Issue 11 – Google AdWords Q & A.
Question: My website already ranks well in natural search results, without my having to pay. Why should I bother with AdWords? Wouldn’t doing so be a waste of my money?
Dave: So-called natural search results are great, if or when you can get them. But there are still plenty of important reasons for sticking with AdWords. Here are five to get you started.
1 – AdWords gives you control. You decide exactly what is displayed, where it’s displayed, when it’s displayed, on which networks, how much you’re prepared to pay for it and where the visitor is taken when they click your ad.
2 – AdWords gives you speed. Getting new (or old) content indexed by the search engines can be a thankless task, sometimes even impossible. New websites may even take twelve months or longer to rank in the search results. AdWords, on the other hand, gets you up and running in minutes.
3 – Your high-ranking natural results position may have disappeared entirely by tomorrow. Not running AdWords would mean that your Google traffic vanishes overnight.
4 – Your ranking in the natural search results will vary widely, according to geographical location, language, browser and personal preferences. If you’re serious about wanting to sell to the world, you need to make sure that you reach them. And stay in control.
5 – If AdWords doesn’t work then your ads won’t be clicked. If they don’t get clicked, you won’t be charged. And never forget that if your ad isn’t there, they’ll be clicking on your competition.
Aaron: If you want your AdWords ad to be in first place, you are in complete control. You pick the keyword, the price you’re prepared to pay and you write the exact ad text, which should be far more captivating then a natural search result listing.
If, on the other hand, you want your natural search results to be in top position, this is no easy task. Natural search results are based on a large number of factors, including website content, how your page ranks amongst other pages with similar content, link popularity and more. You may also enjoy a top position which drives large amounts of traffic to your site, only to have it vanish without warning when Google update their index and ranking algorithms. Working yourself back to the top can be somewhere between extremely difficult and impossible. AdWords can be a more cost-effective route to the top.
Also don’t forget that Google want their visitors to click on the ads. A lot of time and money has been spent on making sure that they’re clicked, including highlighting some of the top performing ads with a (new) different colour background and putting them at the very top of the page.
Question: How do you decide which ads to keep and which to delete? Do you decide based on the CTR * conversion rate, or something else?
Andy Brice – http://www.perfecttableplan.com
Dave: The blessing of AdWords is that you have so much information at your disposal. The curse of AdWords is that you have too much information at your disposal.
The classic mistake is to choose the low performers based on CTR alone. This is considerably less than half the equation.
You need to take into account the ad’s CTR, how it converts, how much you pay for it, and what else the visitor to your site may do after clicking the ad. Once you factor in the keyword’s impact on the equation as well, you start drowning in data at an alarming rate.
Realistically, your decision should be based on a limited number of factors. I usually look at visitor behaviour once the ad is clicked, the cost of bringing them in, CTR and cost per conversion.
I did try to work this out as a weighted formula but I think I passed out (or fell asleep) while trying.
Also don’t forget that your ad isn’t shown in isolation. Behavioural patterns change with time, as do the ads of your competition. So what works today may not tomorrow.
Aaron: In an ideal world, a successful ad is one that sends the most visitors to your website, at a very low price, who then purchase your products. Measuring or achieving that is very difficult. A high click through rate does not necessarily mean a successful ad. For example, if your ad says, “free plasma TV”, I suspect that your click through rate would be extremely high for all the wrong reasons. Unless, that is, you are actually giving away free plasma TVs. If you are, please contact me. [Or better still Dave].
As for working with the conversion rates, Google’s conversion tracking is extremely limited. For example, the person who clicked on your ad and downloaded the product may not be the person who makes the purchase. And even assuming that the downloader and purchaser are the same, Google’s conversion tracking is little more than a fairly basic cookie system. If the cookie cannot be set, the conversion occurs after 30 days of the ad click or the cookies are deleted, then the conversion is not counted.
You might argue that some data is better than none. But how do you know how accurate it is? Is their system catching 40% of the conversions or 90%?
Personally I like to compare the AdWords ad results against web logs. For example I check to see if traffic from the ads is venturing on throughout the website or exiting upon arrival.
It’s also important to remember that this is an ongoing process. Not only will you never have a perfect ad, but you also need to constantly create new ads and delete those that don’t perform well.
Question: I have an AdWords account that has languished, because I haven’t been willing to spend the time on it. I’m pretty sure I’m losing money (or more precisely, not making the money I could be) because of it.
