I have to admit, I don’t know what Seth is talking about. Again, another cool innovation not released in Australia.
It’s 1am on Saturday morning and I’m going to bed, can anyone email me a link or description of this? I am too tired to look for myself. ssebastian@theoddoneout.com.au
Thanks
It’s really easy to make mistakes in AdWords and get disheartened, especially if you’re time starved or don’t quite understand how it works. Today I’ll list the top 5 mistakes that I see
1) Negative Keywords
Negative Keywords are a powerful tool when utilised correctly, when not, they are all but worthless. The key to using negative keywords correctly is from Analytics with the search query report. Find terms where people clicked on your ad and they were obviously looking for something else. You can’t just make the entire term negative, there will be some word in there you want your ads to come up for, or else it wouldn’t be in your keyword list. What tells you they didn’t want your site? That’s what you make a negative keyword.
2) Avoiding Maintenance
Really, what we make the first time is never perfect. Don’t avoid going back and making adjustments to make your campaigns better fit the user behaviour and desires shown in the stats. It will make it more effective and cheapers.
3) Measuring Costs
Well, I want you to do that (and so do you) but don’t forget to measure outcomes as well! The best outcome to measure is the outcome you want, cash in your hand, spread of your idea, subscribers to your tribe. Whatever that is, measure it too.
4) Single AdGroups
It’s almost 2009… There is no excuse for using a single AdGroup with all your keywords dumped together in it. Make a commitment, fix it by January 1.
5) It will happen overnight… Right?
Sadly, no. No matter how good you are (or the person you hire is) it will not give perfect results overnight. Aside from imperfect knowledge of searcher behaviour, the results (cost, ad position etc.) all have historical data taken into consideration. I’m sorry, it takes at least 3 months to get to a stable equilibrium of cost and ad positioning.
If you’re making any of these mistakes, the time to stop is now.
Yahoo! is circling the drain and Microsoft is refusing to buy it. So where is the future competition for Google?
I think it is not in search. The difficulty of making something remarkable enough to train hundreds of millions of internet users to use something other than Google is too high. Yahoo! may have been able to manage it, but I doubt it and they certainly cannot do it now. Microsoft is too corporate, too bureaucratic and (the biggest problem of all) assumes that people are morons, it oversimplifies things and stops us from getting what we really want.
Google is simply too rich, too big, too profitable not to buy any startup that looks like it might have something good in the search department. Competition with Google will not come from search.
So where?
One of the main problems with search is that it cannot handle complex inquiries, if you want a page on laptop bags, it will give you that wonderfully, but if you want a page that asks what the prettiest laptop bag is, you have to search through hundreds and hundreds of pages to get what you want. Seth Godin thinks there is a market for user ranked and generated interpretation of search for complex inquiries.
I don’t know if that can ever compete with Google for online advertising, but it’s looking more promising than anything else I’m going to mention. The rest are what not to do, where the competition will not come from.
It won’t be a pop-up, pop-under or anything that distracts and annoys the user. It won’t come from ISP’s, individual sites or even collaborations of sites. Nothing on the internet is so good that we as users need to put up with crap like that any more.
It won’t come from an individual site, there is no site that is universal enough to be more profitable to build their own advertising platform instead of using adwords (except for pornography, but they have their own advertising aggregators and most other businesses wouldn’t want to be in on that).
It might be a collaboration of sites, dozens of popular sites with frequent content updates, (rudius media (safe for work, but only just) springs to mind). People go to these sites to read content though, it’s just a copy of AdWords’ content network, limited to a few sites, that may be profitable, but it will never compete with Google.
I don’t know where the next online advertising competition will come from, if I did, I’d be building it and ordering my Lamborghini. I’d love to see examples of what you think is great online advertising models and methods, let me know.
Wow. Another big (but hardly surprising) development.
Google has walked away from the controversial deal of selling ads in Yahoo’s search results. It’s interesting. If I was a cynic, I’d suggest that they walked away because in the long term they’ll have the same market share as if they had signed the deal.
So here’s the cynicism. Microsoft bids on Yahoo, Yahoo fights back, asks for too much and Microsoft withdraws the bid. Google seeing Yahoo with a shortage of cash and the expected influx from Microsoft non-existent walks up and gets the deal going, delaying the deal for the US DoJ to inspect. In the 6-8 months of delays Google
While Yahoo does…. Nothing. The incompetent management (everyone good bled out in the Microsoft deal) sits on their arses and waits for Google’s money. Google seeing the competitive environment changed (incrementally, but significantly) realises that Yahoo is dead-in-the-water and that they will have Yahoo’s search share, sooner rather than later, and pulls out of the deal. Leaving Yahoo alone, in poor hands and desperate to sell up.
