TruthyPR

Six projects to do before you do a website redesign: Learn what the user craves to hear

[This is the last part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

The currency of modern online marketing and outreach is content.  There are a lot of other words for it: conversation, social, engagement, storytelling, but it all comes down to one key thing:

You need to tell a compelling story about the work you do that communicates your excitement and passion.  And you need to be able to do it on a regular basis.

To this end every nonprofit (and every brand of any kind) is currently struggling with how to tell their story.  They've been doing broadcast boilerplate so long, they've had to relearn what it means to talk about themselves passionately. 

Figuring out your voice, and what your users crave to hear, is a key part of your strategy that you need to figure out before you do a redesign.  Why?  Because if you do a redesign and then discover that your users want different content than what you thought, your new site navigation, layout, and your entire information architecture will need to be redone.  After you've spent a bunch of money to do it once, do you really want to do it all over again?

Here's a simple process for experimenting with new content that you can do in order to prepare for a redesign. 

Start by applying analytics to your email list

If you've been running an email list for a year, you've probably got a large set of archived emails and their data around.  Pull that list and go through the process of categorizing the content in each email. 

Use your email analytics product to pull reports of people who have clicked something in the last 30 days, in the 30-60 day range, and in the 90+ days category.  These will be your test subjects.

Now pull the content analytics from your website

Another great source of information about what content of yours is most popular is your inbound links, something we've discussed in great detail in previous articles.  Go into Google Webmaster and download the list of your pages fed by external links.  Or Google Analytics and look at your landing pages that are fed by "Referral Traffic" that aren't search engines.

Pull them into Excel and categorize them in classic card sort technique by their topic.  Most of what you write probably falls into just a few buckets.  Is there something that you no longer produce that has a sizable number of inbound links or traffic?  A-ha!

Conduct a simple, telephone interview

Now reach out to the users in each email history category through something simple and cheap, like a personal email, and ask them to talk to you about what they like in your email newsletter.  If you have their phone numbers, you can call them. Ask them what they remember clicking on.  Does it match what they actually clicked on?  Ask them if they read an email newsletter from a similar organization and learn what they like about the others.  You won't copy it, but combined with your domain expertise it should give you ideas.

I've actually been doing this with the latest ebook about measuring Facebook that Shayna and I wrote.  Just randomly, I've been calling people that downloaded it, apologizing for the intrusion and asking for a few minutes for feedback.  Most of them are stunned by this personal touch and complete lack of sales pressure.  When they realize I really just wanted to know what they thought of it, they're more than happy to help me.

Half of them haven't read it yet, but the other half gave me particularly good feedback that Shayna and I are using for our next ebook.  In fact, I have learned that about a third of the material in the ebook was not really well-targeted to the audience.  Surprising?  Useful?  Valuable and inexpensive Research?  Yes Yes and Yes.

Start brainstorming new content and testing it in email and Facebook

At this point its time to come back to your domain expertise.  Learn what it is people might like, and start drafting some examples.  Use them in an email and check the clickthru rate, or post them on your Facebook page and watch the Facebook.com-sourced traffic click through to it.  Over time you'll find some new ideas for your content.

Conclusion

This process is a good one to try on a regular basis, perhaps once or twice a year, in order to avoid getting stuck in a rut doing the same thing without external input.  It requires listening though, which isn't a natural skill for everyone.  Making sure you've done it before a redesign is crucial, though.

Six projects to do before you do a webside redesign: Learn what the Internet sees

[This is the fifth part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

Your website today, no matter how crappy you think it is, has a steady stream of people that come to it, and in some percentage, do what you ask.  They are some of your best traffic.  They came for one of two reasons:

1. Because they searched on a topic in a search engine that didn't include your name, found what you had to say compelling, and then gave you their email address or money; or
2. Because someone on the net linked to one of your pages, which they came to and found what you had to say compelling, and then gave you their email address or money.

Think about the last redesign you did, or one of your colleagues did.  Did anyone spend anytime talking about preserving the incoming link structure?  Did anyone prioritize that list by the largest numbers of backlinks, making placeholder pages or 302 directives to redirect people to the new website?

Did anyone say, "Hey, we're ranked in the top 10 for these topics, we better do our homework to preserve our Google rank when we put up the new design"  ?  Probably not.  Many agencies that do redesigns (and the clients that hire them) are concerned with mockups, designs, content management systems, and content migration. 

And you paid an agency your hard earned money for this treatment?

