Blog & Podcast

Einar and I talk about our approach to the Gary Vee content model—the inverted pyramid approach to creating a ton of content and distributing in every way you can. We talk about how primary content gets broken down into micro content, essentially ads that you put out as posts on social media platforms that point back to your main content. We talk about the role of tertiary content—the long-form posts we write that few people read—and why it shouldn’t be viewed as primary. I go into an example of my in-laws’ bookstore, which has taken this model and seen great success as a local business. Even though this is outside of our client base, it’s a really easy example to grasp and it shows that you don’t need much in the way of original content to draw people to your website and achieve your goals.

The way most people do web navigation is totally wrong. They organize their content by the type of media—wrong. They use drop-downs under each primary category—wrong. They carry over long menus onto mobile devices with small screens—wrong.

How do you avoid all this wrongness? Continue reading this post.

On Episode 8 of our podcast, we talked about the following best practices in website navigation:

1. Be descriptive. Don’t use abstract menu titles like “products” or “services.” Instead, use specific product categories like “compasses” or “maps” and specific services categories like “custom cartography” both to spell out what your website is about for users and for search engines that are looking for specific keywords and key phrases.
2. Avoid format-based navigation. Our think tank clients do this too much. No one is looking for a white paper or a “legislative action plan,” they’re looking for information on a particular topic. Your navigation should focus on the subject matter and guide the user to your most easy-to-understand content on that subject first, giving the option to dive deeper only if they want to. Don’t serve up the 50-page essay first, serve them the 30-second clip of the 60-minute panel discussion about the 50-page essay instead.
3. Avoid drop-down menus. But how can anyone get anywhere without seeing the 12 subcategories we have under “research areas” or “programs” without a drop-down? Either include those subcategories on the “research areas” or “programs” page, or embrace the one acceptable form of drop-down, the mega menu. These menus lay bear the entire navigation of your site, or a whole section of your site, with one mouse-over. Users like them because they make drilling down to exactly what they want fast. We’re not just asserting this stuff, it’s born out by usability studies. Here’s an example of a mega menu from Moosejaw, a clothing and camping gear company:
4. Use human psychology to your advantage. We tend to remember the first thing we see/hear/read and the last thing. These are called the “primacy effect” and “recency bias,” or so we’re told by people who right “turns out” type of articles on user behavior. Comedians have been taking advantage of this for decades, planning out their sets with their best joke last and their second-best joke first, leaving the audience with a positive impression as they’ll forget most of the material between those bookends. Do the same with your menus. Your most important offering should be in that top or left position, while the second-most important should be in the bottom or right position. An example might be “research areas” in the first position while “make a donation” sits at the end of the menu.
5. Always. Be. Optimizing. Remove items that rarely get clicked if they aren’t critical. Rename or relabel things that rarely get clicked if you think they’re important enough to keep around. Move items that get clicked often high up in the menu order. The most important thing is to actually measure this behavior and react to it. Most website owners/managers aren’t doing that at all.
6. For the love of all that’s holy, pay attention to mobile. Don’t be a basic b*tch living like it’s still 2008 and fail to plan for mobile. Letting your CMS stick in a hamburger menu—those three horizontal lines that are supposed to mean something on their own—is really failing your users. Throw them a line and put the word “Menu” or “Topics” or “Policy Research” or something relevant next to that icon so their brains have something familiar to latch onto and guide them to the next action. Also, edit your mobile menus down to what is used on mobile. If your users aren’t visiting certain areas of your website on mobile, or are using them less frequently, don’t include them in the handful of links you stack into that menu.

All of this can be summed up by saying “put the user first.” Don’t try to push some pet project on users through your menu—user your blog, social posts, podcast, videos, and email for that. Don’t try to make your menu clever—be clever in your writing or over cocktails but make menus literal, boring, direct, and stupidly obvious. Use your website on a bunch of devices and try to get to a specific category of content using only the menus. Note when your hand tenses up and your heart wells with rage, that’s a little voice inside you telling you that your menus are awful. Fix them post-haste.

All of this should result in more time on site, more clicks per session, and more users hitting the goals you have for them, like downloading your research, buying a product, or signing up for your lists. All good things.

As we discussed in Episode 7 of our podcast, roadmapping is a strategic planning service that we offer in place of free estimates. You can think of it as architecture for your website. You wouldn’t accept “let’s just start building” as a plan of action for building a house. Instead, you’d hire an architect, or at least use a set of plans that you know had produced a good home for your needs and fit your local environment.

This probably sounds logical enough, but it’s hard to imagine how the analogy plays out. So, how does roadmapping actually help our clients?

