Starting Up by Rahul Pathak

When you’re evaluating whether or not your site is meeting its user engagement goals, it’s important to look at the right metrics. What’s right for a social network is not right for a search engine. By the way, if you’re not measuring things like visits, page views/visit, time on site, etc you should start immediately. (We use Google Analytics at Judy’s Book.)

In my notes from the Facebook Seattle Garage, I mentioned that the FB team was talking a lot about page views/user, time on site etc. These metrics make a ton of sense for them. Their product is all about user’s spending most of their online time adding to their data on Facebook, acquiring new friends, downloading new apps, etc.

If you take Couponlooker, these metrics don’t make a ton of sense. Our goal with couponlooker is to have users come to the site, find the coupon code they are looking for in 1-2 pages and then leave. If we see page views/user and time on site start to spike, it could mean that users just love searching for coupon codes, or more likely, it means that there’s something wrong with our search relevance. In Couponlooker’s case, we’re looking for 2-3 page views/user (1-2 pages of search results and a click) and we’re looking for a relatively short time on site. Our goal is to satisfy a user’s need quickly and have them return in the future. Naturally, repeat visits and direct traffic (users who typed in or bookmarked your site) are the life blood of any site.

At the end of the day, you’ve got to measure your performance to see how you’re doing and to improve. Just make sure you measure the right things. Figure out the optimal user behavior and the metrics should be easy to figure out.

I recently signed up for traffic alerts from Traffic.com. Ostensibly interesting and useful, in reality, totally worthless. They don’t put any information in the subject line of the email and as a result I have to click on each one to see it. Given that most days traffic alerts contain no new information (i.e. usual traffic volume), the alerts get annoying very quickly. While this strategy may inflate open rates in the short term, I’ve got to imagine it leads to higher than usual unsubscribe rates.

If Traffic.com put a summary of the message in the subject line with more details in the body, I would have been much more likely to continue with my subscription. This isn’t hard to do, it just requires putting the user first.

No one is going to be on your site all the time. Using features like email and phone-based messages is a great way to maintain mind-share and to drive repeat visits. For example:

• Email/Phone Alerts
• Newsletters & General updates
• Stats updates - profile views etc

Retailers have embraced this in a big way - every single one I’ve purchased from in the past 6 months sends me a newsletter with ‘exclusive savings.’ While this was great at first, when you start getting three emails a week from Target it starts to get a bit much. While it’s easy enough to turn these off, it would be great to be able to control the frequency with which I received canned information. I’d love to be able to say “I’d like a general newsletter once a month at most, but if you ever have a great deal on Mountain Hardwear Jackets, send me a text message.”

You’ve got to strike a balance between keeping the user informed and being annoying. Ideally, you’d get to the point where you were keeping in touch regularly enough to be meeting business goals but with enough of an interval that user’s would still be looking forward to hearing from you.

If you can put users in control of the type of content they get and how often they get it, I bet you could get more people to give you their email address. If users feel like they control the interaction, they’re more likely to interact with you in the first place.

My home page is a personalized Google page, and overall, I love it. There’s one instance though when it always irritates me just a little bit and that’s when I’m adding new RSS content. The initial flow is fine. You select ‘Add Content’ and then select ‘Add by URL’ at which point you are presented with this box:

Now this doesn’t look so bad at first glance, but the problem is that the leading “http://” is not selected. Why you may ask? It’s because typically, when I want to add a site to my Google homepage, I don’t know it well, so I go there, select and copy the URL to the clipboard and then come to this input form to add it. Instead of clicking “Add URL” and then “Ctrl-V” to paste, I have to select the http:// in the input box so I can paste over it. Every time. If I don’t do it, you get the following:

I get what Google is trying to do. They place the cursor at the end of the http:// so you can begin typing a URL right away. I know Google is very rigorous about their UE so maybe in their testing they found that most people like to type URLs rather than cut-and-paste them in. (I’d be surprised if this were the case, but you never know.) You also don’t want a blank input form because the presence of the ‘http://’ is a cue to the user that they need to put a URL in that input box.