How to find and preserve your inbound links through a redesign

During the redesign process, and preferably at the beginning, you should analyze your site's existing inbound link structure using a tool such as Google Webmaster.  Once you get your website setup in Google Webmaster (GW), you will have access to Google's spider data about your website.  In short, it will be like Google crawled the entire Internet for you and found everyone that linked to you.

Once you do this, you can download a spreadsheet of all those incoming links From GW.  Go to the section "Your Site On The Web" and click "Links To Your Site". 

Up will come a spreadsheet showing how many inbound links there are to your website and to what pages. You can also download the list of all links at the bottom.

When you click on the number on the right, it will bring up the detail of where those links are coming from.  Once again, you can download that table of links at the bottom.

Wow, that's a lot of links to keep updated!  How do I do that?

The easiest thing to do is to have your technical staff maintain the exact page address of anything with an incoming link through the redesign.  If this isn't possible, have them replace the page that was at that address with a server side 301 redirect.  Despite that sounding complicated and technical, it's actually not that hard for someone who configures webservers for a living.

If all of that doesn't work, then you're stuck using one of the less satisfying alternate methods.  There's a discussion of these in the Google Webmaster forums.  What you should not do is create duplicate content, though.

A great deal of your traffic comes in from your existing backlink network, and its worth it to preserve as much of it as possible.

How do I find my well-ranked search engine pages before a redesign?

These are also important.  These are your organic search engine results, and you need to do your best to preserve their rank.  There are three steps:

1. Identify them;
2. Preserve their web address for inbound links; and
3. Preserve their Search Engine Optimization-friendly (SEO-friendly) qualities.

Identifying them can be done through any tool, including your analytics product.  Just produce a report of your top landing pages generated from organic search results.  Preserving their page address is the same process as I describe above for any page with inbound links.

The harder process I can't explain is the preservation of the SEO-friendly qualities of the page.  Keyword density, use of keywords, page crafting and other subtleties are part of the ever-changing SEO landscape.  Any list I would provide would be incomplete.  My advice is to retain a professional for this task if you have well-ranked pages, because if you do, you want to keep them.

You won't get lucky in your redesign if you don't prepare

It may be tempting to press on through your redesign and hope for the best.  Don't.  Google rank and link networks don't maintain themselves without effort, and when you're examining your traffic and wonder why there are less visits, donations, and email signups than before the redesign, this will be one of the culprits.

Six projects to do before you do a webside redesign: Learn what the user can and can't see

[This is the fourth part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

I have an assertion for all of you considering a redesign of your website:

Some significant portion of your web visitors can't see, or don't see portions of your website.  If you redesign the website without knowing what they aren't seeing, you may put crucial information in their blind spot in your redesign.

There are three reasons for users not to see the information you put on a page:

1. They physically can't see it.
2. They don't see it because it's not prominent.
3. They don't look at it because they've been trained not to.

You need to understand the blind spots in your website before you build another one, or you may put information in a blind spot even with the help of a usability expert.  You also need to understand how people read on the web.  Here are the reasons a visitor may not be seeing your content.

Can't see it: Blind visitors

If you're blind, you can't see the images.  If you don't tag your pages carefully to facilitate screenreaders for the blind, these your visitors will miss out.  It's not enough to just add alt tags to all your images, you need to attempt some level of 508 compliance.  The US Government has some great tools for you at section508.gov.

Can't see it: Too small text

If you're text is too small, your visitors may not see what you've written.  You can facilitate this with font sizing functions on your site that let the user make the font bigger.  The major browsers also build this into their functions, but many visitors won't know they've missed it.

Can't see it: Color blind

This affects more people than you think.  There are several different types of color blindness, and you only need to render something difficult to see (as opposed to invisible) in order to get tripped up by this and affect the effectiveness of your site.  Use a tool that shows you the filtered version of your website to spot color blindness problems.

Can't see it: Screen size too small

Do you have a solid understanding of the screen sizes of your visitors?  While you may not want to design your website to satisfy the 5% users that only have 600 pixels of width to view, you need to know enough not to put the most crucial information on the far right.

You can get screen size information from your analytics product.  You can also see an immediate overlay of the site with Google Browser Overlay.  If you run this on truthypr.com, you'll see why I moved my call-to-action column to the left.

Don't see it: not prominent

You can also bury important information in your website by cluttering it with too much material.  I suggest running an eyetracking study on some of your competitors before doing a new design.  People typically advise running a study on a design before you implement it, but a design that only exists on paper (or hasn't been worked into a CMS) is still very theoretical. 

