You are here

Dear eBay, Your Ads Don't Work Because They Suck

March 13, 2013
71

 Do AdWords not work or is eBay bad at AdWords?

Does Google AdWords work? There’s a story making the news this week about how eBay supposedly wasn’t able to make paid search work for them so they stopped using Google AdWords. Ray Fisman, who wrote up the study for Harvard Business Review, said “Their findings suggest that many paid ads generate virtually no increase in sales, and even for ones that do, the sales benefits are far eclipsed by the cost of the ads themselves.”

I couldn’t help but laugh when I read this, and I’m sure many of you did as well, because it’s pretty well-known within the search engine marketing community that eBay is the butt of many jokes when it comes to adopting advertising best practices.

eBay is probably the world’s biggest abuser of an ad writing tactic in AdWords known as Dynamic Keyword Insertion or DKI. As its name suggests, DKI dynamically inserts the user’s query into the headline of your ad, as follows:

eBay Ads on Google, eBay fails at Google Ads

When used correctly, DKI can be a highly effective ad writing technique. But when employed in the way that eBay uses it, the results are disastrous.

Dynamic Keyword Insertion Run Amok

eBay’s AdWords strategy appears to be to pick every possible word in the dictionary and run them on Dynamic Keyword Insertion. For the last 10 years or so, they’ve been running ads on the most ridiculous things including stuff that doesn’t exist or stuff that is illegal to sell. Here are just a few examples:

Need a Perpetual Motion Machine? Shop on eBay and Save! Discount Perpetual Motion Machines!

Example of failed Google advertising by eBay

Feeling lonely? Need a wife?

eBay fail, lonliness ad, AdWords doesn't work for eBay

eBay sells your wife, eBay's AdWords ads aren't working

Or maybe you’re looking for love in all the wrong places?

eBay Paid Search Doesn't Work

Or how about some slaves or Bill Gates while we’re at it.

I could go on but I think it’s safe to say that if you’ve used Google in the last 10 years, you’ve come across a bunch of bizarre eBay ads that make absolutely no sense. So I can’t say I’m shocked that they finally realized they weren’t getting ROI from AdWords.

It’s Not AdWords, It’s You: The Problem with eBay’s Ad Strategy

The problem with eBay’s carpet-bombing ad strategy is that it’s doomed to fail. The way AdWords works is that they allow companies to pick what keywords they want to show their ads on.

Now to ensure that advertisers do a good job at choosing keywords that are relevant to their business (which is important because Google would prefer not to display stupid, irrelevant ads, which annoy users), they give you a discount on ads that have high click-through rates. The flip side of this is that they penalize lousy ads (like ads for slaves or loneliness, etc.) by charging as much as 10x more.

Furthermore, because your ads are so weird, if someone clicks on the ad and goes to the site – they’re not likely to find what they’re looking for (no perpetual motion machines in sight), and won’t convert into a sale. So eBay was paying a high premium for mostly worthless clicks.

These and other factors likely explain your study results. I’m sure Google LOVED your business, but of course you weren’t getting return on that misguided advertising spend.

Attention John Donahoe: I Dare You to Fire Your Ad Managers and Give Me a Shot

So eBay, I hope you can see that it’s not AdWords – it’s your dumb ads that are killing your account.

Success at paid search is possible but you’ll need to invest more time than just creating one universal ad template then jamming every keyword in the world into that one ad using dynamic keyword insertion.

I’ve been doing paid search for 10 years and have made it work for thousands of businesses, and I’m absolutely certain I can make it work for you, too. Here are a five free tips for you:

  • Be more picky with your keywords. Instead of bidding on hopelessly unspecific keywords like “baby,” choose more specific terms that show at least a little intent, like baby bottles, or baby stroller, etc.
  • Consider using Dynamic Search Ads – a new technology from Google that picks your keywords for you automatically.
  • Use negative keywords to prevent your ads from showing up on irrelevant searches. This could save you millions of dollars! No joke!
  • Use remarketing to target shopping cart abandoners. It’s a super-effective way to draw people back onto your site.
  • Grade your AdWords Account for Free

The first 5 tips are on the house. John Donahoe, your AdWords account managers are an embarrassment to your company and your AdWords account is in shambles. Call me and I’ll turn your Google PPC account around for you!

