Author Archive
Posted on August 23, 2012 by Christina Keffer
Keyword Choice!!! (followed by dramatic music)
I have to choose keywords?!!! Noooooooooooooo!
Choosing keywords at the beginning of an SEO campaign is freaking terrifying. The keywords that you choose at the beginning and spend tons of time and energy ranking for had better deliver the kind of qualified traffic you need on the landing pages of your choice — or else.
If you find yourself ranking highly for the keywords you have chosen, and you’re still not seeing the traffic you expect (or is expected of you) things are looking down. If you see a huge uptick in traffic and no consonant uptick in conversions, things are looking down. So down that this puppy got really, realy sad when his keywords didn’t deliver the right kind of traffic.
The worst thing about keyword research is that you have this list you come up with and end up using over the following 3-6 months and every time you look at it you wonder “Really? Is this really the right phrase? Should it have a call to action included? How about the location? Is it too specific to really drive a decent volume of traffic? CAN I CHANGE IT NOW?” and the answer to that last question is always: NO. You can make slight changes, but once you have started the optimization and link building process that is focused around a core set of keywords, deviating from your path will lead to disappointment.
(more…)
View Comments (2 Responses) | Categories: Search Engine Optimization
Posted on June 7, 2012 by Christina Keffer
Contrary to popular belief, LunaMetrics is not usually in the business of sending referral traffic to generic Viagra sites. However, due to some issues with our host, recently, we have been! We consider this a big problem, and not just because they’re not paying for the traffic.
We think we’ve kicked the bug, but if any of you, our dear readers, find yourself on a site that you didn’t expect, please email us so we can fix the problem quicker and get you back to you regularly scheduled Analytics, Social Media or SEO program! Thanks!
View Comments (No Responses) | Categories: Search Engine Optimization
Posted on November 17, 2011 by Christina Keffer
Playing the saddest song in the world...
Over the past few weeks since Google implemented SSL encrypted search results pages that block keyword data for users signed into their Google accounts, SEOs have been pulling their hair and gnashing their teeth. It’s true, and it’s warranted. One of the best SEO KPIs is reporting on the quantity of organic search traffic generated by highly qualified keywords. Now that between 9 and 20% of this data is generic, it’s harder for us to prove our worth and progress. So we’re justifiably upset.
However, as objectively as I can be, I would say that removing data for organic searches has far wider reaching consequences than depriving SEOs of one of their prime KPIs.
One of my friends who owns a great hip hop dance video eCommerce store and I got into a debate last weekend about how it’s not just SEOs who are affected by this change. My argument, which I could not deliver with any eloquence that night, (Friday at 10pm after happy hour and wine at dinner — you do the math.) I will now present here.
Why The (Not Provided) Keyword Hurts the Web
Let me begin by asking a question: Why do webmasters care about rankings? Because rankings=traffic and traffic=some benefit (either monetary or otherwise.) The only thing that webmasters want to do is get their site ranking for relevant terms that drive qualified traffic. To do so, they optimize their sites for those terms.
It is in the webmasters’ best interests to focus on keywords that drive qualified traffic, and to make their site as relevant as possible for this valuable traffic. But how, you may ask, do webmasters know whether they are optimizing for the right things?
What if someone searching for “vice grips” typically wants to buy and someone searching for “locking wrench” typically wants to find out more about the applications for that tool? Both of those words could apply to the same tool, so webmasters look at their conversion data and engagement metrics to find out how to make their site fit the wants and needs of the searcher.
By doing so, they inadvertently give the search engines exactly what they need and want: a page that matches user intent and is valuable to the searcher — who will then use that search engine again because the results were exactly what they were looking for.
The decision to try to rank a product page for vice grips vs. locking wrench has to be informed by keyword data since Google doesn’t provide exact clickthrough data — even through webmasters tools. This data should be statistically significant.
I would hazard a guess, however, that most sites out there don’t have statistically significant traffic over the short term. So even if the site really IS the best result for a specific search, they are already at a disadvantage. Now, however, about 10 – 20% of their already slim keyword data is generalized.
So, the webmaster makes a decision based on personal preference or the opinions of peers in his industry and optimizes his sales page for vice grips. He manages to rank pretty highly for that and can’t understand why people are bouncing so much.
Google, in turn is ranking him appropriately to his level of optimization, but for the wrong keyword and the result is not what the user was looking for so they loose too.