1. The main thing it achieves is getting the right diagnosis of their problems. For example, we’ve had clients come to us because their WordPress site is ugly, counterintuitive, slow, and staff members hate it so they need something better. We’ve also had similar clients come to us begging to switch to WordPress because their current site is ugly, counterintuitive, slow, and staff members hate it. The problems are the same, but the diagnosis are totally opposite. Roadmapping solves this problem by getting to the details of those problems and determining if the problem is really the software platform, or its implementation, design, custom modifications, or something totally different. Roadmapping seeks to get the diagnosis right, rather than move fast on a client-proscribed solution—the “let’s just start building” approach.
2. Roadmapping also means a much smaller commitment, both in terms of time and money for ourselves and our clients. This means we get to know each other and learn if we agree on the direction of a project before we commit to actually building that project. So far, every roadmapping client has decided to move forward with some version of our plan and have found the experience incredibly valuable, but it does give clients the opportunity to work with us in a smaller way without making the commitment to pay for us to execute a plan and actually build their website.
3. Clients start much more confidently after a roadmapping session, especially those who have been burned by a previous contractor. By really getting to know your organization, we have a much better chance at building the website you need, rather than working from a template and hoping it satisfies some of your actual business needs, which is a standard operating procedure for most agencies. They sell the same website that solves maybe 60% of the client’s problems over and over and over again. Clients who have been through that process want to avoid repeating it and can breathe a sigh of relief after roadmapping.
4. A roadmap is also a portable plan. It’s something that has value on its own, not just as a plan for our specific team. A roadmapping plan is specific enough that it can be taken to any agency and give them a clear path to success for your organization. It’s like having plans drawn up by an architect with the option of shopping the build around to multiple construction companies.

This is why we start every project with roadmapping. It produces a plan of action that more accurately addresses the client’s real problems, it decreases their risk in hiring us, it inspires confidence when the build process begins, and it allows that build process to be taken on by ourselves or anyone else.

I recently signed up for a trial subscription of The Boston Globe. I don’t want to continue my service, so I visited the Boston Globe website and signed into my account to cancel the renewal. But of course, that would be too easy.

Try as I may, there was no way to cancel my subscription via The Globe’s account settings page.

So I resorted to using Google. I searched for “unsubscribe from the boston globe” and the first link brought me to a page that said:

We accept cancellations for digital and home delivery subscriptions by phone only.

Why does The Boston Globe accept cancellations only over the phone? The answer is of course because they want to give you the hard sell before they let you go. “Get another six months for only $5.99″ or maybe “For just $19.99 you can get 12 weeks of the Sunday edition along with a tote bag.”

This is, of course, an awful customer experience. Similar to the song and dance you go through every 12 to 24 months to bring your cable bill back down to the introductory subscriber rate. We all hate these phone calls.

The Globe also realizes this, in fact, they are counting on it. Our hatred of these call means we’re unlikely to make these unpleasant phone calls to begin with. Essentially, The Boston Globe is forcing you to make a calculation: do I spend 10 minutes having an unpleasant conversation, or just fork over another $4.99?

But this means there are no longer in the business of selling newspaper, instead, they’re holding your credit card information ransom at the bar end of an annoying and degrading phone call.

This practice is not unique to The Boston Globe, in fact, is become all too common in the newspaper business. It’s become so common, it’s one of the reasons why consumers are hesitant to subscribe to any newspaper. They assume it’s a lobster trap—it’s easy to get in but it’s almost impossible to get out.

This practice violates one of the four credibility factors identified by the Nielsen Norman Group’s Jakob Nielsen. That factor is called “upfront disclosure.” I find this to be awkward phrasing, but it’s a way of stressing not only being honest, but also totally open and straightforward with website users. This means not omitting key information, no fine print, no tacking on fees or extra charges, and it especially means making it easy to cancel your account or delete your data.

A “cancel by phone only” policy doesn’t even pass the sniff test.

By contrast, The Washington Post makes changing your account incredibly simple. By logging into their website you can cancel your account, suspend your account, change your payment options, or do anything else that a consumer would expect to be able to do on any modern, non-hostile website.

Similarly, Google has worked hard to fight against the perception that it is a lobster trap for data by making it easy for consumers to permanently delete accounts and personal information. I can recall when this was started many years ago by an internal Google effort called the “Data Liberation Front.” This group worked as an internal evangelist team dedicated to the notion that users should be able to control their data, which included deleting it entirely.

This may seem counterintuitive at first, but it makes sense that consumers are more likely to use the entrance to your product if there are clearly marked exits.

We should all audit our own businesses for practices similar to this. You may not be doing something as blatant as the Boston Globe’s policy, but you may be making it harder for customers to leave, thereby making it less attractive for new customers to enter into a new relationship with you.

One practice we’ve identified in our own business is web hosting. Even though we support it for a few legacy accounts, we no longer offer hosting to any of our clients. That’s because we never want clients to feel like they have to continue with us because we are holding their website hostage. So even though it may take longer for both us and the client to set up web hosting under their own name, we find it’s worth the greater upfront costs to give our clients a greater sense of security and the feeling of true ownership over their own sites.

