched, IMO.

[tplDrew2]

Thanks earl for your comment. I'm not that silly to assume no one's watching the adword buy, but I also know from experience that clients blindly take a 'better than the benchmark' as a sign of success - especially in the keyword space. Thanks again.

[gddub]

Because his point is that they could spend less and still generate the same income therefore increasing the profit margin.