I’m about ready to jump back into it, but I’m feeling a little overwhelmed. I’d like to manage my account myself, but I’m worried about the initial set-up – that I don’t have the expertise to do all the fine-level tweaking that could make a difference in setting up successful ads and campaigns.
I’d love to employ your company to set up my campaigns for me, but I don’t think I can afford it. So is there some middle ground available? Where I would set things up to the best of my ability, then have your company review my work and make recommendations?
Sue Pichota – http://www.icons-icons.com
Dave: The short answer is that our AdWords services aren’t cheap, and I can of course appreciate that spending over $1000 a month for an account to be managed may not be realistic for some.
We do, however, offer consultation services, currently priced at US $250 per hour. For most accounts, two or three hours should enough time for us to have a look round and send detailed instructions as to how it can be improved.
Details of our consultation package may be found at the following URL: http://www.softwarepromotions.com/consultation/
If you’re going to do the whole thing yourself, a good starting point is to sketch out how your campaigns and ad groups should be structured. Ideally. Once you’ve got a structure and setup you like, you then need to decide whether you can apply the changes to your existing setup, or whether it’s easier to delete everything and start from scratch.
If you’re going to fix what you have, then I strongly recommend using AdWords Editor. Shuffling keywords, moving ads and applying mass changes is infinitely easier and faster than doing so through your web browser.
Aaron: If you prefer to set up your account on your own, I would make the following suggestions:
• Check out Google’s AdWords learning center.
• Don’t mix the content network with Google search and search partners.
• Create more than one ad in each of your ad groups.
• Track your efforts, but don’t get bogged down in the data.
• Keep your campaigns, ad groups, keywords, ads and landing pages focused on the target.
• Don’t let your account work on autopilot.
Question: Lately Google’s been driving me crazy with the “Keyword Inactive for Search” thing. I read all their hints, what goes into the Quality Rating and all the BS I could find on their site. Modified my ads re: their suggestions and for a while all was well.
My prime keyword is “flashcards”. After the modifications above they rated it OK and suggested a minimum bid of .20, which I went with. It had previously been disabled, even though I was bidding .30.
“Great”, I thought, “all this will be worthwhile and even save me a little money”. But NO… within a couple of days “flashcards” was again “Inactive for Search” with a bid of .40 required to turn it on. I queried Google, but they just pointed out that the Quality rating was Poor…. Well, duh, WHY is it poor? I followed all the guidelines. BTW, I have content match turned off across the board.
Dick Bryant – http://www.openwindow.com
Dave: I can’t help but admire the whole concept of Quality Score. It’s little short of brilliant.
The AdWords system involves a careful balance of Google’s two main priorities; relevance and revenue. If enough of Google’s visitors search for a term, click on your ads and stay there, the keyword will be classed as relevant. If, on the other hand, they don’t stay on your site, but instead hit the back button in their browser, you’re classed as less relevant, and so you have to pay more.
Google give you a massive amount of information from within the AdWords account, but there’s also a lot they don’t show you. For example, when people click on your ads, Google know how many of them stay on your website and how many return to Google. They also know how long they spend there before returning.
Wouldn’t that be useful information to have?
The obvious question is why don’t Google provide you with this? Call me a conspiracy theorist if you must, but I don’t think it’s in their interests to. After all, if one of your ads was getting 100 clicks a day at $0.20 per click, and all 100 visitors went back to Google after clicking your ad, would you carry on paying for it?
In your situation, Google have deemed the keyword to have low relevance. Chances are they’re right. You can either set up a landing page that is optimised for this term, decide that the keyword really isn’t a good fit, or keep paying more for the keyword. I don’t recommend the third option.
Aaron: In my experience, whenever I see the “Inactive for search” message, it is usually down to a keyword not fitting the ad group, ads and/or landing page. The keyword “flashcards” is a very broad term. Anyone who searches for “free printable flashcards”, “number flashcards” and many more phrases would see your ads. Is your product relevant to all of them?
In your case, it would appear that Google do not feel that your ads or landing page are relevant to most people searching for “flashcards”. Pay them a little more and they’ll indulge you.
If you’re not prepared to raise your bids, then your only other options are to improve your ads, your landing page or both.
Question: Are results for keywords using Wordtracker as an example, which to me seems to be targeted at Search Engine Optimisation, automatically the best result for adwords?
Peter Muys – http://www.brainstreamer.com
Dave: The idea of services and tools like WordTracker is to find the keywords that people interested in your software are using for the searches. The critical point is that you can’t predict what people may be searching for, but chances are that they’re looking for different phrases than you might expect.
