Website Tripwires: Using Secondary Action to Prove Value

In the 1996 film adaptation of Mission Impossible, Tom Cruise's character repels into a secure room while trying to steal sensitive data from a computer. There are three apparent ways that he can trip the security system:

1. Making any noise

2. Touching walls or the floor

3. (Obviously) being seen

Those holding the sensitive data clearly considered multiple tripwires to determine when and how an intruder had entered the room.



What does Tom Cruise have to do with website analytics and search engine marketing?

Allow me to explain.

Imagine your website visitors are little Tom Cruises. Each visitor is repelling into your site looking to achieve a goal, whether that be to find information, buy your product, or maybe even to steal sensitive data. During each visit, your own system of tripwires should let you know when actions take place - beyond just the final conversion process.

Website Tripwires can be crucial to determining where your audience is in the decision funnel, where holes lie, or what information or events visitors are looking for.

Types of Website Tripwires

When setting up your Analytics solution, consider all areas of your website for secondary conversion. Besides, if there isn't an opportunity for conversion or secondary action on a page, why does the page exist at all?

Tripwires can be leveraged just about anywhere. Here are ten of my favorite uses:

1. Google Analytics Javascript Tracking: Use Google Analytics "onclick" code, or pageview tracking, to measure events such as clicks. Set up pageviews as conversions in your Analytics profile (preferably a new goal set, if it is a secondary goal), and measure away. Note: When setting up onclick code, you will likely have an influx of pageviews and thus your pages per visit will increase.

2. Google Analytics Event Tracking: Google Analytics Event Tracking is similar to the pageview tracking above, but it does not come through as pageviews, rather as "events". The advantage here is not having your pageviews disrupted, and it also allows you to determine categories of events, which is extremely helpful when you are measuring multiple layers of the same action. The drawback may be that your events are tracked almost entirely separate from other data and may be difficult to determine traffic sources, navigation, etc. Try using audience segmentation to find this data.

3. Flash Heat Mapping: If you make use of flash on your website, especially when users can engage within the flash, you may want to consider flash heat mapping as an interesting measure of user activity.

4. Printing/Bookmarking: If a webpage is offering crucial information, users often elect to print or bookmark to keep tabs. Use click-tracking to determine when one of these events takes place, as they may share key data about a user's mindset when browsing your website. For information-rich pages, you may want to add clearer print or bookmark buttons to facilitate action.

5. Share/RSS/Social Buttons: Like the above actions, sharing or feeding information can provide valuable insights regarding what a visitors deems most useful.

6. User Opt-in: With mid-funnel opt-in, you may have the opportunity to capture valuable nuggets of user data, giving you the ability to maintain communication during a longer decision cycle. Consider adding email signups for newsletters or just to get more information if a user is not ready to convert then and there.

7. Information Download: For longer purchase cycles, downloadable content such as case studies, white papers, press releases or information guides can be the bridge from consideration to conversion. Allow these breadcrumbs to be downloaded, and track each one as a lower-funnel tripwire.

8. Video Play: Getting a visitor to sit and watch your video content is no simple task, and could be indicative of a visitor that is ready to convert or move along in the process. Looking at something as simple as time on pages where videos are present could be enough to incite value, but a better option would be to tag video "plays" as tripwire goals.

9. Outbound Linking: If a user is clicking to leave your site via in-text or footer links, what does that say about your content? Why is the user leaving, and is there some information that they expect to find elsewhere? Finding these answers could hold the key to moving users along the funnel.

10. Pure Engagement: Traditional engagement statistics are "low-hanging fruit" tripwires, and though sometimes overstated, can often prove value. Google Analytics allows you to identify certain time on site as a goal conversion, which might lead to further insights regarding which visitors are there for some real engagement.

Ultimately, you should be able to fill in the missing pieces about each visitor's experience on your website by tracking each action as an important event. Connect all of the dots using website tripwires and the full picture will come into view.

That is your mission, if you choose to accept it.

Search Audience Segmentation: Moving Beyond Keywords, Ad Groups, and Campaigns

Who is your searching audience?

One of the great benefits of events like SES is it provides the ability to refine large and sometimes abstract ideas to make them computable in pieces of information.

A standout "idea nugget" that caught my attention was using "mental models" to drive marketing efforts. Although this was brought up in the context of website planning, there was a connection to recent thinking around paid search and ways to take it beyond just keywords and ads and to the heart of advertising - the Audience.