*sigh*
So, why is Google denying analytics users this important information? There is a lot of speculation that it is in an effort to push site owners to depend more on paid search data… up their paid search traffic since they can’t get accurate data through the organic channel anymore.
If this is the case, it seems really short sited to me. The only reason why most people search Google is to get the organic results. Clicking on a paid result is a bonus. If the organic results decline because webmasters don’t have the right information to make education decisions about their site, then Google looses money because people will start using other engines.
If they really are trying to stave off a privacy issue, either present or future, then why is the data still available for paid search? Have people that click on paid links entered into some type of implicit agreement that their search data is less private than those who click on organic results?
Regardless of their motives, the change will detrimentally affect the web in general, search results specifically, and the ability webmasters to make good decisions.
As an SEO, I admit that I’m personally put out by these changes because they make my job harder and who likes that? (See above image.) However, I really and truly believe that the web as a whole will be drastically and negatively affected, especially if more and more people sign in to a Google account while searching.
Have you seen any (not provided) traffic? Let us know how it’s affected you in the comments.
View Comments (14 Responses) | Categories: Analytics, Google Analytics, Search Engine Optimization
Posted on October 20, 2011 by Christina Keffer
Keyword (Not Provided) In GA Keywords Report
You may or may not have noticed something fishy in your Google Analytics Keywords report recently. If you haven’t noticed, go into Google Analytics and do the following:
- Use the Non-Paid Search Traffic Advanced Segment
- Change the Date Range to 10/17 – 10-19
- Go to Traffic Sources > Sources> Organic
- Observe the top three or four keyowrds
Do you see it? If it’s not at the top, it’s in there somewhere. It’s a keyword called (Not Provided.)
Now, if you were training at an SEO event like I was on the 17th and then was out of the office (and largely offline) on the 18th or if you live under a rock somewhere, you might not have heard Google’s official announcement that they will no longer be providing keyword data for organic search results if the user is signed into their Google account.
It’s not just Google Analytics that will be denied this data. By “enhancing” their default user experience for signed in users, Google will be redirecting signed in users to https://www.google.com, thus encrypting the search results page. In analytics, you’ll still be able to see that these signed in users came from the organic search results, but instead of being able to see the actual keywords that they used, you’ll see all that data aggregated under (Not Provided.)
If you’re an SEO who uses the keywords report to prove the validity and efficacy of your work, you’re screaming and gnashing your teeth by this point. If you’re a causal analytics user, you may be asking the question “why do this?”
Well, obviously, it’s to protect the user: “As search becomes an increasingly customized experience, we recognize the growing importance of protecting the personalized search results we deliver.” (excerpt from Google’s official statement)
In what way does hiding the queries that signed in users use to get to your site infringe upon their privacy since all the data is anonymous anyway? That’s a legitimate question by the way. Feel free to answer it in the comments! I’m curious what you all thing. And why, oh WHY are PAID search keywords not affected by this change.
That’s right. You can still see every single keyword that sent traffic through paid search, whether the user is signed in or not — just not organic search. Are users who click on paid search results less safe than users that click on organic results?
*Breathes Deeply*
Long Lasting Repercussions of Keyword (Not Provided)
Google claims that this change will affect only a small percentage of data, since only those who are signed into their Google account when searching will be “protected.” Well, I’m throwing down the benchmark here and now.
So far, since this change launched, LunaMetrics has seen 1% of our keywords clumped into (Not Provided.) A client with substantially larger organic search volume has already seen almost 2% of their organic keywords represented as Not Provided. We shall see how far-reaching these changes actually are in a few weeks when they’re rolled out completely.
Additionally, Google’s placation that only a small percentage of data will be affected because of the amount of people who search while NOT signed is cold comfort to me when they’re trying so very very hard to push the adoption of Google+ on the masses. If they have their way everyone would always be signed into their Google account when online.
I would love to hear the thoughts and concerns, and views of everyone else! Thanks!
View Comments (30 Responses) | Categories: Analytics, Google Analytics, Search Engine Optimization
Posted on September 6, 2011 by Christina Keffer
SEO KPIs
Clients, it turns out, want to see how the efforts they’re paying for are working. We show clients lots of reports that indicate the success of their SEO campaigns, and KPIs are slightly different for every client depending on what kind of a site they have, what their goals are, etc. However, there are three univeral SEO Key performance indicators that we use for everyone.