If you run a study of a competitor website, you'll learn the secrets that are specific to your issues, as well as learn a little about what your competitors don't know about their own audience.  Of course, if you have the money, run it on your new design as well.  Just don't be surprised when the design changes six months after implementation and another eyetracking study reveals a problem.

Don't see it: it looks like an ad

This is something I see commonly on nonprofit websites.  Someone wants to promote something and so they build out a graphical promotion, maybe even an animated gif, to look like an ad, and drop it into every page on the site.

The problem with this is that web visitors have been training themselves to ignore ads for years.   They gloss right over them, which is why the Internet Advertising Bureau has standardized larger and larger ad sizes to be embedded into pages.

Understanding how your existing website, no matter how awful, really works is a key step to creating a better one.  These suggestions won't break the bank, but will help you understand what you need to do.

[This is the fourth part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

Six projects to do before you do a webside redesign: Learn what the user can't do

[This is the third part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

I have an assertion for all of you considering a redesign of your website:

Your website is probably failing you in some way, and you don't know what it is.  If you redesign your website without knowing why it's not fulfilling its goal, you're going to waste a lot of money.  You're redesign is unlikely to randomly fix the problem.

Let me be clear about what I mean by fail: websites that don't have clear goals defined are failing their owners.  And websites that have clear goals but can't get sizable quantities of visitors to complete their goal process are failing their owners.

If you either don't have a goal established, or you aren't measuring that goal, the redesign will be a waste of money and you're website will still be a failure.   So let's assume you have goal already and a notable number of visitors aren't completing it: how can you learn what your users can't do?  Let's start with the simple ones:

Check your site feedback email

I can pretty much guarantee that if you have a problem completing goals on your website, people have complained about it.  Find the person who suffers through the site feedback email and ask them to show you what there is from the last three months.  I remember I once started a relationship with a new nonprofit and asked about their high abandonment rate on the default multipage donate form.  I asked "Do people ever complain about how many steps it takes to give you money?"  The response I got "Oh yeah, all the time, people hate that."

Well, I said, let's change it to a shorter form.  We did.

And the completion rate tripled.

Check your "user voice" tools

I won't repeat the instructions here, but if you installed a tool like 4Q to survey your users, you're set.  A user survey tool will tell you what they came to your site to do and if they could complete it (or why not).

Study Your Goal Funnels

All analytics tools allow you to see what happened to users on their way to your goal, but failing to get there.  Look at this goal funnel below for my first fundraising ebook.    On the left are listed pages that drove people to the download and registration page.  On the right are listed the places people went when they didn't finish and download.
Here's what those links on the right tell you:

• Some just quit (exit), thwarted by the requirement to register.
• Some went to my other ebook about measuring facebook.
• Some went to the about page to see who the heck I am.
• Some went back to the homepage.
• Some went to an article I wrote about measuring social media.

Though you'll never have a 100% conversion rate for a page like this, you can always study the reasons people don't finish by looking where they went and attempt to raise that completion rate by addressing those concerns.

And the project to solve that problem is a lot cheaper than a website redesign.

Six projects to do before you do a webside redesign: Learn what the user does

[This is the second part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

I have an assertion for all of you considering a redesign of your website:

Your website already works well in some area and you just have no idea what that is.  And when you redesign your website without researching first, I guarantee you're going to destroy something you will want.

This doesn't mean that you might discover what's working on your website and then subsequently discard it after doing a cost/benefit analysis, but when you just charge into a redesign, you're not making an informed decision.

Things You're Likely To Break In Your Current Website When You Redesign

Here are some examples of ways in which a website works that you:

• Search Engine Optimization (SEO): You might have incredibly good ranking on a number of search terms that put you #1 in Google and bring you extremely well-performing traffic.
• Heavily linked content (related to SEO): You may have an online presence so old that enormous numbers of people have working, healthy links to your content.
• High prospect rate: Your website may have a design that refers people effectively to the donate page (or your main ask page, whatever it is).  When you redesign, that rate may drop and so will your donations.
• Low abandonment rate: Your donate page may have extremely low abandonment rate, such that most of the people that get there also complete your ask.
• Domain authority (related to SEO) : You may have a very old domain with a high Domain Authority.  Without knowing it, this could be contributing to high Google rankings.
• Existing pro-bono campaigns: You may have existing campaigns, perhaps through the AdCouncil or others, that are sending traffic to specific pages on your site.  If you change your content or domain, you may cause those campaigns to cease functioning.
• Ease of use (usability): Your user base may find it easy to find things, and your navigation may be natural.  But if you re-organize without proper research, you could make it hard for your current audience to find the same things, or even make it worse.

