Monthly Archives: February 2012

We’re probably all guilty of being vague at some point of our lives. Whether it was an explanation to our wives about where we were the night before or a story told in as little detail as possible that leaves everyone confused. In most cases, you can chalk it up to not knowing any better. But it’s not about our mistakes; It’s about how we analyze and correct them that counts.

This same rationale goes for online marketing. When you are too vague in your message, it can easily be misinterpreted. Let’s say you sell collectibles through your online store. Generally speaking, people looking for collectibles are not going to type “collectibles” into the search bar. In most cases, the searcher has a good idea of what they are after. Now don’t get me wrong, the searcher probably isn’t going to search “1966 Vintage Simms Batmobile” either, but that is why you need to meet them halfway. The person searching probably understands that if they are too specific, they won’t be able to find what they are looking for. The same goes for being too vague. This problem could be alleviated by the “two-word keyword” approach. Often times, a one-word keyword is far too general, as in the “collectibles” example above. But when you add descriptors to it, as in “batman collectibles”, it becomes a much better keyword. For example, some better options for keywords, for both the advertiser and the searcher, would be “batman collectibles”, “vintage batman”, “batman toys”, etc. Even though they are pretty general, they still have a central focus.

With broad, one-word keywords (not broad as in the match type but keywords that are just too general), you will still garner clicks from advertising off words like “collectibles”. That doesn’t mean that they will be good clicks. The difference between a good and bad click is night and day. That potential customer that just clicked on your ad can (and will) leave as quickly as they got there if they don’t find what they were looking for. Not only did you not make a customer out of them, but it cost you money for the click. It’s a lose/lose situation.

So even if you have been vague in the past, it is never too late to make a change. Start marketing more effectively today!

Installing SiteWit analytics script on every page of your site is the first step in measuring your visitor’s origination, behavior, and predictability. The tool becomes even more valuable as you set up values and record goals.

When you use SiteWit’s goal tracking feature you’re given a block of script to put on your goal page that looks like this:

This script works great on most html pages, and it’s self-contained with the tracking codes, so it works even on pages that don’t have the tracking code installed. But there are a few cases where you may not be able to put this block of code directly on the page. The cases are: AJAX calls, as well as iframes, hyperlinks, or form posts to 3rd party pages which cannot have the script installed. So here are some options.

AJAX calls

Assuming you already have the tracking script installed on the page, you can simply call sw.set_goal(GOAL_ID); and sw.register_page_view(); with the appropriate goal id at the time of the AJAX call, or better yet at the time of the “on success” callback. Here’s a simple example using jQuery:

Iframes

What happens though when the goal itself is not in a page you control? A lot of websites use iframes to host third party content and/or features like contact forms. Unfortunately, if the page is on a different domain, then most browsers will try to block cross domain communication. One way to get around this is to count the number of times the iframe has loaded and trigger a goal when it has been loaded x number of times (which is presumably how many clicks it takes to accomplish the goal on the remote site). Here is an example of how to achieve this:

This solution is best of course when there are a known number of page clicks to reach the goal (like in a wizard or form environment), and even then there is no way to tell if the user has gotten a failure message instead of the expected goal success. If the iframe contains a page with several success & failure paths or lots of links and/or post backs then the best option is to contact your vendor and have them install the actual goal script in its entirety at the point that the goal occurs.

Out Bound Hyperlinks

Outbound hyperlinks occur whenever your goal is someone visiting another website, opening a file (e.g. pdf, mp3, etc), or clicking a mailto: link. You will need to have the tracking script already installed on the page. All you need to do is add calls for sw.set_goal(GOAL_ID); and sw.register_page_view(); with the appropriate goal id in the onclick attribute of the hyperlink, like this:

Remember to but a ; between commands, if you have existing code in the onclick, put this code first (like in the example above, before the “return true;”.

Forms

If you are using a form that posts to itself and delivers a thank you or confirmation message, you can simply put the standard goal code with that message. As long as the message is not loaded with the initial form, it won’t be run until the form is submitted.