- Traffic: Our major KPI for Search engine optimization is the increase of inbound organic, non-branded traffic month over month (or year over year etc.)
- Rankings: A lot of clients come to us obsessing about rankings and to tell you the truth, we’re obsessed about rankings too. But it’s impossible, simply impossible, to use a 3rd party tool to figure out every single derivation of a short-tail key phrase that might possibly rank. Sure we track a select list of short tail key phrases in order to provide a barometer for the campaign, but the only purpose for getting higher rankings is to increase quality traffic to the website which brings us back to our main KPI.
- Links: As we all know, increasing inbound links plays a huge role in getting a site ranked higher. Therefore, another SEO key performance indicator for us is the quantity of inbound links we generate. There are lots of external tools you can use to track these links like Majestic SEO and Open Site Explorer, but you can use Google Analytics to keep tabs on them too, assuming you build quality links that get traffic as well as raise rankings.
Tracking Traffic In Google Analytics
This is usually ridiculously easy to track through Google Analytics if you use the old interface. If you use the NEW interface, the score card no longer shows comparisons with past data, only comparisons to the site average, which, in my opinion, is super, super lame. You just use Advanced segments to filter out everything but non-paid organic search traffic and run a keywords report and filter out the brand name. Easy right?
Well, what if you’re only responsible for optimizing a certain section of a site. For instance: For internal political reasons, one of our larger clients has a lot of editorial leeway on a specific section of their site where they can make all the on-page optimization changes (titles, metas, copy and headers) changes that we ask for.
For this reason, we’re optimizing just that section of their site. So how do you measure SEO success just for that section? For the purposes of protecting client information, I’m going to use LunaMetrics data instead for this instance. Say I’m responsible for optimizing the Blog section of the site. This section is www.lunametrics.com/blog. So in order to find traffic progress month over month, I’d do the following:
Step 1. Get the correct Date Comparison in Google Analytics. In order to make a clean comparison and take the weekly dips in traffic over the weekends into consideration, I start the date comparison on the same day of the week, even if it means going a bit into the previous month:
Google Analytics Date Comparison
Step 2. Under Traffic Sources, choose Keywords.
Google Analytics Keywords Report
Step 3. Make sure you click the “non-paid” link to make sure you’re filtering out any paid search. We don’t want to take paid search into consideration while we try to judge the efficacy of organic search efforts.
Non Paid Keyowrds
Step 4: Choose the Landing Page secondary dimension
Landing Page Secondary Dimension
Step 5: Using the advanced filter, filter out branded keywords and the landing pages that are in the section you’re considering.
Google Analytics Advanced Filter
Advanced Filter Example
Step 6: Enjoy your report which shows how many people landed on the page in the section you are concerned with through organic search and which keywords they found that page with. Both the number of visits and the quantity of keywords should go up month over month.
Blog Traffic Increases
Tracking Inbound Links in Google Analytics
We use the Referral report as one way to track inbound links from other sites. Get it? Cause the linking sites refer traffic? There are a couple ways to use the referral traffic data. If you’re keeping a list of sites that from which you’re pursuing links, you can simply filter the referrals page for those sites and see if you’ve started getting any traffic from them. Even one visit means someone came from a link on that site.
Referral Traffic in Google Analytics
If you’ve got a strong link-bait campaign going, you can track progress on that by choosing the secondary dimension “landing page” and then filtering for the page that you wrote as link bait.
Referral traffic to Link Bait
Hopefully this has given you some good insights into how to judge the efficacy of your SEO campaign using nice, free, easy-to-use Google Analytics. Happy Analysis!
What KPI’s are you using to measure your success? Let us know in the comments section!
View Comments (3 Responses) | Categories: Analytics, Google Analytics, Search Engine Optimization
Posted on August 18, 2011 by Christina Keffer
Sometimes the simplest things go unnoticed. Recently, a client informed me that they were going to start including QR codes in their print marketing and did I know of a good tracking program. About half an hour into researching the right solution for them, I had an epiphany/head-meet-desk moment when I realized that they were already hooked up with GA and could easily track a QR campaign in their existing account. Here’s how!