If you redesign your website without checking any of these first, you are almost guaranteed to destroy something of value.  It's critical to actually understand what part of a website is working before you undertake a redesign.  The above six items can be checked with the help of a research expert well-versed in the area of web analytics and search engine optimization through a combination of Google Webmaster Tools, Google Analytics, and various SEO tools.

Learn What The Users Are Doing And Avoid Breaking What Works On Your Website In A Redesign

In preparation for a redesign, you should research the following things:

• Your Google juice: What links do you have into you from elsewhere?  What content you've forgotten about that people like?  What terms are you ranking on?  How authoritative is your domain name?
• Your conversion performance: How often do people start to donate?  How often do they finish?  You need to know this, because if your redesign makes your performance worse, you need to revert to early ones or incorporate previous insights.

For your Google rank, you're going to either take a crash course in Search Engine Optimization or hire a guru that can help you assess your current performance.

To measure your donation prospecting and donation completion performance, you can probably do this yourself with a copy of my ebook "3 Fundraising Metrics For Your Nonprofit Website".  I've attached the form below.

The e-book is free! Just sign up for my e-mail list. (The ebook will be emailed to you, so don't give me a fake e-mail address!)
First name	Last name
Organization	Your e-mail
Your telephone

http://www.flickr.com/photos/ebeam/ / CC BY 2.0

Six projects to do before you do a webside redesign: Learn what the user wants to do

[This is the first part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

In the twelve years since I've been doing agency work, here is my entirely unscientific list of the most common reasons people redesign their website:

1. "It looks 'old', we need to update our look" also "We're changing our logo and colors".
2. "We want to use some new technology (usually a content management system)."
3. "We want to create a more engaging experience for the user."

None of these really speak to a deep understanding of the user's desires.  Furthermore, once you understand the user's desires, you may not actually need to do a redesign.  You may need to create more or different kinds of content, or you may need to conduct your work in a different way so you have something substantively different to talk about with your users.  Or you may need to invest in a piece of functionality that would allow your users to immerse themselves in your mission in a more engaging manner. 

But all of those projects are smaller and cheaper than a redesign, and that's not in the agency's interest.

Let's take the stated reasons for a redesign one at a time:

It looks 'old', we need to update our look / We're changing our logo and colors

This is the impetus for a lot of redesigns, but the magnitude of what gets changed is way out of whack for a simple graphics change.  This is one thing to do along the way to a redesign, but it alone is not a good reason.  (These reasons pile up though, as you will see)

We want to use some new technology (usually a content management system)

This is a big one.  It's worth noting that a new CMS doesn't necessarily make things any better from the visitor's perspective.  It's really just about making it easier to post content.  Yes, you may post a little more content than before, but ultimately a successful CMS integration project is invisible to the web visitor, which means they get little, if any benefit.

We want to create a more engaging experience for the user

Probably the most powerful thing you can do before you redesign is to understand what the user wants to do, but in actuality, people go by their gut instinct.  That's not necessary, of course, and the way to do this is to survey your web visitors.  Assuming you don't have a lot of money, it won't be scientific.  However you are not in need of scientifically accurate results to understand what they want, you just need to ask a lot of them.

So how do I learn what the user wants?

There's three tools/practices I'd recommend that aren't going to bust your budget:

1. Random sampling of web visitors for task analysis
2. In-depth telephone/in-person interviews
3. Web-based surveys

Random sampling of web visitors for task analysis

This is "stupid easy" to do.  You want to know the answer to a simple question:

"When the user was on my site today, what were they trying to do?"

Not a hard question.  The easiest and cheapest way to do this is to install the now ubiquitous tool 4Q.  (For a crash course in using 4Q, read my article "How can I survey my users and keep tabs on their website satisfaction for free?")

Once you've done that, you're going to get a large spate of responses to the question "What did you come here today to do, and did you accomplish it?"   Here's a collection of responses to the survey I keep on my website, truthypr.com about why people came to the site today:

This graph said that 80% of the people that came to my site in April were looking for a specific article.  Here's the text of what they said, which I think is the most valuable:

Ah-hah!   While these seem a little like platitudes, the first and third comments suggest that I need to continue working on my writing to carefully translate technical concepts into non-technical explanations.  The third comment emphasizes that frequency is important.