Hit me up on Twitter

 

grade your adwords account

Comments

Rich
Mar 17, 2013

Great story.  Makes sense to me.  And, I laughed my ass off at some of the examples you gave.     

Internet User
May 22, 2013

Ebay is so worried about making to much money, it is progressively going down hill in sales.Their Tech Team is in way over their heads.Their feedback System is a Legal Nightmare waiting to happen.The Big Search Engines now days are Ad Related..pretty much useless now.Rich Companies are to be avoided, exspecialy those that charge fee's at a high percent.America is a bit tiered of paying some rich guy for not working. Freeloading Freeloaders 

Nathan Pabich
May 15, 2013

Some of those screenshots bring me back! Great article, and I can't agree more that DKI is not a good fit for everything. The biggest abuser I see currently in both my personal and professional searches now is Target. That's the problem when you want to sell everything - you may not do it well. Especially in this medium.

Mark Stockton
Apr 26, 2013

Several companies do this. JustEat are my favourite culprits, they allow words such as "Manure" to be dynamically insert into their adwords adverts. Not sure I would want to associate manure with food...

Anonymous
Apr 19, 2013

Google adwords plain and simple is great to make google money but it fails to make produce paying customers.Google adwords sucks!

Tim Fouracre
Apr 03, 2013

We did some analysis of Google Adwords when we realised we had spent a cumulative £100k since we started our SEM campaign.The CPA has crept up over that time so it's definitely worth spending on ads as soon as possible particularly while your industry is in its infancy.... before it gets too competitve and expensive.Lessons learned from spending £100k on Adwords. 

Luis
Mar 25, 2013

My question is... During 10 years and spending and huge ammount of dollars, the Google Account Manager couldn't suggest how to improve or work in the ads?

Yaron Lapidot
Mar 25, 2013

You found 5,6,...100 wrong search terms out of millions, this is drop in the ocean.Managing such a huge campain, can't be done without a few mistakes or more.I think EBAY's biggest mistake, is competing in search terms with a bad conversion rate.If EBAY will focus only on the profitable terms, they will sell less but will earn more.Less is More... :-)

Jared
Mar 16, 2013

I've seen some of eBay's Adword ads and I have to admit they are pretty bad and you can see why they're not seeing a return from them

JoJo
Oct 18, 2014

That's hilarious that they were selling babies and Bill Gates. :D I wish I got a discount on him.

Beatrix
Jul 01, 2014

Really good article. But be honest, you have to play with google not against.

Anonymous
May 29, 2014

google ads is a joke they bulid a web page them they try to sale to u and and never advertise the way they promese to you dont dont do google ads if you dont want disappointed by them i really had a bad experience with them

kim
May 02, 2014

The examples are outdated, from many years ago. You can tell from the way Google SRP looks.

seo
Mar 06, 2014

Dynamic Keyword Insertion is very helpful big campaigns with many keywords.

Andrew Delville
Feb 25, 2014

It looks like someone in the eBay marketing department screwed up with this one!

Vedran Koren
Nov 04, 2013

Excellent article. You just hit the core of the problem.  

Alan Mitchell
Apr 15, 2013

eBay's paid search ads are always making the headlines. Here's another humerous example:http://www.semsamurai.com/2010/01/adwords-dynamic-keyword-insertion-fail/I also think it's wrong for eBay to write of brand ads just because they didn't happen to make them work.I've found 13 tangible benefits of bidding on your brand terms: http://www.calculatemarketing.com/blog/techniques/13-reasons-why-ebay-are-wrong/ .I'm sure there are many more.

vv
Jan 31, 2015

Google adwords suck. It's not worth the money for ppc's they ask for.

Cynthia
Mar 13, 2013

Who ever said you can't buy happiness obviously hasn't tried searching for it on eBay. It's there, next to love and success.  

Larry Kim
Mar 14, 2013

it is the most bizzare adwords strategy ever.