Google Analytics URL Creator
Thoughtfully, Google provides a fool-proof way of creating unique urls with urchin tracking modules (or UTM) parameters appended to the URL so that you can track the QR campaign in Google Analytics. Here’s a screen shot of the setup:
Now shorten that LONG URL
After you create the URL, plug it into your favorite URL shortener. Why should I? You may ask. It’s true, it does add another step to the process, but it’ll be well worth your while. See, the more characters a url has, the more information needs to be embedded in the QR Code. This means that the actual QR Code ends up being much more dense, for lack of a better word. That means it’ll be harder for a mobile device to scan it.
When you’re making QR codes for print advertising purposes, it’s likely that that QR Code isn’t going to take up much real estate. That makes this denseness issue even more important. Plus, using a URL shortener like Bit.ly is super quick and easy.
Dense, hard to scan QR Code
Open, Easy To Scan QR Code
That Shortened URL Into Your Favorite QR Code Generator!
I like to use QRSTUFF because you can download hi res versions of the QR Code that will print nicely. Save the file and give it to your publisher and every time someone scans that flier you will be able to sort by Campaign in Google Analytics.
View Comments (44 Responses) | Categories: Analytics, Google Analytics, Social Media
Posted on May 3, 2011 by Christina Keffer
Recently, I spoke at SMX Toronto about three under-appreciated or often overlooked video ranking factors and how to optimize YouTube videos to encourage the them to show up in universal rankings.
One of the factors I talked about was one that any SEO should already really be familiar with: Inbound links.
It seems to me that, either because of the unique format of a YouTube video page, or for some other reason, people forget that it’s just a web page like any other web page on your site or any site. When you want your site to show up on the SERPs and you’ve already taken care of any on-page issues and optimization, you build links.
Fortunately, building links for YouTube videos is way easier than building links for most types of content. However, there here are some ways to smooth the path:
Finding Potential Links for Your Youtube Video
One of the best examples of low hanging fruit in the link building world is competitive analysis. Finding out who is linking to your competitors’ sites and then figuring out who of those people would also link to your site. You can do this kind of analysis on YouTube as well, to a certain extent. The major difference between doing competitive analysis for a YouTube video page is that you can’t use the same tools you’d use if you were analyzing the backlinks to a competitors’ website. If you did an Open Site Explorer report on Youtube.com, it would be worse than worthless, and doing it for a specific video page on YouTube is also not that great. Same with Majestic SEO. So how, might you ask, can you find inbound links to a competitor video that’s ranking really high? There are a few ways:
1. Using The YouTube Share Button for Competitive Link Analysis
Under the video that you’d like to research is the YouTube Share button.
When you click on it, you get the YouTube shortened URL. Next, simply do a Google search to see who has that link on their page. The resulting SERP has a lot of social media sites listed on it, but if you scroll past the Twitter and FaceBook results you get to some nice blog pages and such that offer dofollow links.
The awesome upshot of this is that you can get a whole boatload of potential twitter followers and facebook fans. After all, if that Tweeter is sharing this particular link then why not yours? Talk about views….
2. Researching YouTube Embeds
Very similarly to the link research, you can also find out who has embedded the competing video. Under that share button, there is also the option to Embed:
Copy that code and paste it into a Google search and you get the following list of sites that embed your competitors’ videos. If these sites don’t all belong to said competitor, there’s a pretty good chance that they will embed your video as well (and include the embedded YouTube link as well.)
3. Getting Links From Within Youtube.
Just like with every website, internal links matter too. Here’s a good way to find out who all is linking to the competing video from within Youtube:
- In Google, set advanced search results to display 100 results.
- Site Search Advanced Search Operator: site:youtube.com (your keyword) Channel
- Export search results into an excel file including title and meta info. (Outwit is my favorite SERPs collection tool.)
- Contact and start relationships with people in the resulting list.
Here’s an example of the information I was able to grab with Outwit and format in Excel:
The highlighted entries are the ones that I thought would be the most worthwhile to start a relationship with. Why? Look at those channel view statistics. By the way, if you include the word “views” in your advanced search operator search, you’ll get much more detailed information with regards to channel views in the meta descriptions that you can pick apart, at least most of the time.
So there you have it, 3 ways to increase the actual links to your Youtube video page. Go forth and optimize!