Alone, one or two comments is not anything to base a future decision on, but it does reveal that I should be asking my users "how often do you expect to hear from me?"  "Is weekly enough or not enough?"  And "what tools are useful in communicating these concepts?  Screenshots?  Videos?"

That last comment suggests that when I'm thinking about verticals, I should think about nonprofit charter schools, which is something that never occurred to me before.

Here's another one from a previous month:

The last one emphasizes an "arts nonprofit" vertical.

Once you've got a flavor for why people there, you can formulate questions that will begin to tease out whether or not you're missing something in your offerings, which leads me to...

In-depth interviews by phone or in person

These are the most valuable information you can get.  You want to make sure you get people who are honest and candid with you, which often takes an in-person interview if that's logistically possible.  You're interested in the initially obvious questions:

1. What do you like about the website?
2. What parts do you use most, why?
3. What frustrates you about it?

But then you want to move onto other, more important positioning questions, such as:

• What other organizations in this space do you follow?

I want to stop there, because the value in this is not to simply write down the name of a competing organization and move on.  You already know who your competitors are, and what your interviewee is going to say when you ask that question.

You want to know why they follow them.   You want the interviewee to grade both your organization and the competitor organization on the common mission, and then learn what they're doing better than you.

And then steal those ideas and improve upon them.

This is the most valuable thing you can learn, and will often go beyond the website's execution and to the way that your organization actually conducts their work.

Another question you should be asking is this:

• If you ran organization XXX with a mission to do YYY, what would you do?

You need to let the participant think big, and approach the mission focus with fresh eyes.  If you get a pat or political answer ("Just what you're doing today") then you need to find another participant.

Web-based surveys

I'm not a big fan of using these to get substantive free-form data from the visitor.  Whenever I read raw survey data (yes I am a nerd) I feel like the person was just filling out the form as fast as possible to get it over with.  That's not valuable to you, and it annoys the user. 

And what will I learn from this study of the user?

I suspect that you'll learn something like the following things:

1. Our competitor has a program that resonates better with supporters as being more mission-focused than something we do.
2. Our good work is not being explained well enough, and we need to do a better job of that.

Both of these things can be solved with business changes and editorial changes, but neither requires a whole site redesign. 

Wait, if learning what the users wants to do might not lead to a site redesign, why do agencies always make it the first step of a redesign plan?

Because web agencies like to sell redesign projects and clients like to buy them. 

Redesigns are big dollar engagements and make great portfolio fodder.  Also most sizable web agencies employ a marketing intern whose job is to submit client work for awards, and web redesigns are great award show material, which makes for great publicity.

Unfortunately I've yet to meet a digital agency that does their research and says "You don't need a redesign, let's give back that money".  Usually the agency concludes that the visitors are desperately craving a "better site" and charge full steam ahead with the redesign.  The client is covered from criticism because the research proved they needed a redesign, and the vendor feels justified as well.

If you went to the doctor complaining about a pain in the butt and he said, "We're going to run some tests but regardless of what they say I'm going to start planning a round of chemo for colon cancer", you'd freak out and go find a new doctor.  Why don't you hold your web agency to the same standard?

Do the research, figure out what the user wants more of, and then execute that.

[This is the first part of a six part series on projects you should do before you commit to a redesign.  To learn why redesigns are expensive and often serve the agency much better than the client, read my introduction.]

Don't redesign your nonprofit website! Six projects to do before you do a website redesign

I'm going to tell you a little secret that nobody in the web agency world wants to own up to:

Quite often, website redesigns are not very good for the client

They over-promise, under-deliver, run over budget and over schedule, and don't actually improve the performance of the client website (mostly because nobody bothers to measure the performance to begin with). 

I appreciate that this announcement isn't as momentous as the lead spokesperson for the health insurance industry turning whistleblower, but it's not something a lot of agencies want to admit because website redesigns are the largest engagement (by dollar) that an agency gets with a client.  And this brings me to my second secret:

Five and six figure website redesigns are often cost justified in the name of increased donations, but are never actually measured.

I've seen this promised a lot, but rarely do I see a vendor who even knew what the performance numbers were when saying it.  That should ring alarm bells in your head.

Not all website redesigns are bad, but they can be an expensive and painful process if not scoped down viciously, and for most of the 12+ years I've spent in agency work watching both my colleagues, competitors, and clients conduct them, everyone has a motivation to increase their scope.  I say this with a significant amount of humility, given that I have pitched for, led, won, and executed website redesigns for clients for two thirds of my professional life.

