Archive for September 2006
This week I’ve enjoyed glowing encouragement from users and mentors… But a refreshingly critical conversation Seth Sternberg of Meebo stands out as one of the more constructive chats I’ve had in a while. Here’s some background, but more importantly, I hope this blog may lead to even more tough input like Seth’s.. comment away!!!
About eight months ago, Seth left Stanford GSB to pursue Meebo full time. He was just a few months and a handful of credits shy of graduating. Many classmates (and some renowned prof’s) thought he was crazy. He quickly proved folks wrong. Soon Seth and team won over Sequoia, and in just a few months they’ve fostered the fastest growing and most passionate IM communities I’ve seen since Skype.
Seth and i have exchanged a few email threads over the past year, but mostly along the personal context of entrepreneurship while at the GSB. This week, I finally had the opportunity to show Seth what we’ve been working on.
Seth was fairly critical about the concept of social search, but agreed with our general idea behind “searching together”.. We dove into a mini-demo, and Seth dished out a healthy dose of critical advice:
- Vision simplicity: Seth patiently let me describe our search community… and then he laid into me… “that is way too complex – do you really expect an average person to get that?!”
- Feature simplicity: Seth questioned why certain stuff was on our pages.. “do you REALLY need that text there?” he quizzed… He gave example of at meebo they found that the ‘Send’ button on IM clients was used ~1% of the time (people just use ‘Enter’ on their keyboard), so they took it out. He referred to Google as another example: Despite their growing portfolio of major products, Google has shown relative restraint in not pushing their other properties by cluttering their SERP.. (though maybe still not enough restraint?)
- The Two shot chance: Seth commented that he felt every new web product has two shots at getting “blogger buzz”..after that, the product and community must to carry and spread itself.
Building a community can be like lifting up rocks on a mountainside to explore for bugs (or truffles, or whatever folks look for under rocks)… you need do some work, look hard, and treat each like it’s own discovery. Your hands will be dirty (or in this case, calloused from too much email), but you’ll be pleasantly surprised by the real person you can find behind each outreach.
With our platform in place, and (of course) a company name settled, September’s focus has been to expand our community. We’re moving beyond just GV’s to build a credible & authentic group of real users. The challenge is to do it discretely – in a secret pre-launch beta site. This way, when we do “launch”, visitors will discover an experience and a real community to search with.
So, how can we discretely sign up hundreds of people who will keep us secret, but who will also buy our vision, bring us into their private “search life”, and forgive us for any early shortfalls? Given that many early adopters tend to blog about new sites they try, this can be a scary challenge.
One approach I’ve enjoyed is to reach out, just say hello, and be authentically interested in someone’s life… the “lifting rocks” analogy applies because each outreach is a unique challenge, takes time, attention, and a more studious approach than any traditional mass-outreach.
I started with our own users… contacting certain clipclip.org users, and opening up a 1:1 dialog. Effectively the email said “would you like to try our search community? Can we trust you to keep us secret?”. The dialog wasn’t just a mass mailing, but included taking each interested user and pre-discovering the person behind that username and email address. I studied users’ clips, searched their backgrounds on our search site, and (in most cases) tried to discover my own genuine interest in involving this person in our community. I did it over and over, with every user we invited. It was a huge amount of email (thanks Stanford J) but I felt more authentic about my invite being selective, and it seems the response was encouraging…
“Your appeal to early adopter and selectiveness got me.”
“I’m honored that I can help”
“I appreciate the way you talk to people you don’t even know (read “me”) ! It goes beyond the skills of sales person”.
Many of these emails have continued on to a thoughtful exchange of feedback, collaboration around expanding community.
The big question is how well this approach can scale and can it expand it to other online communities? Last week, I tried a posting on craigslist (after all, craigslist users have been great focus group members) “thank you craigslist, users for all your support so far …we’d love to have you join our search community….”. Again, it took work, but was fruitful, and many of our newest, most generous members are craigslist users.
It’s been couple of days now since the change. Do you like it? Do you not like it?
No, I’m not talking about my new hairstyle honey, I’m talking about the search results rankings moved to the left from the right! So? Do you think it’s better or worse? For those who joined the alpha recently and for those who can’t recall how it looked before, here’s a screenshot from BEFORE.
My thoughts below.
- It’s easier to compare the engines and really shows the meta value (if you don’t know what meta means don’t worry because I didn’t either until a month and a half ago but it means it aggregates all the results from multiple search engines… I think)
- Harder to scan the result titles because of the extra columns on the left.
- High (or higher) learning curve -> OMG! I don’t know what I’m looking at.. because this is so different from Google!
High learning curve has to do also with all the up and down arrows, avatars, the grids, the frames… yes I do realize that. That is why this experiment was an especially scary one because I was very careful not to break too much of the conventions such as the colors – blue for the title, black for the summary text, green for the url.
But I am already breaking the convention big time by using a table to compare the results which none of the other meta search engines did. I wasn’t sure if it would work and was also paranoid that maybe there was a reason other meta search engines did not do it this way: they don’t want to look too different from the conventional engines.
You could argue that I’m breaking the convention anyway so might as well do it all the way, but changing the rankings from the right to the left was like jumping off a cliff for me. It was scary. Thus I haven’t changed the grey box pointing still to the right column. :p
Any comments? My husband will say “yes, honey it looks great!” without lifting his head up….
Once you log in to url.com, we remember your password until you logout.
Well, that’s the theory anyway. Several users have been raising the issue that the “remember my password” feature isn’t working for them. I think I’ve found the scenario where this may be happening:
If you’re logged in to our service from two different urls e.g.
our super duper secret url 😉 and our earlier staging site url, it’s only going to
remember your password for the most recent url.