But that’s the cynicism.
I don’t know any other explanation for it, but one big questions hangs in my mind. It’s one for the textbooks I think, an example to sit next to Nike and the sweatshops.
Why Did Yahoo!’s Management Do Nothing? Why did they sit there and twiddle their thumbs hoping the deal would come off? Is this the best contingency plan they could come up with?
Damn.
All in all I feel sorry for Yahoo’s staff. I wish the best to them and hope they get to join some other company where their skills will be put to great use and they leaed happy and successful lives.
Yahoo! is dead. The lesson: Contingency plans, have one.
My bet is Microsoft will buy Yahoo, but for a lot less than the offer that was withdrawn. Your opinions?
Wowee… What a fortnight.
Firstly I disappear for two weeks, with no explanation or updates (unless people are following me on twitter). Secondly some big changes have occurred within AdWords in the US, but not yet released in Australia. Thirdly, my girlfriend has submitted her honours thesis in Law at the Australian National University. Not relevant for any readers, but I am so proud of her I might burst, so I have to say something.
I might explain my 2 week absence by saying I’ve been workig on another project, which was finalised yesterday and we hope that it all goes well. It has nothing to do with AdWords, or the internet at all, so I’d rather not discuss it here.
This being a blog about AdWords, I think I might skip the third point and focus on the second.
Universal search and Adwords combined. Wow. That’s quite big. I haven’t had a chance to play with it yet because it’s not released for Australia and none of my US clients would benefit from it.
I can see one argument against it (aside from the generic “something changed and it ruined my business model”) and plenty for it. I’ll start with the negative.
Spam. The bane of the internet. With badly targeted universal search ads it is very, very possible that searchers may end up with completely irrelevant ads that fill a lot more of their screen than currently. This is a Bad Thing ™.
Positives. More relevancy. A client in Australia runs a location targeted campaign to promote a bricks and mortar business. We’ve theorise that most of the bounce rate is from searchers who are looking for something location based, adding a map to the ads for the business would, assuming the theory is correct, give the searcher more information, enabling self selection of clicks, so the searchers looking for a business in a different geographic region will not click on the ad. This is a plus.
Improved search quality. Ok, it conflicts with the spam conjecture, but this is Google. Their business (and huge profits) are only because they give the best search results to users. Google need to continue this and will only implement something if it allows better search results too users.
Lastly, improved CTR’s (for good ads). This is a Good Thing ™ for advertisers and for Google in general. Improved CTR’s give Google more money (on average) per search term, and give advertisers who are relevant a big boost in results.
Overall, I think we can discount the spam complaint in the long term, and ignore the loud, yelling screaming forum posts about universal search adwords destroying businesses (not because I think destroying businesses is a good thing, but I think it will only destroy irrelevant businesses. Self selection is a great thing) and it will lead to greater long term success, for good advertisers and for Google.
I’m excited about it, and can’t wait to get started.
Now on another note, I’m thinking about this blog. Black with white text is nice enough, but from a design perspective… it sucks. It doesn’t look professional. I’m going to be experimenting with some different designs. If you see something you like (or don’t like) I’d love the feedback. Also, I promise not to be absent for long periods again. This project took too much out of me.
Thanks for being patient.
There are two wrongs to be righted today (three if you count my grammar) and all are my fault. The first is my lack of posts for the last two days, I’d love to tell you it was something exciting but, quite simply, it was not. I was writing other work until the wee hours of the morning and was too tired. I apologise (but I can’t promise it won’t happen again).
The second wrong is also my fault. I have simply not written enough about the reason I started this blog, to try to convince people the best strategy to get cheaper CPC’s and higher QS’s is to have tight keyword clusters.
So today I’ll tell you a story with some client data (names, keywords and industries changed to protect the innocent, but if she’s been reading my reports, they’ll know it’s them).
3 months ago I was contacted and had a discussion with a person who owns a very promising company, doing some remarkable work but neither the time nor experience to implement and run a campaign that does the company or service justice. It was agreed that I would work with them and this is what happened.