Whether you’re using keyword research for SEO or Google AdWords, these services are ideal. The only real difference is that you’re not as concerned by the “extra” data. All you need are the relevant keywords.
Aaron: I use Wordtracker for AdWords keyword research. While your AdWords account does provide a keyword tool, I find it to be as good as useless. It’s highly inaccurate and often gives a lot of irrelevant results.
Question: I heard Dave Collins speak at the Shareware Convention last year. I remember he said to monitor the Adwords campaigns and that it was a mistake to set them up and not monitor them. The problem I have is that I am monitoring my campaign. But I cannot seem to get definitive data to direct my actions.
For example, I’ll receive reports that indicate one ad outperforms the other ads. So I’ll think maybe I should get rid of the other ads and use the higher performing ad. But then I’ll receive reports showing the higher performing ad performed worse than the other ads. I cannot get consistent data to draw any conclusions.
I also have the same ads going to different landing pages. One week, the ads will perform better with a Google-specific landing page. But on different weeks, the same ad will perform better going to the home page.
Please help! Do you have any suggestions on how to make sense of it and direct my actions to get more consistent data?
Kim Murdock – http://www.magsoftwrx.com
Dave: This is quite a common situation. The trends in your actual data will change on an ongoing basis, as the people searching, their search terms and search behaviour patterns will be in constant flux.
There are a number of steps that you can take to make sense of all the data.
First of all, make sure that you’re looking at a big enough data sample. One day’s data is never enough, and at the very minimum you should be giving your ads and keywords a full week to generate data. Sometimes longer, depending on the volume of impressions and clicks you receive.
Secondly, when you’re looking at data, look in units of seven days (7, 14, 21 etc) so as to make sure you get an accurate sample. Nine days could include one weekend or two, which can have a major impact on what you’re looking at.
Thirdly, make sure that you allow enough time for your changes to really kick in. New ads, for example, can sometimes take a long time to be approved for the content network. Making alterations seven days after adding new ads may not be anywhere near long enough.
Fourthly, don’t alter too much at the same time. If, for example, you log into your account today and add new keywords, new ads, different bids, different budgets and point to new landing pages, you won’t be able to point to the reason for your success or failure in a few weeks’ time. The AdWords system is complicated enough without running in circles, chasing your own tail.
And finally, the murkier the results, the more time you need to look at. A one week pattern may be completely misleading, but if you look at data for the past two or three months, you should get a better idea of what’s working and what isn’t.
Aaron: I think you are going to have a problem finding consistent data. The AdWords traffic is constantly changing. What worked this week, might not work next week or next month. And something that failed this week might actually work next week or next month.
What is important is to always experiment by optimizing what you have. Look at the last 7 days; look at the last 30 days. Purge what is not working and expand on what is working.
It’s also important not to get lost in the data and keep things in perspective.
Question: What percentage of clicks do you think are fraudulent? How much do you think the average small developer spends on wasted clicks?
Sharon Housley – http://www.notepage.net
Dave: This is an impossible question to answer with any accuracy. Generally speaking, I think that most accounts will experience some level of click fraud, but I assume that isn’t quite as precise an answer as you hoped for!
Google’s approach to handling click fraud is of critical importance to their profits. Being seen as indifferent and dismissive risks advertisers being scared away. Being overly active might also have the same effect, by drawing attention to something that the advertisers haven’t (yet) worried about.
The bottom line is that Google are undoubtedly aware of how widespread click fraud could become, and how damaging that could be for them. Because of this, they are definitely taking active steps to detect and prevent it.
If advertisers lose faith in the AdWords system, they’ll move away en-masse, and I believe that a serious loss in confidence could destroy Google in a matter of months. Ultimately, they have a lot more to lose from click fraud than most of their advertisers.
It’s also worth mentioning that I’ve usually been very impressed at how Google respond to claims of fraud. I have presented them with irrefutable evidence that click fraud took place for one of our clients, and to their credit, they eventually agreed and compensated us more than reasonably.
The bottom line: keep an eye on it, but sleep well at night. Google are doing the same, on both counts.
Question(s): (a) Currently I only run my Adwords in English speaking countries. Do you recommend branching out further and, if so, which countries are amenable to support only in the English language?
(b) I currently split my ads into separate campaigns for U.S. and non-U.S.
English speaking countries. I have found that ads that work well for one don’t always work well for the other. Given question (a), are there further geographic regions you recommend splitting the campaigns across?