Philip Cohen
Mar 13, 2013

I would not bother applying for any job at eBay—they have no concept of competence ...The ugly reality for consumers dealing with the clunky, manipulative, unscrupulous eBay/PayPal complex ..."eBay-Facilitated Shill Bidding Fraud on eBay Auctions: Case Study #5" ... http://bit.ly/11F2eas  

dennisg
Mar 14, 2013

Are you kidding me. Those screenshots are really old, and must be picked from the Webarchives. Your five "free" tips are a joke, as these have been incorporated into the bidding algorythms for years.I posted a response on the Inbound discussion here. But before you make claims that you would do better than the team at eBay, you should realize the story might be more than just marketing optimization. There might be more going on here, stuff you have not been exposed to...

Larry Kim
Mar 14, 2013

Hi Dennis - thanks for the comment. just a few comments. (1) Quality Score calculations take into consideration historical performance of your entire account.  (2) while i don't know everything that is going on in the ebay account, we can infer that there are big problems with keyword selection ad copy. also here are a bunch of recent ebay "top ads" from mid february this year that you can see are still leveraging DKI - http://www.ispionage.com/research/default.aspx?q=ebay.com&lc=1&type=0#tab-1,kadtab-0 

Anonymous
Mar 14, 2013

I suspect ebay does not even have a conventional QS. Did you know there are Adwords accounts with 'averaged' or pre-scored QS? Bet you did not and probably will swear they do not exist but they do and the reality is ebay probably doesn't get a QS like you do. Not to mention the terms you mention are retardedly small and have some to do with match types. Overall for products they probably crush it, their branded terms crush it, so to say their QS is low is cherry picking examples. But I bet ebay negotiated their way out of QS a long time ago.

Larry Kim
Mar 14, 2013

I've never heard of this but if they did have some pre-negotiated quality score and still couldn't make it work then double shame on eBay search marketing managers.

Ionut Mircea Oprea
Mar 15, 2013

Dear Aggresive,Dude take a breath of fresh air and be specific. I have never heard of pre-scored QS either (that is not to say it's not true) just that you haven't given an example and that takes away some credibility. As for the key words, it makes all the sense in the world to target long tail and specific key words for better QS (where you say "retardedly small" I say specific and optimised for conversion).  

Chris Zaharias
Mar 14, 2013

I agree 100% with dennisg. Larry, you've painted yourself the greater fool by saying SEM success is just aboutusing better SEM tactics, when the study is about attribution and is quite powerful in its findings. As someone who's beenin the SEM game even longer than you, I suggest you read all 25 pages of the research report, and I rather suspectyou'll change your looney tune. 

Victor Pan
Mar 14, 2013

Hey Chris,You're a stats guy, and I respect that. I think eBay has a case of fundamental attribution error, and you can't deny that keyword targetting and ad copy were key variables left out of the analysis. Larry's point is still a very valid one once you look at eBay's choice of ads. eBay's study results should be viewed as a case study within eBay and not a model for other businesses.

Larry Kim
Mar 14, 2013

How can anyone be a "greater fool" than the SEM marketing managers at ebay: http://www.vaughns-1-pagers.com/internet/ebay-ads.htmnot possible.

Nick Danenberg
Mar 15, 2013

 I'm not sure that Larry did say that SEM success (or, at least in this present eBay example) was solely about SEM tactics.And to be sure there would be a number of other variables involved that are unknown to us outsiders, in fact unknowable by outsiders, that would influence their results and decision making.But that does not negate the premise that improving SEM tactics would improve SEM performance / success. Of course it would! But would it improve performance enough to overcome the invisible 'hurdle rate' that eBay must work to in order to see a worthwhile ROI from SEM? Again, that is unknown to an outsider.Using our Dynamic Creative software e-channel Search doubled eBay's pre-conversion metrics (Clicks, CTR) for eBay Deals Australia in a joint trial we conducted with them last year. This was achieved not though using DKI, but the insertion of more relevant information, like Price and Savings amounts into the ad. There was no access to conversion data or ROI, however, but certainly SEM tacttics were responsible for greater SEM success. Again, are such 2x improvements enough? Only eBay can know.One other point to make is that there are other aspects to consider regarding advertising effectiveness and the measurement thereof, when advertising at the scale of eBay. For instance, the salience, reminding aspect of advertising (like TV or outdoor) that support a large base of sales, which has a substantial *inertia* component. ie Coca Cola's sales wouldn't immediately drop if it ceased advertising. Measuring the incremental effectiveness of individual campaigns over the short term could lead them to make sub-optimal decisions in the longer term.