This week I'm going to tell you common myths about website redesigns, and then over the next six weeks I'm going to give you six data-driven projects that you should do long before a website redesign.  These projects might even make it moot to do a redesign.  At the very least, they provide the sound research you would need if you actually wanted to do a redesign well.  These projects are:

1. Learn what the user wants to do: Purpose and task analysis of existing visitors
2. Learn what the user does: Examination of your analytics
3. Learn what the user can't do: Website goal quantification and optimization
4. Learn what the user can and can't see: Visibility analysis of your existing design
5. Learn what the Internet sees: Backlink and keyword preservation
6. Learn what the user craves to hear: Exploring new editorial vehicles

(I should definitely tip my hat to Jono Smith at the Event360 blog and Brad Porterfield from the Latino Community Association.  Last week Jono interviewed me and Brad prodded me to say more on this topic, which led to a beginning of this essay in the comments of my interview.)

So, onto the top myths about website redesigns.

Myth: We're just going to update our look and tweak our brand

Although it does occasionally happen that an organization does a purely aesthetic redo, this is usually the camel nose in the tent.  Suddenly you're also putting in a new content management system.  And then another department wants to add an advocacy tool.  And somebody else wants to create a new content area. 

This happens in nonprofits especially because of the need for broad consensus to move forward with the redesign and the lack of focus on a single goal.  When you start adding items to your website redesign like a Christmas tree, you destroy the ability to judge cost recovery.  Much like a fast moving bill in Congress, people tend to add their well meaning initiative onto the project to make sure it gets done and the end result is awful.

Every item you add to the redesign adds a level of complexity and additional critical points of failure that can delay the project or overrun the budget.  The knowledge that adding scope adds complexity and cost is not rocket science.  But often agencies, especially those with a lot of headcount, will assert that if they start to fall behind they can add more people to a project.

Literally, since the time of mainframe computers, it has been known that adding more people to an already late project makes it even later.  That large agency staff is not going to save you.  I recommend that if you have a number of discrete parts of your website you want to build out, that you execute those in discrete chunks.  This way you can justify the expenditure of the money based upon the needs of the audience.

Myth: Our website looks old, we need a new look and feel

Honestly, that isn't going to do it.  This is especially true now that organizations are judged by the work they do in an ongoing manner and what they say.  Changing your colors or your aesthetic will give you a new look and feel, but it isn't going to cause existing and potential donors to feel better about you.  Changing the work you do, or describing it in a way that is more appealing to donors is what will change their perception of you.  Doing this is "Learn what the user craves to hear: Exploring new editorial vehicles".

(Unless you're website looks like one of these, in which case you really do need some aesthetic work.)

Myth: An easier to navigate website will pay for itself in increased donations, sales, etc.

This is ironic, because vendors don't even ask what your current conversion rate is before they write you a proposal.  You can only make this statement if you know three things:

1. Your conversion rate
2. You existing performance
3. Your likely increase in performance

Let's do a simple math problem:

Question:

You have a nonprofit website that had 10 million unique visitors and took in $100,000 in donations last year.  It converts unique visitors into donors at the rate of 1%, meaning that of every 100 unique visitors, 1 becomes a donor.  If a vendor comes to you and says they can raise your conversion rate by 15% to 1.15% with a redesign that costs $50,000.  Should you do it?

Answer:

Raising your conversion rate, assuming the size and quality of your traffic remains equal, is going to raise your annual donation revenue to $115,000.  That's an increase of $15,000, for which you are paying $50,000.  In a little over three years you will recoup your original investment, and then maybe start to see profits from it. I don't think that's a good use of $50,000 this year.

This assumes that the 15% number is correct, which is surely a blind guess.  It turns out there are mathematically solid ways to increase your conversion rate, and they don't involve redesigning your entire website.  This is described in "Learn what the user can't do: website goal quantification and optimization".

Problem:

You feel the need to redesign your website to increase a poor conversion rate.

My Solution:

Undertake a website optimization project first, you will achieve most of the major gains at a lower cost.

Myth: Our vendor has a methodology to ensure that we produce a website that serves our needs

If you are truly letting your decisions be data driven, you don't sign up for an entire website redesign at once.  I wonder how often these methodologies justify the need for a website redesign?  The answer is probably often, especially if the agency holding the website redesign contract did your research.