View Comments (19 Responses) | Categories: Search Engine Optimization, Social Media
Posted on March 8, 2011 by Christina Keffer
One of the most challenging aspects of being a consultant is being able to convey your progress to your clients (and occasionally, boss) in an easily-digestible, understandable manner. This usually involves paring down reports to their barest essentials — stripping away the extraneous in order to reach a core of actual insightful data. It involves keeping everyone focused on overall trends instead of micro-analyzing on a weekly or even daily basis. After all, most website changes, especially SEO-related changes take time to really come to fruition.
The idea is to avoid the possibility of the client freaking out when traffic drops 30% after a bout of 301 redirects to new search engine friendly URLs. The rankings (and traffic) will come back in a few weeks if everything is allowed to shake down adequately, but it won’t work if the process is reversed due to yanking out the new urls and going back to the status quo because the party in charge was obsession over daily or weekly traffic levels instead of looking at the bigger, overall picture.
In this post, I’ll talk about data presentation and the most important success metrics (as I see them) for SEO. The examples I’ll use are real, albeit sanitized for client confidentiality.
SEO Success Metrics
Usually, when a client comes to us, they’re utterly obsessed with rankings. They have a shortlist of about 3 or 4 keywords or phrases, usually highly competitive head terms, and those terms are all they care about. They are preconditioned to judge the efficacy of SEO efforts by this one metric. It takes a long time to wean them of their dependably on keyword rank tracking.
It’s not that rankings are not important. Far from it. In order to achieve the REAL goal, rankings have to be achieved and upheld. But the 30,000 foot view model of reporting doesn’t jive with multiple reports and goals. In order to distill insight from the data puke that is GA reports, it’s important to pick one, or at most two metrics for success to report on. So lets cut out the middle man (rankings) and focus on what really matters to clients — even if they don’t know it: Traffic and Conversions.
The overall goal of optimizing ones site for the search engines is to get higher quantities of good quality, converting traffic from the organic search engine results page. Once you understand that, figuring out how to set up your reporting is easy. You look at traffic, you look at organic traffic, and you look at conversions and you focus on trends.
Google Analytics Data Presentation vs. Excel
Google Analytics is pretty awesome. To those of us who use it every day, it is intuitive, well organized and offers some pretty awesome ways to slice and dice data. It’s really easy to forget that our clients, who in many cases are not as tech-savvy or comfortable with browser based programs as we are can be overwhelmed by the choices and actually dislike the visualizations that we find so incredibly useful.
I’ve been delivering monthly reports to clients for years in every format you can think of. I’ve created reports on the GA dashboard, exported them into pdfs, ccvs, excel sheets… you name it, I’ve done it. All in an effort to find the best way to show them what they NEED to know without distracting them by what they don’t really want to know. The report that I have the most success with is actually the simplest excel report imaginable. It’s a month to month comparison of overall traffic, un-paid search engine traffic, unbranded search engine traffic and the quantity of goal completions due to search engine traffic. These are the resulting charts I send out monthly:
This graph represents all traffic from all channels on the website. I include it to give clients context for the organic traffic and also because it’s something they invariably want to and should see. It’s important, however to make it clear that this graph does not represent SEO efforts specifically, rather overall performance.
This graph represents organic traffic and unbranded traffic. The importance of organic traffic as a metric for success is pretty obvious. However, I think that the true measure of success for SEO is unbranded organic traffic. After all, a site that’s not ranking for its own brand name, with some notable exceptions, has some pretty big problems. I find it hard to take credit for branded traffic.
This last graph is an example of goal completion broken out by month over time. The spec download was this client’s most important goal. Using GA annotations, it’s easy for me to explain the sudden jump: we created a landing page for the download and optimized it for the appropriate keywords. I don’t have to include the ranking performance of the specific keywords for this report because the results are obvious.
Date Ranges
You’ll notice that I’m using entire months instead of, say, 30 day chunks of time starting on the same day of the week. Does this skew the data slightly? Yes. However, since it’s my goal to focus on traffic and conversion trends rather than specifics, I find this acceptable. There are cases when the difference in traffic from weekdays to weekends is so drastic that I’ll change this methodology, but those cases are rare.
The 30,000 foot view is a great way to spot trouble areas and places where success is really evident (always a great thing to point out to clients and your boss.)