Admittedly, this is no breakthrough - there has been plenty said before about the importance of audience segmentation in advertising. That said, when is the last time you actually planned, executed, measured, and optimized a PPC program with all of your different audiences in mind - as opposed to, say, the best overall converting keyword?

Moving Beyond Labels

The traditional structure of a PPC account has its clear uses: control, organization, and targeting, to name a few. Unfortunately, the presence of these labels can also be confining for strategic thought, and can limit the ability to make the next big leap in your results.

Thinking of your clicks in terms of mental models can be both liberating and beneficial to your results. So how do you get started?

Here are some questions you might want to ask yourself:

1. What are the goals of my landing page or website? As we all know, goals are paramount. But we also know that not all clicking audiences will convert, so the key is understanding which clicking audiences are converting, and how. If they are not completing the primary action, then what are the secondary goals of the website?

2. Does my existing account already target audiences that I'm not aware of yet? Whether or not you've realized it, your paid search program most likely reaches a number of audience types, all of whom are looking for different things. Take a close look at your campaigns, ad groups, and keywords, and consider what the users are looking for in regards to your website, regardless of the account labels that have been established.

3. Are all of my audiences created equal? Some users that are clicking through to your website are more or less valuable than others, based on their intent and the likelihood of conversion. Knowing which "clickers" are where in your conversion funnel is an essential step in asserting value to each type.

Next steps

Once you understand and identify your audiences, there are a number of actions you can take to help your program succeed.

1. Opportunity: Now that you know the users clicking on your ads, you may witness some new opportunities that had not presented themselves before. Is there something this audience likes/dislikes that your can leverage? New keywords to test?

2. Navigation Paths: The audience segmentation exercises may very well end up informing how you view your website. Different audiences will likely pursue unique paths to get to their conversion; What can you do to make their experience more effective? Are your audiences doing what you expect, or are they providing surprises that can be used for website optimizations?

3. Account Segmentation: Although you have been freed strategically from the limitations of a PPC account, you are still confined by the labels in practice. However, this doesn't mean you can't use them to your advantage. After looking at the results from your audiences, is one core audience type being squeezed out by others due to budget flow equality? It might make sense to give certain primary audiences their own budget or targeting to ensure tighter control and more resources, while limiting the presence of lower-quality audiences.

4. Landing Page Messaging: If your audiences are looking for different things when clicking, why not give them different messages upon landing on your website? Ways to achieve this could be by simply sending them to different pages, or perhaps recognizing their needs and changing small landing page copy, messaging, or calls to action to better give them the confidence that they made the correct decision by clicking.

5. Reporting and Analysis: Audience segmentation in search can't just be used for account improvements, but can also be helpful when reporting and presenting materials to others, as it provides neat packages in which to place results; and if your PPC audiences are in line with website mental models and results, even better! When reporting results, try to illustrate the story of each one of your PPC audiences, and prove why each one is valuable, or how they can be better served through marketing efforts.

Segmenting audiences in your paid search accounts can be a powerful way to re-evaluate your programs, and can lead to positive results.

Do you have any additional thoughts about PPC audience segmentation? Feel free to share in the comments below.

Is Web Design Holding Your Landing Page Optimization Back - or Vice Versa?

Ssssh...Landing Page Optimization in progress.

By now you've surely read that landing page optimization is the key to improving conversions...

...easy, right?

Well, that's what they'd like you to believe. But unless you are working at a shop that works strictly on testing and optimization for landing pages (working through these shops has plenty of drawbacks - but that's for another time), initiating and practicing landing page optimization (I'm calling it LiPO from now on, because it makes total sense) can be arduous and a bit of a hassle.

When working in an agency or even an in-house setting there are plenty of issues that can hold you back when trying to test; among them is the science/art of Web Design.

Is True Web Design Dead?

There was a blog post being passed around recently - especially by the creative folks - recalling the story of a Google designer who was forced to test 41 shades of blue which effectively led to the epiphany that web design is dead (or at

Could this be the future of Web Design?

least that was my interpretation). He ultimately quit.

This was all some sort of vindication for the apparent futility and inanity of testing landing pages, and a bleak vision into the future for web designers (gasp!).