There’s a small tweak to the box that makes it better and Wordpress uses it. They present you with an input box where the ‘http://’ is pre-selected.

The benefits of this approach are subtle, but important. The http:// is present and stands out. The fact that it is pre-selected means that if you copy and paste a URL directly from the browser bar, it overwrites the existing text so you don’t get the Google error case shown in the second screenshot. Also, if you type a URL, you overwrite the text but the dialog box is smart enough to know that you entered a URL and to treat it as such.

Now I agree that this is a tiny, tiny nit but the fact remains that it annoys me every time I add content to my Google home page and in contrast, I get the warm fuzzies every time I use the Wordpress input box. User experience works this way. You have to help your users win in small ways and they add up to making them happier. Joel has a great article in which he talks about this topic:

Put people in direct control of the stuff around them and they will, more or less, on average, be happier. It explains why some people like stick shifts, it explains why lethargic user interfaces make you frustrated and depressed, and it explains why people get so goddamn mad when Sony decides to install viruses on their computers just because they tried to listen to a CD.

If you’re a software designer, this is it. This is your big chance to do something meaningful to improve the world. Design software that puts the user in control and you’ll increase happiness, even if your product is the most boring accounting software imaginable. You can do this at the most microscopic levels:

• The bookkeeping software I’ve been using for the last six years makes a beep when you record a transaction.
• The Apple iPod includes a tiny internal speaker so that the thumbwheel sounds like it’s clicking when you rotate it.
• The Sonos digital music system has a handheld controller with a motion sensor built in. The instant you pick it up, the screen lights up.

It’s really hard to focus on tiny details when there are huge swaths of functionality that are missing staring you in the face, but ultimately, the little details are going to be the things that make a difference. If you can deliver enough small, positive surprises to your users (think gorgeously wrapped power cables, a click wheel that clicks, headphones with cords that match the device) they are going to love your product.

I just got back from the Facebook Developer Garage in Seattle. These are local events started by a local sponsor for people interested in Facebook apps. There were a handful of people from Facebook including the Senior Platform Manager who spoke.

Overall, a really interesting event. Some of my impressions:

• Facebook is an incredible phenomenon - 33 million users, adding 100,000+ every day.
• The platform is real - 2000 apps in 2 months. 75% of active users have at least one app installed
• The talk is “open platform” but Facebook is in charge. They throttle things like invites and the number of notifications apps can put in the mini-feed (1 per user per day.) Like any other application built on a free api - you have no rights. You get what you pay for.
• Lots of Ruby developers in attendance. Probably the single largest group which is sort of crazy in Microsoft country.
• Virtual currency and Ad Networks are interesting right now. However, money is changing hands across Facebook apps. They’re encouraging more app developers to seek to get outside ad dollars into Facebook.
• Top app is seeing $20 CPM; $5-10 is more reasonable.
• Engagement is the key metric. Pageviews/user, repeat visits, time on site. People are more focused on raw numbers right now.

I think the talk about engagement is critical. Anyone will try your app once. The key is how often they come back and actually use your product. This is something we’re very focused on at Judy’s Book. Unless people are choosing to opt in to the experience you’re creating after their initial trial, you’re nowhere.

I mentioned in an earlier post about how Kayak provided alternatives for people who clicked off their site and then came back without finding what they were looking for. This is great in my opinion but shouldn’t be limited to off-site actions. Within your site, users should never hit a dead end.

This means, for example, if they conduct a search that returns no results, you should be telling them that, but then also providing other results or content that might be of interest. If you don’t have any other content of interest, you could even provide links off site that would run their searches on other sites. By the way, another great option in the specific instance of an unresolved search is providing the option to sign up for an alert when you do have matching content.

They key thing is to enable the user to continue to make progress towards his or her goal. If you can’t deliver on the user’s immediate need but can provide suggestions that will help them do so, you are still able to create a positive impression in the user’s mind.