Jim Banks
Mar 14, 2013

Great article Larry. Good luck with the account if you win it.My issue often as a competitor of eBay is they never seemed to factor any of the normal points that advertisers look at.It was like "those rules don't apply to us", which as we all know was never going to be the way Google operated and QS and all the other factors were going to matter just as much to eBay.Market conditions, competitor landscape, ad variation, none of these seemed to apply to eBay.It's almost like their strategy is to be #1 on the side ads using the same format title/description no matter which category it is in and day of the week/time of day.They must be sitting on petabytes of data which should allow a much more granular campaign.I get that with the scale trying to be as detailed as other advertisers may not be possible, but I'd say that over 70% of the keywords they use even get an impression, never mind a click.I use a service called Terapeak which shows me all the successful auctions, title's/description in each of the markets eBay operate in.You can see product research, competitor research, category research, what's hot. It color codes the sales volume by category, so I don't understand why they don't use that data as a means of generating a sensible robust strategy.We won't be changing their strategy, but at least we know what we are dealing with as a competitor.

Nick Stamoulis
Mar 14, 2013

Hey Larry,Great article and so very true!  It is amazing to see Ebay bid on everything with poor ads as you pointed out.  Often times I see Amazon as a huge ofender in bidding on everything,  but their ads and targeting seem very good.  Any thoughts to look at Amazon's PPC program? 

Larry Kim
Mar 14, 2013

amazon started out with a carpet-bombing strategy like ebay but later evolved to be more targeted / relevant.

Aaron Levy
Mar 15, 2013

My understanding was this was primarily because of Amazon's affiliate policy. In the past, they let affiliates/associates bid on whatever they wanted. Folks would bid on anything under the sun to get clicks & drop their cookie, assuming that a person would someday buy. Once amazon dropped the policy, their ads cleaned up.  

Damian
Mar 14, 2013

it's at least slightly ironic that the company that pioneered one bidding algorithm couldn't get its head around another

Larry Kim
Mar 14, 2013

when i explain keyword bidding to new clients / people i use ebay as an example (lol).

Andrew Goodman
Mar 14, 2013

We worked for a division of PayPal once. Did a great job. :)It's clear that what eBay should be doing is cherrypicking a smaller volume of better-performing, better-managed inventory.I was never convinced that, even if it were well-priced, that the endless DKI advertising was helping eBay's brand. Rather, it seemed to me it was hurting it, given that everyone already knows about eBay.You're right to point out that Quality Score issues must be seriously driving up the ad cost.Unless Google had been manipulating QS for eBay's benefit, and every now and then there is a dispute, which is essentially over money and favorable treatment, and then eBay makes a big show of "pulling" its ads.I'll leave it there... maybe that's the "stuff we have not been exposed to"... of course Google and eBay are frenemies, so it's a bit like the case of how unbiased can Google really be in ending free product feed referrals when Amazon -- a sort of competitor -- was the biggest beneficiary in the past.If Google has the power decide the prices their direct competitors pay for ads, that's a problem! But that's precisely why you need to know how to fly under the radar somewhat. And tactically, it's important to build a diversified, "anti-fragile" digital lead gen portfolio.  

Bogdan
Mar 14, 2013

This would have been a great article if the screenshots were not outdated.In the first one, where you are searching for "baby", you can see Froogle in Google Search's menu, a product which has been renamed in 2007. On top of that, the screenshots don't have a consistent image, which leads me to think they are copied from some old internet archive.Nice try, Larry Kim!

Larry Kim
Mar 14, 2013

have you really never seen dumb ebay ads for buying "dead popes" or other ridiculous things? 