Clearly the anecdote provided was extreme, but what can't be ignored is the fact that designers are not usually responsible for producing, recording, or reporting on the results from their landing pages.

Unlike TV, where many departments share the responsibility of driving results, and where results are more abstract, online results can be infinitely measured and are thus held to a higher standard of results generation (not to mention online users are further down the conversion funnel than television viewers, so the importance of conversion is paramount).

Unfortunately, websites can look pretty and still completely fail to meet expectations - and guess who's held accountable? Not the designers.

With all of that said, web designers remain a key player in the LiPO process - they cannot and should not be marginalized. This is especially true if and when you are trying to strike that delicate balance between brand message and conversion generation. It is crucial that all parties understand the unique characteristics of the web, and what the goals are from the outset.

So what's the solution?

1. When testing, make it clear to the designer what the best usability practices are, what elements you want to test, and why it is so important to do so, but...

2. ...be careful not to step on the "creative toes" so they have freedom to do what they do best. Do not tell them how to do their jobs. There's a reason they were hired to design for the web, and you weren't.

3. Understand what the brand message is, and what needs to remain in the design to maintain it. Even if the website is supposed to drive results, it shouldn't abandon core brand message principles. You may just have to work with certain colors or images.

4. Allow for at least one initial A/B test (read: completely different layouts). This gives the designer some freedom to come up with multiple designs, and sets you off on the right path.

5. See if it is possible to design layout templates so that the elements can be moved and changed without compromising the integrity of the design.

6. Take one or two elements that the designer really wants, and test against something you want. You never know, this could be a humbling experience for the tester as well!

7. Get the designers in on the action. Ask for their professional opinion, draw from their experiences, and make sure to take their bets on the winning design!

In the end, landing page testing does not need to spell the death of web design - they can go hand in hand to make both a better looking and smarter future.

Have you experienced any unique challenges with Web Design and Landing Page Optimization? Share your experiences in the comments section below!

Web Design and LiPO Testing Can Work Together.

PPC Brand Personality Disorder Part 2: Dealing with Smaller Brands

PPC Brand Personality #2: The Ankle-Biter

Defined (by Deco Dig, of course) as a small company whose brand awareness is relatively low in its respective advertising market/s, The Ankle-Biter may very well compete with PPC Brand Personality 1: The Kleenex, among others, but with a much lower brand awareness and budget. As a result of this, s/he must be more resourceful when it comes to search marketing tactics.

In short, The Ankle-Biter is the Underdog; the Little Company that Could; the "Rudy", if you will.

How do you know if you are the Ankle-Biter? To answer this question, you have a great measurement gauge at your disposal in paid search data. I tend to think that if your branded keywords cannot match 10-20% of your general terms in clicks with unlimited budget, you fall into this category. There are exceptions (e.g. really, really high volume and/or expensive general keyword terms) but for the most part I have seen this to be true. Either way, if searches on your brand name do not constitute a noticeable portion of search traffic, be it paid or organic, you might be the AB.

If you have come to the realization that your brand/client is The AB, then listen up, because there are always methods of taking it to the big guys.

1. The first and most obvious tactic is to actually bid on competitors' brand names. Some of you may say that this technique is "immoral" (morality in advertising? psh!). My argument regarding this is threefold:

• This is a topic that has been argued in a legal setting, most notably by Geico insurance, and the right to bid on brand names was never deemed illegal. This means that the technique remains moral by legal standards, which is good enough for me.
• Google has strict trademark rules and regulations which stop an ad from actually calling out competition in ad copy. Therefore you are simply capitalizing on an opportunity to reach the active searcher, or a user who is clearly interested in what your brand has to offer. If the searcher is looking for something, and you can offer it, it is Google's responsibility to offer that option. If the user clicks on your ad, it is the competitor's fault for not closing the deal with his/her ad when the advantage was clearly there.
• The common argument against buying competitor's names is that by doing so, you are essentially piggy-backing on the popularity of a brand that has spent countless amounts of money/time/effort building said brand. This is true, and there is no way around it. Big whoop. If your offer is better for the consumer, I don't know how this can be immoral, considering the bottom line should be all about the consumer (that's right!). If we are talking about brand recognition of the Kleenex in particular, there really is no way of getting around bidding on his/her name, since the majority of searches will have this name in mind due to the synonymy of the name and the product.