View Comments (5 Responses) | Categories: Analytics, Search Engine Optimization
Posted on January 13, 2011 by Christina Keffer
Opinions are like…well we all know what they’re like. Everyone has one. Here is my opinion about the link anchor text debate that has been going on since Pubcon ’10. (At least that’s when the debate popped up on my radar. Apparently Greg Boser’s been talking about this since 2009)
In one corner of the ring, we have those who obsessively build anchor text-rich links with exact phrase anchor text that looks unnatural as all get-out but adds boatloads of relevancy to whatever page the link points to. In the other corner, we have those who seem to have forsaken the building of purposefully keyword rich anchor text links in favor of pursuing any kind of link from high authority, relevant pages.
And then there’s the rest of us who have been patiently working on a nice well-rounded backlink portfolio that includes both anchor text-rich links and domain name links from powerful pages and some directory links and some blog comments and all the lovely natural links that are generated by a nice, good quality link bait generation campaign.
See, in my opinion, anchor text is still really important. It’s hard to get (quality, organic-looking, contextual) keyword-rich links. If it’s hard, it’s not easily controllable, and in my experience, the search engines tend to place a lot of emphasis on elements that webmasters and we evil system-gaming SEOs can’t easily influence.
That being said, I was there when Matt Cutts said that they were going to start taking notice when a website’s link profile contains only keyword optimized links. The thing is, if you’re doing a good job creating link bait, this is never going to be a problem. You’ll have people linking to the site naturally — probably using the domain name, brand name or the word “here” or somesuch.
Those links will far outnumber the ones you spend your precious time on generating. But let me tell you something, if I spend the hour or two (on average) of building a relationship with someone so that they’ll link to a site from a nice, juicy page, you bet your ass I’m going to ask for a link with relevant anchor text.
And I bet, given the option of having a link with good anchor text from an authoritative but non-related page vs. having a non keyword rich link from a page with similar authority, most SEOs will still choose the keyword rich link. I would.
View Comments (10 Responses) | Categories: Industry News, Search Engine Optimization
Posted on January 11, 2011 by Christina Keffer
The other day, I was hunting on the internet for lunch and I stumbled onto a restaurant I hadn’t ordered from in a long time. It turns out that they had added an “order online option” which automatically gave them the edge on all the other Pizza mongers in Pittsburgh (with the exception of Dominos, who has one of the best order online setups I have EVER seen. Unfortunately, I’m not the biggest fan of Dominos pizza.)
The checkout process was like a cheap date: not pretty, but super easy. And at the very end, the best part: A tip calculator!
Now let me momentarily digress. My ability to do math in my head is probably rivaled by that of – oh I don’t know- a crack addicted chihuahua. Because of this handicap, I have a pretty odd method of tip-leaving that is based more on the type of service that I’m tipping for than the actual percentage of whatever I paid for in the first place.
Usually (always….) this really works out well for whoever I’m tipping. For instance. Bartenders get $1 and whatever change I get for every drink, even if it’s just Coke. (If I don’t have a tab. If I have a tab, the tips probably average way more per drink since by that point I’m a drunk crack addicted chihuahua.)
For dinners, it’s typically between $10 and $20 depending on how often my water glass gets filled up, regardless of whether I’m at Tom’s Diner or Bravo’s. If I’m eating at a really nice place, it’s likely that the Boy is paying, and I needn’t bother about such things at tips. For delivery drivers, it’s usually at least $3 if not $5, depending on the weather, the challenges my location poses, the length of time they have to drive etc.
Now back to the story at hand. This awesome tip calculator not only calculated my tip based on percentage of what I paid for my pasta, but also allowed me to simply click on the right number. In this case, 20% was all of $2.90. 20% is also, well, a fair tip for someone who just had to cross a bridge to get to me on a relatively nice day. So that is what I left them.