The original campaign was run in 2 countries with 2 campaigns and 4 ad groups and 4401 keywords, 3 adgroups in 1 campaign and 1 in the second (smaller) campaign with an average CPC of $2.07. The service offered is quite complicated and 4 ad groups is unable to communicate with all needs and desires of their customers.
4401 keywords is very broad, but gave great insight into the targeted categories of their customer, and researching the provided competitor websites and their own helped to distinguish what they did that their competitors did not (and vice versa) as well enhancing the insight. Google Analytics and AdWords data helped to make sense of everything.
STOP
At this point, there was heaps of data, quantitative and qualitative, all about their customer, competitors, what works and what doesn’t. There’s a big temptation to look at it and pull your hair out and this is the hardest part of the entire job. Don’t despair, work at it over time, take a break to run around in the sun for a while. I promise you, the rest is easy once you get past this point.
RESUME
So with this information we now get to categorise. Categorise what? It depends on the customer. At this point everything is up in the air, you might want to categorise by product offering (I’m looking at you retailers) or by the emotions and feel of the customer (car manufacturers and luxury goods) or the problem you’re trying to solve for your customer (B2B services and tradeswomen). It depends on what the data is telling you (and because some of the data is qualitative, your intuition helps).
This clients information was categorised by problem solved and by the feel of the customer. The existing keywords helped to make this decision as there were little pockets of high click through rate keywords and high content network clicks in each adgroup which were similar.
The account ended up with 1 campaign targeting both geographic regions (a mistake, I’ll explain why later) and 6 ad groups, each with their own little cluster and ads specifically written for the adgroup (which were approved on first application! I’m very proud of that).
Fast Forward 1 Week
All is going well. CTR’s are up to 0.82% and the CPC’s were down to $1.93. Wait! That’s not very good at all, I mean sure, it’s better than it was, but I’m a professional, I charge money for what I do. A 10% decrease in price is not acceptable. What had happened? What went wrong? All CTR’s were up, content click costs were well down. I must have missed something.
I had. One geographic region was much cheaper than another. I had missed it, a most amateur mistake. Ughh. Still, 10 minutes later and the campaign was replicated and seperated geographically.
Fast Forward 1 Month
Ah, now it’s brilliant. CTR’s are up to 1.35% and CPC’s down to $1.79. That’s better.
Current Situation
With minor tweaking to the campaign, rewriting some ads and tightening ad groups the results are even better still. CTR’s are hovering above 2% and CPC’s are down to $1.05 (and heading for under $1! – Yes). Brilliant work.
Why?
So, the reason this worked so well is solely because the keyword pockets have ben branched out into their own little adgroups with ads written only for them and targeted at the specific user. You need to do the same thing with your campaign, it takes a lot of time and effort, but how much better would your business be if CPC’s halved in 3 months?
That same $150 per day will get you twice the clicks. It’s worth the effort, but more than that, it’s the only way to do it (although, I’m fairly sure if you seduced Matt Cutts… *shudder*).
*Disclaimer: Matt Cutts is a brilliant guy, incredibly smart and happily married. I love his blog and and… don’t delist me Matt. Please.
Fabulous Freeform Fridays are your opportunity to ask me to clarify something I’ve written about, write something on a topic you want to read about or just plain ask me some questions. It’s late today because it’s a long and complicated question. I hope I’ve covered it well enough, but it’s something I’ve had to think about a lot. I welcome your feedback, thoughts or disagreements. Email or in the comments.
Today’s Fabulous Freeform Friday was suggested by Rebecca McMahon.
Hi Sebastian,
I know Seasonality exists in my business [...] How do I take advantage of this?
Hi Rebecca, Your question touches one of my favourite concepts, and one that I loathe for being out of date. Unfortunately my favourite one is complicated, technical mathematically and boring to most people, so I’ll be brief on that and spend a bit of time deconstructing the concept I hate. I’ll start with what I like.
Forecasting
Wow, someone likes forecasting… I am such a nerd. There are 2 ways to do it, empirically with lots of numbers, which I won’t cover because it deserves an entire blog on it’s own (if you know a good one, send it to me!). What I will talk about, briefly, is the forecasting based on gut and experience.
You know there’s more sales in December and that January is slower. You know March and April is a great time for chocolate. You know this, because you’ve seen it. You know your business, your industry, you know when it is busier and when it’s not.
Write it down.
That’s forecasting by gut. It gets better, more accurate when more people get together, but also more open to bias.