If you are posting to a page that is outside your control, then you will need to record the goal after all the validation has been completed, but before the form is sent off. In the case where your form is validated by calling a function from the onsubmit attribute of the form, the goal code be executed at the point where a the final return value would happen within the form, like so:

If you are using hooking into the event after the fact with something like the popular jQuery Validation plugin by Jörn Zaefferer, it would look something like this:

Hopefully this will give you or your developers some ideas on how to implement the goal code when faced with complicated scenarios. And of course if they ever need help, SiteWit support is waiting for their call.

First let me say that if you develop an application or an API and you would like to put restrictions on how it is used, that is your right. That being said, some requirements make little sense. Case in point is Google’s Required Minimum Functionality (RMF).

The idea here is that you can use one of their API’s, for example the Adwords API, as long as you utilize a minimum set of functionality that they think is most important. Seems like a fairly simple request right? Well, of the 153 features that Google touts on its list of Adwords functionality, 58% of them are required. So a majority of the available functionality is included in the required minimum. I would also like to point out that this API is not free to use. Companies pay for every call they make to it so shouldn’t they be given a choice as to what calls they make and what features they implement?

Now let’s be clear on what “required” means. It does not mean that your tool utilizes these functionalities to, let’s say, automatically optimize campaigns through the required features for those clients that do not want to manage every nuance of their online marketing campaigns. “Required” means the user must be able to find the feature easily and make changes to it as they see fit.

So if your businesses’ goal is to handle all those nuances behind the scenes and minimize the amount of work your clients need to do to their campaigns, then you are out of luck. If your goal is also to present a clean interface that does not overwhelm first time users, well, sorry, you’re out of luck there too.  I think the obvious use of an API should be to distill information in a different way for certain segments of people and allow them to interact with your service in a different, hopefully better, manner than they can through the interface offerings that already exist. Perhaps I’m wrong but it seems that requiring developers to basically recreate a tool you have already created is a bit short sighted.

With that being said, I don’t think Google is trying to be difficult.  I do think they need to rely on their API to allow other companies to bring them more clients from markets they have not yet tapped. I do believe all of the features listed in the RMF are important when running a successful internet marketing campaign. I just don’t believe that a company should be punished for creating an innovative way for people, who are new to internet advertising, to easily create campaigns without worrying about all the ancillary settings.

Let’s look at an Automated Bid Management feature for example. Auto Bidding is generally based on algorithms that look at a desired outcome, such as a particular position for an ad, and experiment frequently with the bid to attain that position at the best possible price. This feature would obviously utilize the “Edit keyword max CPC” feature that is required by the Google RMF. But just using it would not satisfy the requirement. One would also have to allow the user to be able to change the individual Keyword Max CPC.  Never mind that this is already set by them at the Ad Group level or that the whole point of the service they are paying for is to manage those settings. Never mind that changes made by the user to this setting may throw off the algorithm and cause it to perform at a less than optimal level.

Another example would be a company that builds an application that provides reporting on a user’s Adwords performance in a different way than Google offers.  Let’s say that company polled their target market and found that their customers are only interested in Average CPC, do not care about how it breaks up by device or ad network, and they do not want to see total impressions. Again, sorry, they all have to be included because these items are RMF. It is this kind of rigid regulation that kills potential innovation.

I understand having requirements to protect users (i.e. security requirements) is important.  But I do not understand stifling innovation with an arbitrary list of required items. If a company develops a tool using the Adwords API and they tailor it in a way that brings more people into Google’s advertising fold, then everyone wins. The company earns money, Google earns money, and the client is reaching a potentially new client base. And if that same company does not provide enough functionality or people do not like the way they are presented information, then that company isn’t going to stay around for very long.

Policies like these seem short sighted and Google is not the only company to implement them. I remember some arbitrary rules laid out by some big companies in the past and the problem is that the more of those rules a company has the more people they irritate and then suddenly everyone is looking for an alternative to whatever product that company provides. I know that Google is king right now and they can do whatever they want.  I can think of a few companies that used to have this same philosophy.  But one day they woke up to a world where they were no longer king and realized it is much, much harder to return to the top than it was to get there in the first place.