When we mentioned our new url, the Firefox search plugin was still pointing to the old one. So, this may have been have caused the bug.
Anyway, the search plugin should now be updated to the new secret url, so hopefully this bug won’t show up again.
Hopefully, we’ll be moving to one URL (pun intended) soon. But do let us know if you’re still being logged off unexpectedly (the more details the better)
I hope I never think about a company name again. Chao and I have collaborated since way back in 1999, and we’ve never had a single contentious discussion till we butted heads on the topic of company name – and by default association – our URL. It was painful, hugely educational experience. Hopefully this post will summarize just a few of our realizations
Founders instinctually care too much about the name.
Names mean a lot. Names are synonyms for all great things a brand represents. In every job I’ve had, I became proud of that logo on my business cards…
I fooled myself to believe that the killer name will give us a head start to everything great about our brand identity… a name people would want on their business card.
Thanks to a rattle from Chao, I’ve matured to a strong viewpoint that -in a startup context- we expect way too much of our name.
Yes, the name is and always will be a sacred word to us. Sure, it will ultimately be a single-word embodiment of our brand identity, but today, it is just one little tile in the mosaic of brand identity. Here’s the kicker… early on – that mosaic is sparsely populated with just traces of ideas, some alpha code, but no history, no customers, and no passionate community. When you see a mosaic of one tile, you just see one tile – so you want that tile to be shiny, beautiful on it’s own. This is the paradox of a startup naming exercise.
When naming a startup, we need to (and I failed at this) elevate the other parts of our identity– our team, principles, values, innovations. And put blinders on this name thing. Be comfortable with the name, but don’t expect it to somehow embody the existence of the brand you aspire to become.
Where’s the value in a URL? Probably mostly in the SEO
So here’s the paradox – we will develop our brand identity over time, and grow into any name we choose – but a domain name seems to have instant ramifications. Or does it?
Clearly, the market reflects a strong opinion that domain names matter. I’ve heard the industry has surpassed $2 Billion, and transactions like wiki.com’s recent $3M sale makes this easy to believe. Nonetheless, as I combed through the bone yards of names passed over by legions of squatters, I began to question exactly what strategic value a URL really carries. I found boatloads of propaganda, but almost no compelling analysis. Maybe I didn’t search hard enough.
By strategic – I mean how does a fancy domain name really cause users to come to your site? And how does it really help them come back? (Caution – I’m nibbling at a broader study of customer acquisition and retention.. which is way beyond scope of any single post).
I now believe that – outside of traffic-driving SEO optimization – a URL is unlikely to be a significant driver of user acquisition or retention. Furthermore, the logical claims of “usability” and “memorability” may be way over-inflated and deserve closer analysis.
SEO does matter. A search-keyword optimized domain name will likely help you GET traffic. Google, etc still factor domain names in organic search keyword relevance, therefore – for the right target segments – it makes sense that you will get users merely because of your URL. It seems to be a fairly well analyzed concept and – if it fits your model – it may rationalize a hefty price. Shoes.com is a great example of this. Bike.com (recently sold for $500k) is another. However, I do wonder if the weight of URL as a factor in search relevance will dilute as search engines continue to refine algorithms with such wizardry as social search, etc. Then again – in a perfect economic context – is paying for a domain akin to bidding on adwords and thereby a fair indicator of transaction-related search relevance?
Now, on to the two other aft-discussed, but (as far as i can tell) unstudied “values” of fancy URL’s.
Claim 1- “A more usable (short, easy to type, phonetic spelling) domain name is it more will make people more likely to come back…” On face, I’m inclined to agree that a 4-letter URL is more “usable” than 5-letter URL…but will it drive retention? Do we truly not come back to a site because typing the URL is a pain? Now that search contributes a vast percentage of quasi-direct navigation, what is the true traffic lost to URL misspellings, or sheer laziness in typing long URL’s?
Claim 2- “People come and come back because the domain is memorable”. This one frustrates me on two levels. First, what REALLY is memorable anyway? Was “eBay” really more memorable than, say “svxx”? Second, even if a particular word/url is more memorable than another – does this truly make someone more inclined to return? What ever happened to users remembering the great experience they had on your site? In August, I ran a little test on our Guiding Voices group (an awesome team of 50 friends, family, and mentors who’ve become a sounding board for our search project) where I flashed all with a list of potential domain names, then at the end of a survey, asked what domains they remembered most. The results were disappointing – there was hardly any consensus amongst the group as to what people remembered most. I would LOVE to see more research on this topic. If someone forgets your exact URL – won’t they just type it into their favorite search engine anyway?
Quickly, I want to cast a quick vote for acronym-driven URLs, which treat the URL more as a phone number than a true brand. VRBO – “Vacation Rentals by Owner” is my favorite example of this. Their NAME is cumbersome, their URL is about as distinctive and memorable as a third-grader’s scribble. Yet the company has absolutely romped their nearest competitor (and a vanity URL owner) Homeaway.com.
Finally, are community names different?
Given the extreme levels of stewardship that strong communities take on their service, it could be true that names mean more for a web community. For example, the communities themselves apparently often contemplate the name of their service – I found over a dozen cases where the Yahoo Answers community has asked itself “why is this called Yahoo Answers” . At eBay, i explained Pierre’s “echo Bay story” hundreds of times. Are users proud of eBay the name? Do people think they’re talking to Craig on craigslist? Did they at first?
Sadly, again, I found little research on this topic.
So, where did we land? Perhaps you’ll crack the secret 🙂