It tells you when some products are more effective, but how do you take advantage of that? Again, it depends on your product and consumer, but the vast majority follow this sort of trend
Attention – Consumer notices your brand or product
Interest – Consumer gets interested in your brand or product
Desire – Consumer wants your product
Action – Consumer buys your product
This happens over time… except it’s not quite right. Seth talks about flipping the funnel, I’m going to change the funnel (not to say I disagree with Seth, he is, as always, correct. I’m merely changing the interpretation).
Problem - The consumer exists with the problem in a semi-equilibrium. They may be aware of the problem or not.
Problem Recognition – The consumer recognises the problem, conducts a cost/benefit analysis (costs and benefits may be fiscal, they may not) and decides whether to confront the problem.
Solution Search – The consumer searches for the solution
Desire - They find a (hopefully your) solution that they like
Action - They buy a (hopefully your) solution
When Seth says flipping the funnel, he is talking about getting customers to recognise problems in other, non-customers and providing a solution and desire (the problem may be as simple as the non-customer doesn’t have an iPhone, or Little Miss-Matched socks).
Answering the question
The thing with the funnel, flipped or not, is that it takes time. There’s no point trying to get people to recogise their problem, or getting existing customers to point out other people’s problems (which you have solutions to) the week you have forecasted sales to go up. It doesn’t work.
You need to prepare in advance, either with AdWords advertising or whatever form you use. It takes time, to build the relationship, to let consumers search for the solution, ask their friends or even recognise a problem.
How long depends on your industry, I hope you know how long. If you don’t, do some research now.
Hope this helps everyone and again, I welcome your comments (or to suggest a topic for next week).
Continuing in my theme from Monday’s post.
Adaptation is a fairly simple concept, an ad is doing really well in one segment (normally geographic location), lets transfer the ad to another, but make slight tweaks to suit that market better.
Not hard, the problem comes in the tweaks. 2 lately have been really annoying me, both average ads for average brands targeting the average male (I’m sure they would disagree…) I haven’t been able to find the originals, nor the adaptations online, so you’ll have to take my word for it.
The first is an ad for a Ryobi electric tool range that uses one battery. Kind of a boring subject I know. The second is for a Castrol oil that supposedly lasts 45% longer at maximum horsepower (the kind of claim that is so ludicrous that any respectable engine-head knows cannot be true without cheating the test, and non engine-heads don’t care about).
Both of these ads were launched for the US. You can tell because the imagery is US-focused (Pontiac cars and badges, generic huge drink, “American Dream” houses). The accents the actors have are Australian. So far, so good, a successful adaptation. Unfortunately, they’re dubbed. Badly.
I don’t like a non-Australian accents in ads, it’s a personal bias and most definitely product specific (tourism etc. get away with it, along with many others. Average products don’t). I think it’s a form of patriotism, or my mother complained about American accents in ads too much (I also don’t like strine (crocodile dundee/croc hunter/Australian sportsperson accents) either. That’s because it makes Australians sound uneducated).
I got off point. I meant to say I don’t like non-Australian accents in ads, but the only thing worse than that is obviously pandering to the target audience. In the ads above, it’s almost like someone at an advertising agency read a little bit about adaptation and decided to do it, without considering the quality of the ad.
In short, do it right, or don’t do it (sound general business advice in my mind).
So… AdWords. Adaptation?
We’re lucky. We have great segmentation (by location) and with Google’s Ad Planner (still in beta last I looked) some pretty good segmentation by demographics in the content network. This gives us data, data is great!
This data means we know how our ads are going and can test adaptations by each segment and use ones that work! There’s no fear of your ads showing up in a blog somewhere as an example of what not to do!
I’m deviating slightly from my usual topic of AdWords this week, I’ll be touching very briefly on some thoughts I’ve been having on everyday marketing. Hopefully I can project something back to AdWords, but it might not always happen.
So today I want to talk about mass marketing, I can hear the screams now. I promise I’m not getting all Ogilvy-like on you, I think mass marketing is dead. However, watching the Formula One over the weekend got me thinking about Vodafone (and Red Bull, but I’m a Hamilton fan).
Why on earth do Vodafone sponsor so many sports events, and motorsport in particular? What do a bunch of atheletes and cars offer to Vodafone?
Seth has been writing about the death of the mass market for a long time (and so has everyone else). I ask if sport is the last, large barrier remaining before the grey haired monstrosity that brought average to everyone goes away?