In general, it is good practice to make your website fast and usable. But did you know that it also affects your SEM efforts? Your quality scores for your ads and keywords in both AdCenter and AdWords take elements of your landing page’s performance into account.

“Well,” you ask, “what can I, a lowly website owner and potential or current advertiser, do about this dilemma?” or “How do I know if page speed and user experience are issues on my site?”

Answers to these questions vary in range from highly technical solutions to very simple changes that can have an impact on your advertising and overall user happiness in general. You want happy users, don’t you? Their happiness brings you word of mouth referrals to compliment your paid advertising, stretching the reach and value of your advertising budget.

“How do I know if page speed and user experience are issues on my site?”

Page Speed

The general answer here will most likely be: They are. Now, that is a pretty big statement from anyone, let alone a single engineer. And since I fancy myself a scientist, I’ll pass you off to a way to determine the imperfectness of any website known to man. Google Page Speed(https://developers.google.com/pagespeed/). Head there and type the address of your site in the box and click analyze. You should see something. Even Google, Facebook, CNN, and AOL have room for improvement. If your score is higher than 80 and you have no high priority, you’re generally OK. If you only have low priority items, you’re great, I’m a liar and you can stop reading about page speed (You should still check out my thoughts on user experience or the editors may have me flogged for writing an irrelevant blog post).

For more info, read Google’s help page on load time and quality:
http://support.google.com/adwords/bin/answer.py?hl=en&answer=87144

User Experience

Page speed is part of a user’s experience. No one wants to wait for a graphics heavy pages to load. Users also don’t need to be bombarded with everything under the sun that you have to offer. Remember that the user searched for a specific thing that matched one of your keywords, so show them that. Never clutter your landing pages with unrelated material and please, PLEASE, don’t put ads on your landing pages. AdCenter recommends not placing “a lot of advertising on your landing pages” but, in my humble opinion, don’t put any. If you are buying the keyword “teddy bears,” don’t land your potential users on a page selling toy trucks and don’t land your users on the home page unless all you do is sell teddy bears. Also, try to avoid redirects on your landing pages. Instead, land users directly on the relevant page itself. And as always, make your landing page relevant to your ads and keywords. If your landing page is informational, don’t try to pepper in keywords. Just write it normally and if your chosen keywords aren’t relevant, maybe you have the wrong keywords.

You can read more about user experience on landing pages at
http://advertising.microsoft.com/small-business/product-help/adcenter/topic?query=moonshot_conc_aboutlandingpageuserexp.htm
and
http://support.google.com/adwords/bin/answer.py?hl=en&answer=46675

“Well,” you ask, “what can I, a lowly website owner and potential or current advertiser, do about this dilemma?”

Simple page speed improvements

• Be specific: Remove clutter from landing pages. This focuses the user on what you are selling (what you are paying advertising dollars for) and don’t distract users with other offers on the landing page
• Land directly on the desired page: Don’t redirect through a special page, land users on the correct page directly
• Optimize images: If you are using images, make sure they are good for the web. Create thumbnails for large images and specify the sizes in the HTML, if possible. Don’t just specify the size in the <img> tag and leave the image as-is. A 5 megabyte, 15 megapixel jpeg included like so <img src=”my5megapixel.jpg” width=”200″ height=”150″> still requires the browser to download the entire 5 megabyte image and it resizes only after it is downloaded.

Simple user experience improvements:

• Land directly on the desired page: Don’t redirect through a special page, land users on the correct page directly
• Keep your landing pages free of irrelevant information
• Keep ads off of your landing pages. If you want users to see other offerings on your site, create other ads for those offerings. If you are using services like AdSense, don’t put them on your landing pages (this should be very obvious since it could potentially give other sites your advertising budget anyway).

Obviously, this is no small subject and this article just scratches the surface. You can read into the suggestions from Google’s Page Speed tool to find more information about specfic site speed ups. Check out the other links in this post to continue learning about why page speed and user experience are important parts of online advertising.