We want clients to know that they can walk away from us at any time. It’s always better for each client engagement to stand on its own merits and not to be influenced by long-term contracts or other hostile business practices. That way every engagement has a clear sense of mutual benefit, rather than lamentable obligation.

We never want to plant the seed of doubt in a client’s mind as to whether they are staying with us because our work is good or because we have some leverage over them.

Long-term customer relationships are not built on “cancel by phone only” policies. That’s the kind of tactic employed by a jealous boyfriend, not a self-confident business.

So be confident in what you provide to clients or customers, and leave the lobster traps to fishermen in Maine.

Modern specialized hosting offers users automatic software updates, daily backups, one-click restore of those backups, and additional security focused around the specific CMS (like WordPress or Drupal) the hosting supports. Services like WP Engine, Pantheon, or Flywheel also offer realtime chat support, setting them further apart from their discount counterparts that usually offer slow-response ticket system or outsourced phone support.

All of this is available for between $20 and $30 a month in most cases, which may seem like a lot compared to $10 discount hosting, but that discount comes at a steep cost.

Essentially, discount hosting is like buying your child a goldfish—you’re setting yourself up for tragedy and heartache. That’s because your discount hosting is bound to result in one of three things:

• Site breach because of out-of-date, vulnerable software
  • This could mean defacement of your site, causing embarrassment
  • This could also mean data theft, alienating your customers
• Data loss because of a software update gone wrong or user error
• Software slowly becoming incompatible with un-updated server software

Site breaches are more common than you might think and don’t require your site to be an explicit target. That’s because most website attacks are carried out by automated bots, programs that scour the web looking for outdated sites running software with known vulnerabilities. These bots some deface website for lulz, fill them with spam links to sell fake Louis Vuitton handbags, or steal customer data, often to spam them. None of these are great realities for your business to face, especially when these problems can happen anytime, including when you need your website to be collecting leads, displaying your work, taking donations, or otherwise serving a vital business function.

Data loss is very common for modern websites that have no backups and can happen for a wide variety of reasons. One of the most common is a simple user error. With so many people in an organization editing websites that employ modern content management systems like WordPress or Drupal, pages and posts can find themselves deleted by mistake. Though these systems employ trashcans or some other metaphor that prevents immediate and irreparable deletion, we find that clients somehow find a way to delete data permanently all the same. Having some way of rolling back deletes outside of the system, provides an additional layer of insurance.

Software updates can also cause data loss when an error happens during the update process. This is why we back up our clients’ sites before we apply any patch or update and always test after an update is applied. This is just a good common-sense procedure but is unavailable with discount hosts that offer no backup process.

The most common problem is software slowly aging underneath your updated CMS. Websites run on layers, or “stacks,” of software, each supporting the next layer. Websites running Drupal or WordPress, for example, rely on Linux, Apache, MySQL, and PHP or a variation of those basic layers to support the CMS. Discount hosts don’t keep these layers updated for you, which means that after as little as a month has passed, your site is running on outdated software. When major version changes occur, such as the recent switchover from PHP 5.6 to PHP 7.2, websites running on older underlying software break down.

So you have to ask yourself if discount hosting is really worth it. Is it worth all this trouble? Is it worth allowing your website to be vulnerable to attack, running without a backup, or slowly becoming more and more out-of-date?

If you cannot justify spending an additional $250 a year to prevent these outcomes, then you either have to readjust the way to calculate the return on investment (ROI) your website provides, or you should consider shutting down your site.

If your website is capturing leads or sales, processing donations, generating media attention that lead to sales or donations, or creating some other quantifiable benefit, then it’s simply a matter of whether or not it creates more than $240 worth of value. Given that low, low figure, the answer to this is usually a no-brainer.

But if you do have a hobby site, or you’re starting a fledgling business that really needs to pinch every penny, you’re best off using an all-in-one hosted services like Squarespace or WordPress.com. These services are usually priced pretty similarly to discount hosting but don’t have any of the disadvantages of having to maintain a server. They are essentially like premium, platform-specific hosts but are even more integrated. Their primary disadvantage is that they offer a one-size-fits-all solution that can’t be customized or can only be customized in limited, superficial ways, but that’s probably an acceptable trade-off when you’re just starting out.

TLDR: leave discount, self-managed hosting to nerds who can maintain their own machines. You need something that is maintained for you. At a $240 annual premium over the discount offerings, these services simply offer too good a value to be turned down. If your website generates any value at all for your business, it’s worth it to have the peace-of-mind and avoid the interruptions that come with leaving your website vulnerable.