Think about it, most sport is self selecting. I like freeride mountain biking, (unforunately, not me) but it’s not popular and could never be described as a mass market. But some sport, some sport is universal. We may not like it, but if we watch tv, read the paper (even online) we know what’s happening and we have some exposure to the brand who sponsors it. In Australia it’s cricket and rugby league.
So is sponsorship the last remaining mass market? If it is, how long until we, the consuming and ad avoiding public, break that last barrier?
What’s this? It’s a Google Trends search volume query for Lenovo. There’s an obvious jump in August… Did something happen then? I think something did, I can’t really remember what though… 08.08.08 comes to mind for some reason.
Ok, so the Olympics. Biggest sporting event on the calender. Lenovo won’t release figures, but they spent at least $80, probably more than $100 million US from the research I have done. I hope they’re getting that much worth out of it.
Lenovo isn’t sponsoring any more Olympics. The London Olympics only have two thirds of the sponsors that Beijing had, are we coming to the end already?
So it’s not all doom and gloom. Sport is a fantastic social object and, for the foreseeable future, will have access to most people. I want to know what’s going on with my team, my sport, my player, my racer. In doing so I will always see Fiat, Marlboro, Vodafone, Fox and all the other brands they are sponsored by.
I think there will come a time when the sponsors can isolate and measure the return of their sponsorship. By the time that happens, I hope the sports I like are self financing. Yes, for now, sport is the last frontier for the mass market. The time is limited and I hope they’re planning for the break.
How does this affect us as AdWords advertisers? Not many of us sponsor a sports team (not many… ha, try essentially none). I guess it doesn’t.
Share this post!
Fabulous Freeform Fridays are your opportunity to ask me to clarify something I’ve written about, write something on a topic you want to read about or just plain ask me some questions.
Today’s Fabulous Freeform Friday was suggested by Jonothan Wright.
Is the long tail worth targeting?
Interesting question that many people just assume rather than looking at the data. I’m saying yes, but I think Jonothan wanted more than just my opinion.
The start of the official thought of the long tail (and the term itself) is really recent, just October 2004 by Chris Anderson in Wired Magazine it’s a relatively long article, but should be required reading for all business people, and particularly online and marketing types.
The long tail is that bit at the end of the graph, the yellow piece on this chart
(from wikipedia)
Green is your generic terms “Car” “Boat” etc. whereas yellow is “inlet manifold 1964 corolla” car is going to get a lot more traffic, but if you have an ad for “inlet manifold 1964 corolla” and someone searches it, you’ve got a sale.
Not a lot of people search long tail terms, that’s the point of them. Some businesses make a fortune from long tail terms (rumours about Amazons long tail range from 25% to 45% of total sales) so there is definitely potential.
What does it mean for you?
Well the yellow section of that graph just keeps going on and on and on and with two big changes to the world (ok, our world) the green section is peaking earlier and the yellow section gets fatter and fatter, meaning the yellow section actually contains more than the green.
The two changes
1. The end of the mass market
2. Search users becoming more sophisticated
With these two changes, more people are searching more long tail keywords, meaning more value from the time you put into making the long tail campaign.
The main benefit to the long tail is its price. If you had to pay the same for a conversion on “car” as on “inlet manifold 1964 corolla” it’s a relatively easy choice to ignore the long tail keywords because of the extra work they require.
It is extremely unlikely for that to ever happen however.
So here’s what to do for long tails in AdWords (and it’s boring, sorry). What we need to do is search as many long tail terms as we can think of that are relevant to us, and see what ads show. This tells us how many competitors, how well they’re using their AdWords and how many are using the long tail (crafting ads specifically for that search or using their generic ads)
This gives us an indication of what we need to do to target the long tail.
If they’re using targeted ads, get ready for a long night or two writing ads and crafting adgroups for each tiny little cluster of keywords. Be prepared to run search query reports weekly and update the ads each time.
If they aren’t using targeted ads, you can get away with being slightly more generic and targeting “parts for old cars” instead of “inlet manifold 1964 corolla”.
The trick to using it is to be more targeted than your competition.
Of course, if you have no competition, it’s really easy.
Fabulous Freeform Fridays are your opportunity to ask questions. Something you want to know, want to hear my opinions or even have a discussion started? It’s all in your hands, post a comment on this (or any other) post if you have something you want to talk about or you can email me.
Share this post!