However, before going this route, you should be aware of the potential pitfalls:

• Make sure your boss/legal dept./client is okay with this technique. Despite my points above, some companies just aren't comfortable with the so-called "hyper competitive" method of buying other brand names. If your client or boss is the type that likes to be involved in all aspects of the marketing, run this by him/her first.
• If your industry is somewhat tightly knit, or if you know the other parties, perhaps there is an agreement to not take this route. I have, in fact, had clients who did not want to bid on their peers' names due to this. Fair enough.
• You probably don't want to use keyword insertion ads when bidding on competitor's names. Not only do you want to avoid displaying competitors' brand names in any ad text (including body copy) due to trademark issues, you also want to steer clear of misleading searchers with your ads - for quality score, CTR, bounce rate, and conversion rate purposes.
• If the competition is paying attention, they will always come before you in the results. No worries there, just make sure your message is clear, concise, and enthralling enough to catch the user's attention and wedge that seed of doubt about the big brand.
• Don't be surprised to see the competition send it right back at you, depending on the growth of your product and the scope of the campaigns. If you are truly the Ankle-Biter, this should not matter as your brand terms will not be getting much play. However, you should be sure that you bid on your own brand keywords nonetheless, and in such cases refer to the Kleenex techniques and throw in competitive advantage ad points.
• Remember, this is still all about your goals and those of your client/company. Be careful with bidding on company terms that deal mostly in something that will not convert for you. For example, if the majority of users search for The Kleenex because they are looking for a video that you cannot offer or compete with, make sure to refine your keywords or add negative keywords. Purchasing competitors' keywords should increase the odds of goal conversions; you do not want inadvertent clicks.

With all of the above in mind, here are the best practices for bidding on your competitors brand names:

• Someone once said "If the competition is paying attention, they will always come before you." I'll take it to the next level - if the competition is not paying attention, go for that top spot and make them pay! There is nothing more demoralizing that having some other brand name show up in your search results, and there is no more effective way for the Ankle-Biter to take a chunk out of the competitor's searcher base that to show up ahead of it for searches.

  

  Don't be afraid to take it to the Big Guys

• Your search text ads should clearly state the competitive advantage of your product/brand over the other. Whether it be a defined price point, award, proximity, etc. make it obvious (without belittling the competition) that your ad should be clicked on instead. Again, you should highlight your strengths, not the competitors' weaknesses. This isn't a smear campaign, after all.
• Just like if you were the Kleenex, you should bid on the various misspellings of very prevalent brand names. By doing this, you will also drastically increase the chance that you are bidding on highly searched on terms that your competitors haven't considered. Bonus!
• Local, local, local. As a smaller brand, you most likely have a regional advantage over the competition. If so, let it be heard! I have seen state/city names in headlines work very effectively. Also Google Maps is a useful tool if you have a physical location.
• On the same level as the last bullet, perhaps a competitive advantage is the fact that you are so much smaller than the big guys. Tout you level of customer service due to your size, or the speediness of delivery, or the "family owned and operated" tag if applicable. Anything to make big guys or girls seem large and impersonal.

Clearly, the technique of buying you opponents' brand names in paid search can be a delicate balance of morality and artful marketing. As always, do what you feel is right for your brand. As the Ankle-Biter, you are looking up at The Kleenexes of the world and paid search provides a ladder more readily than any other medium.

2. Spend more time on long-tail keywords.

Long tail keywords aren't extinct from The Anklebiter's list of tactics

The quandaries of long-tail terms, or keyword phrases that are long and detailed, have been well documented. In fact, it seems to be yet another case of a "finger trap" search marketing issue, whereas there needs to be some give and take in order to come up with optimal solutions.

In the case of the Ankle-Biter, long-tail keyword phrases are probably more important than in any other case.

Why is this?

Put bluntly, it is because you simply cannot compete with the big guys on high volume keywords. Or, perhaps more importantly, trying to compete with those dense keywords will ruin your budget and have your Ankle-Biting client gnawing on you because they are not seeing results from their paid search investments.

Furthermore, a recent Hitwise study found that the average number of search keywords is growing as users become more knowledgeable about how to search.