(c) I just signed up for Google Checkout and now have the little Checkout logo beneath my Adwords listings. Does it improve or deter click-throughs?
(d): What are the factors you consider when deciding whether to create a new Group or a new Campaign?
Nicholas Hebb – http://www.breezetree.com
Dave: I don’t know which countries are amenable to English language only support, but if a person is searching in English, the odds are probably in your favour. Choose the countries, select your ads to only display in English, and let Google handle the delivery complications.
As to whether you should expand into non-English speaking countries, why not? Our company only works in English, yet we have worked with companies from more than 30 different countries. If there are people in non-English countries searching for what you sell, in English, then why not give it to them? Try it, track your results and decide how to proceed.
As for how to split the campaigns, there are an almost unlimited number of possibilities. The main issue, as always, is time. Bearing in mind you want to track everything you do, try to avoid setting yourself up for data overload.
As for the Google Checkout logo? I have no idea. If it attracts attention then why not? But the advantage (if there is one) may be quite short lived. As an AdWords tactic, it if works, I predict an increasing number of people signing up for Google checkout, which may result in many (or even most) ads soon having that logo. You have to hand it to Google.
To answer your final question, campaigns give far more control than ad groups, but handling twenty campaigns for forty keywords would be somewhat time consuming.
Consider the analogy of a filing cabinet. For obvious reasons you wouldn’t want a separate drawer for every bank account, utility bill, licence, warranty, insurance policy etc. Nor would you want “anything to do with money” in one single folder.
Where budgetary control is needed you want a campaign. Where streamlining is needed, ad groups are the better option. Pick and choose as you see fit, according to your criteria.
Aaron: For questions (a) and (b), I think this boils down to time. In an ideal world, it would be great to have a campaign set up for each language, and each country, and then each of these would be further split into three more campaigns; one for Google search, one for the search partners and one for the content network.
You could go even further with an ad group for each keyword, where ads would be based on that one single keyword. But the time and effort that you would need to manage this would be incredible.
I would be surprised to find that people in Germany, France or China do not perform searches for flowcharting software in English. If your ads are not being displayed to those countries, you may be missing out on sales opportunities.
For your question on when to make a campaign and when to make an ad group, it’s all about control and time. Campaigns give you all the control, while ad groups offer flexibility through shared control. At the campaign level, you can choose your budget, the languages, geographic region, the various networks and more. At the ad group level, you share with all other groups within that campaign.
Let’s say you have one campaign where you have chosen the Google search, search partners and content network. You will not be able to focus 75% of your daily budget on the Google search and search partners and have only 25% of the daily budget for the content network. The only way to set separate budgets would be to have separate campaigns for each.
The problem with separation is time. The more campaigns, the more ad groups, and the more time it takes to manage your account. Finding the right balance is key.
ON A LIGHTER NOTE:
Those of you who actually read this part of the newsletter may remember that my wife and I gave birth to our first baby at the start of this year. You could argue that Karin did the “giving birth” bit more than me, but let’s not nitpick.
Personalisation is supposed to be good for business, so here are a few facts about our little girl:
(1) She had her very own licensed version of Office 2007 months before it was released. Karin always rolls her eyes when I mention this to people, which I admit may be more often than is healthy. The midwife really didn’t seem to care.
(2) Whenever I sit her down in front of my PC she waits for around 30 seconds before crying. Or screaming, Or both. It doesn’t bode well for our expansion plans. Her typing skills aren’t great either.
(3) She obviously has her own domain name, but so far has chosen not to use it.
(4) She has her heart set on my Nikon D70. Who am I to refuse her? I guess I’ll just have to replace it with a D200.
(5) She has already earned money, despite being less than four months old. I’m not convinced she’s profit making though.
The Competitive Edge newsletter is a monthly in-depth look at the issues faced by independent software developers today.
Common Sense Disclaimer: The advice, views, experiences, opinions and other information shared in the Competitive Edge newsletter are researched to the best of our ability. At the time of publication, we believe them to be correct.
SoftwarePromotions can not and will not guarantee that any of the information here is accurate, and/or will work for your company. In plain English, use this information at your own risk.
All information herein is offered “as is” and without warranty of any kind.
To the greatest extent allowed by law, SoftwarePromotions, nor its employees nor contributors, are not responsible for any loss, injury, or damage, direct or consequential, resulting from your choosing to use any of the information and/or advice presented here.