I have done website redesigns before where it became obvious midway through the project that the budget needed to be shifted to producing more content focused on user needs and less on the work of the actual look and feel.  In those cases I took money away from a redesign and gave it to writers, some of which benefit the company, and some of which didn't, but I don't think this happens very often.  When I did it as one of the owners of the company with the contract, I had the gravitas to tell my partners that this was the right move.  I doubt that happens very much, though.

Problem:

When you sign a website redesign contract that covers the whole process from research through redesign and launch, your agency has an incentive to skew the research so that it recommends a redesign.  And depending on how hard you fought to get money for their contract, you have a bias to see it that way too.

My solution:

If your vendor thinks so highly of their website research and planning services, do one of the six projects I mention above with them, but don't contract for the rest.  Immerse yourself in the data and see what that produces.  Also, unless your CFO is trying to spend a grant in a hurry, she'll thank you for spreading out your spending a little longer.

Myth: Our vendor has a methodology to ensure that we will launch on time and on budget

Any vendor with a website redesign methodology has clients with horror stories about late projects in which the total cost went up because of added scope, or the things they were able to launch with were reduced, which is another way of increasing the price.  Just ask around.

If you do your "redesign" piecemeal and focus on what you learn about the user before taking the next step, you're unlikely to miss anything, or to run far over budget.

Look for my description in the coming six weeks about each of these projects.  You (or your CFO) will thank me.

http://www.flickr.com/photos/stevendepolo/ / CC BY 2.0

Yield Software: Optimizing paid search campaigns, organic pages, and landing pages

1. Paid Search: It can help you optimize your paid search campaigns (including Google Grant campaigns);
2. Organic Search: It can help you track and optimize your organic search terms, and give you a sense of what terms you should be working on; and
3. Landing page optimization: It has a very convenient graphical interface to let you conduct multivariate landing page optimization.

To do all these tasks, I believe you'd still need an analytics guru, or an analytics-enthusiast staffer around, but if you do, these can help make their time much more efficient.  I've made a short 2.5 minute video with a quick intro to the product by the product manager herself.

Thanks!

Actionable Analytics: Keyword research that brings more traffic to your website

How many times have you had someone give you a "keyword report" from your analytics product and not known what to do with it?  Top keyword lists fall into the realm of analytics data whose purpose is not immediately obvious.  Also in this realm are visitor "traffic reports" and "top content" reports.  Unless you have an active program going to generate traffic from your keywords, then such a report is useless to you.

For example, I once had a client say to me:

Client: "We had xx thousands of visitors come to our website last month.  Is that good?" 

Me: "Do you have any kind of organic marketing program that can influence that number?"

Client: "No."

Me: "Do you have a paid media campaign that gets scaled up or down based upon your traffic?"

Client: "No."

Me: "Do you have any commitments to funders or partners about how much traffic will visit any part of your website?"

Client: "No."

Me: "Then it doesn't matter, you can ignore it for now."

Data like visits, visitors, top content, and top keywords only matter as a barometer for measuring the ongoing work you are doing that may effect them, or if you have some kind of commitment to maintain a particular performance level.  If you have no intention of doing anything to generate more traffic, it's mostly a wasted act to obsess over your traffic reports to learn anything.

Today I'm going to show you where to find the top organic keywords report, which shows what keywords are delivering visitors to your site from search engines.  Then I'll show you how to research these keywords to find similar ones you should be incorporating into your content so that you can get even more traffic.

Of course, every reporting period you can look at the keyword report to measure your progress, which is what makes it an actionable analytic.

Step 1: Find the top keywords/search phrases bringing traffic to your site

This is trivially easy.  On Google Analytics just go to Traffic Sources->Keywords and make sure you click "non-paid".  You should have a report that looks something like this if you select "Landing Page" as the second column item:

If you're using Omniture SiteCatalyst, go to "Traffic Sources->Search Engines - Natural".

Find one or more keywords or phrases that is actually appropriate to you, such as "fundraising metrics" is for me.  If you're constantly writing and publishing, that is an organic search engine optimization program.  How do we find more words similar to this that people search for?