Here is what I suggest:

• Use the tools at your disposal to come up with the most significant long term phrases that searchers use. This would include keyword tools, competitive research (look in competitors' search ads and websites) and most importantly any historical data that you may have, whether it be organic website searches, in-site searches, or search query reports.
• If you consistently run out of daily budget, consider a freeze on the high-volume terms. Just for a couple of days. Do it. You may learn more from pausing keywords than adding them. By freeing up funds that would otherwise go towards dense, expensive, "blob" keywords that arguably are less measurable due to their girth, you can glean magical insights from the more specific keywords that will cost much less. With the right keyword list, you should at least come close to hitting your budget, but with many more clicks (depending on the size of you budget)! This is great because it A) lets you see which specific phrases are more effective when it comes to goal conversion, B) tells you what your ceiling is on the low-cost phrases, and C) you may end up with some helpful insights about how your users are finding you (or trying to find you!) and you can then expand on a particular phrase that works best.
• Depending on your product/offer, you should definitely be bidding on detailed product names and types. These will be extremely likely to convert post-click, they will cost less, and you have a better opportunity of showing up at the top of results pages. Continually look at how users are searching for your products and expand your long-tail keyword lists to include all variations of search.

As the Ankle-biter, you may be better positioned to take advantage of fast changing consumer mindset and search tendencies. Leverage your ability to be nimble and you'll see results that the Big Brands have trouble emulating.

Remember, just because your brand awareness is low, it doesn't mean you can't compete with valuable brand searches in paid search and let searchers know you are there for them!

Feel free to add your thoughts or ideas below on how branded terms can be leveraged in paid search.

PPC Brand Personality Disorder: How to Deal with Big Brands in Paid Search

Brand keywords have caused me to think. A lot.

For each client there is clearly a different "right" approach regarding how to use the brand name when frolicking through the land of paid search.

Your first response to this is, predictably, "Yea...and...?", or more descriptively, "isn't that the point of being paid for this? To think about the strategy?"

Yes and yes.

But that doesn't mean I can't write blog posts about it to find different ways to circumvent this whole "thinking" process. Then we can all spend more time on more important things like...facebooking...or whatever the kids do nowadays.

There are a plethora of important reasons why you should be using brand terms in your paid search. These keywords can obviously explain a lot about brand awareness (the "awareness gauge") and they can send extremely valuable and inexpensive traffic with users at the tail end of the purchasing cycle (depending on your goals), plus the added benefit of being able to see just how many people misspell (read: mangle) the most innocuous of company names for the sake of search.

However, branded terms can also be somewhat enigmatic. And the question remains, are the branded terms that you are using really helping your results? if so, how much? Again, it depends on your goals and your client.

In the next few posts I'll try to neatly package the ways that I've used branded terms by profiling the different types of brands you may end up dealing with, and the helpful tactics that I have seen.

Brand Personality: The Kleenex

Everyone knows your name; you are synonymous with the general product you represent (Kleenex = Tissue). When others somewhat resemble you, you often hear things like "OMG, you remind me so much of the Kleenex!", or, "Did the Kleenex get a hair cut?", or even, "Yo! It's Kleenex!".

The Kleenex's goal is most likely not to build on existing popularity, but to make sure everyone knows it is still there and is still the sh*t, sometimes by buying new shoes and showing them off. Some call this "defensive marketing" which, for ubiquitous brand names, means it is not about getting to the top, but more about staying on top while others chuck grenades at your text ads.

Here are the problems with being the Kleenex: Everyone knows you (good), which means there are a lot of people riding your coattails to success and potentially driving up your costs (not good). This also means that everyone tries to be like you, so you can end up blending in with the crowd if you're not careful, especially in search results. Furthermore, you are an easy target for undercutting, and people saying that they are the better version of you.

If you are the Kleenex, here are ways to make sure you are getting the best results out of your paid search:

1. You must always remind people that you are the real deal. Make it a priority to be at the top of the page. It seems obvious aloud, but how often do you actually look at the daily positions for your branded terms? Sure, your quality score should be handing you the top positioning on a platter (a delicious, delicious platter), but don't look away - that position is coveted by the little guys who just want a piece of you and/or want to steal your squeeze. They should be willing to bid the big bucks to get above you, since capturing the users at this time in the purchasing cycle is a rare thing.