If there’s a feature on your site that’s not constantly improving, you need to bury it. This is really hard to do because current users of the feature don’t want you to take it away and internally, you’ve probably got a lot invested in it. Common things that come up:

• “What about people who are using it today?”
• “Let’s just leave it up there.”
• “We have more important things to worry about.”

The bottom line is that if something isn’t improving, then it’s becoming obsolete. The best thing you can do is tell users you’re sorry, it didn’t work out with that feature, you’ll be de-commissioning it in 4 weeks and then take it down. People will be unhappy but it’s far better to take your medicine early than live with a gradual decline into stagnation.

The first release of a feature is often just a foundation for the iteration that is going to be required for the feature to be successful. Just sticking up something like a link to sign up for an alert isn’t sufficient. You’ve got to merchandise it, track how it’s being used, improve the parts that aren’t working until it hums.

Adopting the mindset that you are going to need to invest in features to make sure they disappear and just work helps avoid the disappointment that can result if you just expect it to catch fire the instant you release it.

Another benefit of continuous improvement is that people who do adopt the feature early find themselves enjoying its evolution. Ideally you would use their feedback to make it better, so they also feel like you’re being responsive and that your product is alive.

Things do explode out of the gate from time to time, but it’s the exception, not the rule. Focusing on making a feature better after its initial release requires having the discipline to work on the next new, shiny piece of functionality.

What is going to make people seek out your website, not just click on a link in search result pages? That’s the key question that you need to focus on. Dave asked me this earlier this week and it’s exactly the right perspective.

In the long run, I think that the best way for a website to maximize its SEO performance is for it to maximize its value to users. This isn’t a reason to ignore SEO best practices; you can’t afford to do that. The bottom line, however, is that if you are a site users care about, search engines will figure out how to send people your way.

Making sure your pages are set up well and that navigation and page titles reflect the content is important, but you need to view SEO traffic as gravy. Now don’t get me wrong - there’s nothing wrong with liking a lot of gravy, but there’s got to be something underneath it. (And yes, I realize that’s a horrendous metaphor.)

With all that being said, Google is an important driver of traffic (duh) and you should make sure that you design a positive experience for users who find you through search. Succeeding at this creates a great virtuous cycle. You get more users, search engines view you as a higher quality site, you show up in more search results, you get even more users…you get the idea.

This sounds awesome. What’s the catch? It’s the whole “design a positive experience” part. No matter how you approach your business, you can’t get away from delivering value to users. Google has to be able to understand your site and index it, but it’s your users’ opinion of your site you need to worry about.

I was trying to get some information on using a Cingular Blackberry in India and I decided to post a question in LinkedIn’s Q&A section (login required). Within 24 hours, I got some great responses, information about flight training and a strong sense of appreciation for what LinkedIn is doing at the moment. My best answer is posted below.

If EDGE/GPRS, connectivity should not be an issue as long as you have international permissions provisioned on your account. It will definitely be charged at roaming rates (x cost per MB). Local rates may be less expensive (requiring a different sim). However, the BlackBerry only downloads 8-10kb of data intially (usually the text portion of the email). The large attachment resides on the server until you “retrieve” it. So if you stay away from the attachment download, it is a very inexpensive method to get email.

You will also want to check your carrier’s website for international coverage and roaming agreements in the particular area you are traveling to.

LinkedIn’s workflow is pretty good. They enable closing a question, replying to responders and selecting a response as the best response. This allows you to have a dialog and ask follow up questions and also to provide kudos to thoughtful answers. Email notification is used to inform you of the evolution of your Q&A session. They also found a way to monetize these pages with sponsorship banners across the top of the page. Overall, nicely done.

We had a similar feature on Judy’s Book called Expert Shopper that we launched but chose not to invest in. Sticking up a Q&A feature isn’t sufficient - you need to work to ensure that questions get answered quickly. If a user floats for longer than a day without any responses, they lose faith in the feature. On the flip side, if you can guarantee a response, you have an opportunity to create significant positive impact.

Q&A is a powerful addition to a site but it requires an active community or alternatively, a small but dedicated staff who are responsible for answering questions as they arise.