Step 2: Use the Google Wonder Wheel to find similar keywords and phrases

There are indeed a lot of ways to do keyword research.  There are paid tools, and free tools, but one of the first you should look at it the Google Wonder Wheel.  Go take your phrase, like "fundraising metrics" and run it through Google.   Click the "Show Options" link above your search results and you'll see a line of options down the left.  Look all the way down to where it says "Related Searches", and click it.  Your results will look something like this:

From this I picked up the following related search phrases:

1. donation metrics
2. fundraising measurements
3. charity metrics
4. metrics nonprofit fundraising

Copy these down in a list with the original "fundraising metrics" and set it aside.  Now click the Wonder Wheel option right below "Related Searches" and you'll see another way of viewing related searches visually.  (I love the graphical presentation of the Wonder Wheel)  It will look something like this:

The results won't be different, but you can easily expand out to explore related terms.  Don't go too crazy, remember your core term.  After looking at a few of these, I concluded that I should also add "fundraising analysis" to my list.  Let's review the list of phrases:

1. fundraising metrics (the phrase I'm currently using)
2. donation metrics
3. fundraising measurements
   
4. charity metrics
   
5. metrics nonprofit fundraising
6. fundraising analysis

Now we're ready to research them.

Step 3: Evaluate keywords with Google Ad Keywords

The best tool for this is the "Keyword" tool that is built into Google AdWords.  Go get a free AdWords account and find it under "Tools->Keyword Tool".  Copy and paste your list of current and potential phrases into it, and the Keyword tool will tell you incredibly useful things about those phrases, in particular, what is the monthly volume of searches.  Here are the results I got:

This is a great big "DOH!" moment for me.  I have used the phrase "fundraising metrics" in my latest ebook about nonprofit fundraising web analytics without checking to see if that's a highly searched phrase in Google.  Now that I look, I should have used the phrase "fundraising analytics".  That has 10 times more searches than "fundraising metrics" does.

Lesson learned.

Be careful about letting the AdWords tool suggest additional phrases to you.  It casts a very wide net, and you'll find that it suggests phrases that are not appropriate to your content, which can be a waste of time if you're buying ads.  (If you're developing the organic content for it, then it's a fine idea, though it's suggestions can include extremely competitive and generic keywords.)

Once you've identified some phrases that look like they have potential, you can start to incorporate them into your writing, which is the first step of any organic search engine campaign.  Also if you're going to start doing any AdWords purchases, whether its through a Google Nonprofit Grant or through actual budgeted ad dollars, you already know what phrases you want to be advertising on.

Actionable Analytics: Where is your search engine traffic coming from, and where is it going on your site?

We've all seen this keyword report that shows you the top search engine keywords that drove traffic to your website are.  It's found in Google Analytics under Traffic Sources->Keywords, and once you click "non-paid", it shows you all your organic keywords that are generating traffic from search engines to your website.

This report contains one of the great fallacies of modern web analytics, which is that it defaults to displaying this information ranked on this page by things like visits, pages per visit, and time on site. 

Nobody puts up a website to maximize "time on site" or "visits", they want results!  So click on that tab labeled "Goal Set 1" where you've configured Google Analytics to measure things like sales, e-mail signups, and leads, and then resort your top keywords by "Goal Conversion Rate":

Now this is useful.  Which keywords are most "profitable" for you, yielding the results that justify maintaining your website? 

This is useful information if you're buying keywords because that would tell you how much to pay for them, and whether or not you should buy more or less of them.  But we're talking about organic keywords, now, and organic keywords don't exist on their own.  Every organic keyword exists in tandem with a landing page on your site.  After all, if you like (or don't like) your organic search results, and want to either emulate or improve them, how would you know where to start?  Which of your pages are performing well that you can scrutinize?

The solution comes in a feature rolled out in the latest release of Google Analytics.  Select "Landing Page" from the green-highlighted pulldown menu:

The resulting chart is almost perfect, except that there are a lot of single visits with perfect conversion metrics.  That's too small a sample to draw conclusions from. 

Use the "Advanced Filter" at the bottom to drop out all keyword-landing page combinations that are too small.  While you should use a higher threshhold, I just dropped out every combo with less than 5 visits.  Using the filter is easy, just click "Advanced Filter", and set Visits > 5.

 

.....

The final report is perfect. 

You can use it to look at which combinations of landing pages and keywords are working well, compare those to pages that aren't working well, and either re-edit the copy on the landing pages, change the ask on those pages to raise it's yield, or use the techniques that seem to be working on those pages to create new landing page-keyword combinations.

If you're interested in spending more time on increasing your search engine traffic, and you should, since it's often your best performing traffic, I'll be trying out and reviewing a new product in 2010, Enquisite, to optimize organic search on a number of clients.  (And yes, the people from Enquisite are giving me free access to their tool to try it out.)