Larry Kim
Mar 14, 2013

Hi Bogden, take a look at this recent crawl of ebay Ads from Ispionage: http://www.ispionage.com/research/default.aspx?q=ebay.com&lc=1&type=0#tab-1,kadtab-0(or use spyfu or keyword spy, etc.)You'll see that there are a bunch of strange ads for "childlife sandbox", "keyword bible", "spy bug", etc. The date of the ads are February 18, 2013.

searchengineman
Mar 14, 2013

Cough..Cough..Before you jump on Larry's thesis... I think we've all seen wierd E-Bay Ads dated or not.  But also rememberLarry (Wordstream) is the same guy who sucked back..almost 1% of Adwords keywords activitiesand reported the activities..to us. (Thank you) .. so he may know what he's talking about...A simple check in say keywordspy could more than verify...wonky Ads.  Out of curiosity I wonder who exactly was the man(company-ies) behind the curtain..running that Adwords campaign?Because the scary thing is they're probably running other Fortune-500 companies, into the ground. Searchengineman 

Anonymous
Mar 15, 2013

My suspicion is the Ebay is less at fault here than their (terrible) agency. Of course eBay should have been letting them go for lack of performance, but we don't know what other businesses they are partnering in together (the agency might do a great job in other aspects of the marketing spent, who knows).So yes, I would love to know who that agency is :)

Eloi
Mar 14, 2013

Ebay (and Ask.com) have been for the past 5-6 years the prime examples of what not to do in SEM.Bad tactics, bad overall strategy, no taking advantage of new features and the most perfoming tools in the arsenal ... This is why nothing happened when they turned it off in a third of the country. They had a really bad performing campaign to start with.

Larry Kim
Mar 14, 2013

i'm mostly surprised that it took them 10 years to figure out that their campaigns weren't producing ROI.

Eloi
Mar 14, 2013

indeed. I have a feeling the marketing budget was (bluntly) split by channel, and not by performance.

Andrew
Mar 14, 2013

I think the larger issue with eBay AdWords, is users already know what to expect from eBay and that determines if they will click the ad or not. A lot of my friends and co-workers don't like eBay and think of it as a shady place for people selling out of their basement and wouldn't click and ad to eBay regardless of the ad's copy or specific relevance to their search. Personally, I buy and sell a lot of products on eBay, but wouldn't click an eBay ad. I would go to eBay directly in my browser instead of Google for certain product searches. I can say the same for Amazon ads and I'd go to Amazon directly if i wanted to see what they have for me. eBay (and Amazon) search ads may serve more as a reminder to users to go to that market place, rather than a "this is what you want and we have it". 

Claire
Mar 14, 2013

People who don't click the ad aren't the problem. People who click the ad and don't convert are the problem.

Ryan Jones
Mar 14, 2013

Did you even read the story? EBay said that ads for the term "ebay" that were clearly navigational didn't work for them in paid search.  Your examples are non-branded terms.  Two totally different things.  

Elisa Gabbert
Mar 14, 2013

The report did NOT cover only branded keywords: "Second, the effectiveness of SEM for non-branded keywords is small for a large and well-known brand like eBay and the channel has been ineffective on average." This is in the 2nd paragraph of the introduction. See for yourself: http://conference.nber.org/confer/2013/EoDs13/Tadelis.pdf

Joe Preston
Mar 14, 2013

While I'm familiar with eBay's firehose techniques with Adwords, competed against them for years, I don't think their poor strategy has much to do with their published paper.Ebay asserts that branded search terms aren't incremental when you have organic position on those queries. Of course, ebay may find that their branded queries are more expensive than a lot of other people's branded queries, because google is going to suggest "ebay shoes" to other people trying to buy shoe keywords. But, in any event, their study showed that when they had a high ranking organic result for "ebay <whatever>", it didn't move the needle to have the ad appearing on that page.Second, they showed that their non-branded sales were really really low. I'm not sure why it took them a decade to figure this out, but they weren't talking about the efficiency of their buy on non-branded, but all they are saying is that the sum total of their non-branded ads are negligible in terms of producing conversions. This wasn't about the efficiency of the non-branded ads on ROI or ROAS, this was simply about how many transactions the advertising produced. I have always assumed, as you did, that eBay's advertising was monstrously inefficient, and they disclose in the paper, that they manage a portfolio of 170 million keywords, so no wonder. But, I have also always assumed that they negotiate some kind of bulk rate with Google on the DKI advertising they do, since it's a very large buy in what is, effectively, remnant space, so I suspect they have always been comfortable with the investment, which is why it took them 11 years to realize it was useless. 

Pages

Leave a Comment