Ways to remind the searching audience of your awesomeness:

• Long story short: Bid more. If your name is penetrating enough to fall into the Kleenex personality type, you have the money to spare. Besides, the clicks for these searches generally cost less anyways. If you're just starting out with the campaign and you're missing the results history desired for top billing, you might want to bid heavier at first to secure that spot and high CTR before dropping CPC.
• Use "Official Site" in your ads and constantly place your brand name or product name (depending on the search terms) into the ad copy and headline.
• Be sure that your quality scores are as high as high can be for your branded keywords. This means placing your brand name in the headlines of ads and ensuring that the ad text is as relevant as possible to the keyword (duh). This could also mean splitting up your Company/Brand/Product terms into multiple ad groups for categorization and ease of matching ad messages to searches.
• Test the registered trademark symbol in your ads next to your brand/products and measure any results on CTR.
• Use position preferences if your average position is not at the top. You might want to just test this one out for a few days, since it will not just put you in top positions for each, it will just not show you in poor positions. You should only use this tool if you believe that appearing low in results will either adversly affect your brand name or will bring down your CTR, in which case you might take a hit in quality score or CPC.
• Also, most search engines have trademark laws that should be adhered to across the board. Make a competitive sweep once a month just to make sure ads aren't using your name and/or products in ads. If so, file a complaint with the search engine to remove your name from competitors' ads. To deflect this potential up front, you might want to check with the search engines and make sure your trademarked name is registered with the network so competitors will be thwarted when writing their malevolent ads.
• NEW UPDATE: Use Google Sitelinks if you can. This will give your branded ads additional real estate and will allow you to link to multiple pages from one ad. Huge advantage for big brands.

Reminding the user that you are "the one and only Kleenex", and that all others bow down to you, lends an incredible amount of credibility and plants the seed of doubt about all of those posers on the page.


2. Cover misspellings with their own ad group. You'd be amazed at the number of ways people can misspell your famous name. Clean-ex, kleen ex, cleanix, chleanox...the list is infinite. Again, you do not want competitors stealing your thunder, so missing out on a few of these terms could add up and could make you lose out on many customers.

• Check out the search query report if you are using Google Adwords, look into the most popular organic searches in your analytics program, or check out review sites where people comment on your product (and make sure they are saying nice things!).
• Moreover, setting the misspellings aside in their own ad group will help you keep track of which misspellings are happening the most. Knowing this, you may want to gear some ads specifically to these misspellings, or even throw in very prevalent misspellings strategically on your website to help with quality score and SEO.

3. Give brand terms a separate budget by making them a new campaign. This could be a big one. It depends on what your users are using your brand name for, and it could be troublesome if your targeting/strategy gets hairy. It could also pay off.

For example, if you are a bank, or any website in which users constantly return to sign in, how much of your paid search budget should be used up by these users coming in through text ads?

If you are sending users to a site where customers log in, here are a few approaches:

• Limit budget, but don't get rid of branded terms entirely. You do not want to completely abandon these keywords since many users may want sign up instead of sign in. All you have to do is put thought into what types of users you are willing to pay for, and what they are searching on (so simple yet so effective!).
• If you are truly concerned about this, add negative keywords such as "login", "signin", "existing user", or "customer", but make sure you are the #1 organic search result for these searches.
• On the same note, look into excluding terms that deal with customer service, troubleshooting, etc. If you are not a company that focuses a lot of effort on these items, the users coming to complain or troubleshoot your product can use organic, no use paying for it.

On the other hand, if your brand terms are the most efficient and effective way that users not only get to your site but also end up converting, you don't want the more expensive broad terms eating up your whole budget while the branded terms are left in the dust, do you?! For this reason, test the brand search budget ceiling for these branded terms by tinkering with the budget for about a week or two, and come up with a figure that allows these keywords to fly free. Not only will this drive down CPCs (which the client loves!) but it will also increase overall clicks, conversion rate, and CPA (which the client also loves! Good times).

The bottom line is this: Branded terms are very inexpensive and can be invaluable, especially if you use them correctly. In most cases, it is not worth it to leave your brand and product names out of the campaign, if only because they cost next to nothing. For the "big gun" brands such as our friend the Kleenex, defensive marketing tactics are a necessity in the online search environment, since everyone wants your position and customers!

Constantly remind searchers of just how important you are ("kind of a big deal") and why your official-ness is more awesome than being a brand freeloader.

Stay tuned for the next post, Brand Personality: The Anklebiter.