Had this nifty tip calculation feature not existed, that driver would have gotten a much higher tip. This got the hamsters in my carb-starved mind sluggishly running in their wheels. This poor low-tipped driver might only be an example of a greater issue. As website usability testers (as well as SEOs and analysts) we deal with usability every day, and the general rule is: the more usable the better. But is this really true? Might occasionally limiting a user’s possible actions actually result in MORE money? Well, probably, yes. In some cases. Like the following:
The Pros And Cons of a Really User Friendly Website
Sometimes, the very things that may make a site super-usable can also have some unforeseen and unwanted repercussions. However, it’s a thin and wavering gray line. Lots of usability experts/the books of information they publish preach that we should include easy, intuitive navigation, lots of leading information etc. and they are totally right. Except when they’re wrong. Look and PPC landing pages for instance. You (usually) don’t want a full horizontal and vertical navigation with bread crumbs and everything else included on the landing page. That would make it super usable, but it might also lead the targeted traffic to pages less suited to actual conversion. They might like what they see, but offering them more choices at that point might lose you the revenue. That is a great example of how usability best practices can lose you money. Here are a few more:
Decision Paralysis
Have you ever been faced with so many options that you decide not to pick anything? I have. Recently. I usually find myself making a midnight run to the nearest Walmart the day before any given holiday to scrape up last gems in the picked over card racks. It normally takes me 10 minutes, because there are a limited amount of cards that don’t have moving parts or sound effects left. It makes my decision easy. Two days ago, I was being completely, uncharacteristically, ridiculously forward-thinking and found myself looking at Valentine’s day cards.
Faced with an entire AISLE full of cards, I found myself completely incapable of making a decision. They were grouped together, but only by relationship (daughter to mother, husband to wife, child to parent etc.) It would have been way more useful if they had been subdivided further into genre (daughter to mother/funny or husband to wife/ risque) That would have limited the quantity of cards that I would have seen initially, but would have made the entire process more manageable, and I would have walked out of there with a girlfriend to boyfriend/funny card and no problem.
There’s a very fine line between too many easily accessible options and not enough as any medium to large sized e-commerce site webmaster will tell you.
Lower Average Conversion Value
This idea of usability adversely affecting conversion value brings us back to the thing I noticed with the tip calculator, but there are many other cases where additional choices may lead to lower conversion size. Here are some possibilities:
- Service Package Size: Many different packages of services. Lots of SAAS companies have this problem. They have the “Deluxe” package, followed by the Really Deluxe, Super Deluxe, Awesome Deluxe, Super Awesome Deluxe, Giant Enterprise Deluxe with a Cookie, and maybe some more after that. One one hand, breaking out services in this way creates more customized packages that are likely to be attractive to almost every conceivable type of client. On the other hand, offering too many graded choices will often cause clients that would potentially want the Awesome Deluxe package to choose the Super Deluxe one instead because they want to save that extra $50 a month. Users typically choose the middle of 3 packages. You can lump your services together however you want behind the scenes, but don’t sacrifice order size just to try to reach everyone at once on the conversion page of your site.
- Shipping Options: Offering a half dozen shipping and insurance options and carriers is often unnecessary and may cost you revenue, even if the process of choosing these is dead easy and intuitive on the website. Better by far to include insurance (if necessary) in a flat rate shipping cost and add one other rush option at a premium (or a similar two choice process.) If the user made it all the way to the shipment selection portion of the checkout, they’re rather invested and as long as the cost isn’t way outside of their expectations (which you can manage along the way), they will happily pay. You don’t WANT them opting out of an option that can make you money, and by providing those options you’re inviting them to do that.
- Loss leaders: A website is not a restaurant, and the happy hour model is not always effective. Faced with the choice between an inexpensive single product and a higher-priced gift basket containing that product on the same page, there is a much lower likelihood that the user will choose the more expensive gift basket. Giving them a lower cost alternative in the menu might be a reasonable alternative.
Information Overload
Everyone knows copy is really important to a website. From an SEO perspective, it gives the search engines something to index. From a usability perspective, it gives the user needed information to encourage them through the buying funnel. From a… well you get the point. You need copy. Arguably, on a truly nice and usable site, the copy will be formatted in a way that is not overwhelming (broken up by pictures, bullet points, ordered lists etc.) However, sometimes, the sheer volume of product description can actually turn users off. Make sure that you’re taking all phases of the buying funnel into consideration when designing how massive copy will appear on the site. Take care of your impulsive, ready-to-buy clients up top and then let the other information settle into tabs or at least below your strong calls to action. Otherwise, you risk overwhelming your user.
Hopefully this post was somewhat helpful. Even though the common sense, often repeated maxims are present, I thought it was interesting to look at how, even if you do everything right, you might still lose money. I think it’s helpful to occasionally remember that there are two sides to a usable site. You want to make sure that you have a site that CONVERTS.
In other news, I had a hard time finding good, general examples. Anyone know of any sites that are so usable you want to hit them with your keyboard? Thanks!
View Comments (10 Responses) | Categories: Conversion, Usability
Traffic
Analysis
Action
Subscribe to this blog's feed