In my interview with Brian Phillips of the Texas Public Policy Foundation, we covered a lot of ground, but one of the last things he shared stuck in my mind. It’s something simple and something you’ve heard before, but some simple and oft-repeated things warrant yet more repeating when they are also willfully ignored. Phillips said:

One of the things I tell my staff all the time—or at least the analysts around here—is “there’s no such thing as the general public.” We used to ask all of our analysts—they would write an op-ed or they would have an idea for an op-ed or they’d write something and I’m getting through it—and I’m like, “Who’s this for? Who is our audience here?”

“Oh, it’s for the general public.”

I’m like, “Well, if you’re writing for everyone, you’re writing for no one.”

How is Phillips solving this problem with his team? They’re leaving the office and meeting their audience.

Cody Library, San Antonio, Texas (Texas Public Policy Foundation)

Texas Public Policy Foundation has been taking their show on the road, touring the great state of Texas meeting with 100 or 200 people at a time. This gives them a platform to share their message about good public policy, but more importantly, to listen to the people they meet. Phillips highlighted the importance of this early in our discussion:

You can talk about property tax all you want or spending or budget, but to actually hear somebody talk about the issues they’re having—between the property taxes going up every year and making it difficult for them to save for the future or pay for college or even put food on the table—those are really, really important stories that we as policymakers or even researchers need to hear.

How do Phillips and his team know who their audience is? Because they’ve looked them in the face and heard their concerns.

Again, is this wisdom anything new? No, but so few people execute on this truth. You have to know your audience, but it’s easier to stay in the office and write op-eds for other policy wonks.

By listening to his audience, Phillips is also ensuring his message is being heard. It’s counterintuitive to some, but listening shows that you respect your audience. When people feel respected, they’re more open to listening. As Phillips put it:

Going into those communities, it says that your voice is important, your feedback is important, this community is important.

TPPF is giving so much deference to the audience they are trading the typical policy wonk presentation for a town-hall-style Q&A. Each presenter is limited to 10 minutes of speaking time and allowed only 3 slides as part of their presentation. 30 to 40 minutes of each event is reserved for answering questions from the audience.

That Q&A helps to expose what’s working and what’s not working in the presentations. During their post-game analysis of each event, TPPF staff talk about what could have gone better, how the message can be shifted, how they can come at an issue from a better angle to get at the heart of what matters to people.

Phillips says history working in political campaigns has helped him be frank in his feedback:

You identify stories that work and that don’t work. I’ve been doing communications and critiquing candidates and things like that for a long time. So, I don’t have any problem going up to people and saying, “Hey, that story didn’t work,” or “When you said it this way, I think people are misinterpreting what you mean. Let’s fix it.”

Along with this feedback at the ground-level, Phillips and TPPF are making use of polling to get a macro sense of the public policy debate in Texas. This helps them avoid costly errors, like pushing a message to an audience that you might assume is sympathetic to your cause, but doesn’t care about the issue as much as another audience might:

We found that on the property tax issue, the urban centers were really where that issue resonated, and then not so much in the rural areas. So, don’t spend a lot of money going and talking to middle of nowhere West Texas about property tax because they just don’t think it’s an issue. Go into urban Austin or Houston and have that issue. If we’re going to do an event out there in West Texas, don’t bring your property tax guy. Bring your education guy or bring somebody to talk about property rights or something like that because they really care about these issues.

So while polling is costly—Phillips quoted a figure of $1500 per questions for state-wide issue polls—it’s also expensive to give the wrong message to the wrong message over and over again.

Not using polling before you launch a major public policy campaign because you “can’t afford it” seems to make as much sense as not using an architect/engineer before you build a building because you “can’t afford it.” You’re saving money in the near term only to have a project fall apart in the long term. The only difference between a policy campaign and a building is that it’s painfully obvious when a building has fallen apart. Builders can’t hide behind lame excuses. Neither should you.

If cost is really a concern, Phillips suggests using Facebook as a quick way to get feedback and help your audience guide your messaging:

It’s a really cheap and dirty way to test engagement over, especially because you can change the message so quickly. If something is not working within the first two or three or four hours, and you’re not having at least your typical results on your post, then you can change it out.

Based on my conversation with Brian Phillips here are some key takeaways for public policy organizations:

1. Writing without an audience in mind is a waste of time
2. Take your message to your audience in-person wherever possible
3. Listen to your audience and adjust your messaging based on what you hear
4. Use polling or some other form of testing as a tool to help you launch a campaign and adjust as you go

Whether you hold a forum, do an AMA on Reddit, post question to people on Facebook or Twitter, or conduct a national poll, you need to be asking your audience questions that will help you hone your message. If you’re not doing this, if you’re only talking at people rather than listening to them and adjusting based on feedback, you can be as right as right can be on policy and your message will go nowhere.

Use Brian Phillips and the Texas Public Policy Foundation as your example and start engaging with your audience today.