Mahalo, The Little Human-Powered Engine That Might
You can leave a response, or trackback from your own site.
In Search Engine news, entrepreneur and SEO pariah Jason Calacanis has announced the alpha release of the “world’s first human-powered search engine, Mahalo.
The concept of the engine is that paid Mahalo.com editors will manually review and include sites that are relevant to the top 10,000 search terms that people use to find websites, using the Google index as the basis point for retrieval of said sites and “curating the Google index”, as Calacanis puts it. The SEO community in general has been dismissive of the idea, writing it off as just another attempt by Calacanis to gather publicity and possibly Google Adsense revenue. The following links outline some of the more common criticisms:
- Battelle Media post on Mahalo (not the post, but the comments)
- TechCrunch Announcement (again, not the post, but the comments)
- Steven Bradley’s Mahalo post (Steven seems to be the only person who took the time to form any kind of intelligent commentary on the matter, so I recommend you read his post, even though he disagrees with my take on the situation.)
As you can see from the posts/comments listed above, the majority of people do not support the concept and are prepared to summarily dismiss it outright. However, I’m not prepared to do that just yet. There is simply too much potential for this idea to take off for me to do so. Here are some of the more positive aspects, based on what I’ve seen so far:
Positive Aspects of the Mahalo Concept
Non-Mahalo Search Results are Provided by Google
The counterargument could be made that people could just use Google if they want to see Google results, but most SERPs are accompanied by advertisements across the top and right side of the page. In either case, the search engine results don’t appear immediately below the search query itself. However, in Mahalo’s case, the listings above the search engine results are presumably placed there by staff members with no personal bias.
This probably isn’t a good long-term solution, but as a short-term hook to attract an initial userbase it might well prove to be effective.
SEOs Don’t Think It’s Worth Spamming
A direct quote from Andy Hagans’ blog post on the Mahalo subject, including Hagans’ own formatting:
My question is: why bother spamming it, when it won’t gain enough users to send any real traffic?
Most people miss the point completely and agree with Andy, not realizing that there is a simple means to spam Mahalo: spam Google and be successful at it. If the default results are presented by Google, then some SEOs may consider this a worthwhile venture, since the same results will appear on Google SERPs.
IMPORTANT NOTE: I do not condone, nor practice, nor endorse the idea of spamming Google or any other engine. In fact, I am strongly against the spamming of search engines in general since it detracts from the user experience. I am simply pointing out that there is a means to gain potential traffic from Mahalo with no additional effort required on the part of the spammer. (More on this below…and let’s see if Andy’s as good as his word about the link).
Jason Calacanis’ Awareness of the SEO Community
Jason Calacanis is aware of the SEO community mindset, trends, and opinions; this will allow him and his team to potentially come up with ideas and concepts to counter the sillier SEO tricks (e.g. link exchanges) and thereby present results that would be more relevant to users.
For example, the idea of human-edited Mahalo results allows for a presumably objective evaluation of each site and the determination as to its worthiness of inclusion.
The Name “Mahalo”
Mahalo (Hawaiian for “thank you”) is a very pleasant word, both in tone and in context.
The Mahalo Site Layout
The Mahalo site layout is organized, yet inviting. The colours are calming, and the shapes used throughout the site give it a certain flow.
Mahalo is in Alpha
Mahalo is clearly in alpha mode, which means that there will be issues. Most people would not see this as a positive, but the implication here is that the Mahalo team is clearly aware that there will be issues and are looking for people to find those issues and fix them. The open disclosure of such information will ultimately serve them in good stead, although it doesn’t appear to have done so at this stage.
Jason Calacanis = The SEO’s Version of Vince McMahon
Almost every SEO hates Jason and rips into him at every opportunity, and inevitably falls into the trap Jason has brilliantly set. “You can love me, you can hate me, but just make sure you spell my name right.”
If Calacanis was destined to fail immediately, this post wouldn’t exist and the comments from the SEO community wouldn’t either. It’s Vince McMahon Marketing at its very finest.
Major Financial Backing
Investors in the Mahalo project include CBS, Dallas Mavericks owner Mark Cuban, and Ted Leonsis of AOL fame.
Precedent has been established in the past for large corporate investment in failed projects that had no revenue/income plan (e.g. the “dotbomb” era of the late 1990s-2000). However, this is something that someone like Jason Calacanis will be acutely aware of, and he’s not likely to come up with an idea that has no long-term financial goal.
What Mahalo Needs to Improve Upon
Relevancy and the Apparent Use of the “OR” Operator
It appears from an examination of Mahalo results that the Mahalo results make use of the “OR” operator as opposed to the “AND” operator. Instead of presenting human-powered results for Search AND Friendly AND Engine AND Layouts, for example, they present human-powered results for Search OR Friendly OR Engine OR Layouts; as a result, a page about Thomas the Tank Engine appears for all searches containing the word “Engine”.
CREDIT: Bill Hartzer has independently made the same discovery and posted as such on the TechCrunch blog post listed above.
Get Out of “MSN Circa 2002″ Mode
I suspect this is the next step for Calacanis, but it is a step that needs to be mentioned nonetheless. As it stands right now, the Mahalo concept is similar to the concept that the MSN search engine used in 2002; directory listings, followed by search engine results from another party. Mahalo differs slightly in that the human-powered directory results are their own, whereas MSN’s directory results were powered by LookSmart.
At some point, Mahalo will need to create its own spider and present its own results with its own algorithm; however, this can wait until Mahalo has gained sufficient popularity.
By detaching itself from Google, Mahalo will create another engine and, in the short term, decrease the likelihood of spam; until it reaches the status of The Big 3 (Google, MSN, and Yahoo!), SEOs will be less likely to try and spam Mahalo for ranking purposes.
Remove “Today’s Top 20 Searches” for the Time Being
The top searches on the Mahalo website pertain to the Mahalo engine itself and to Jason Calacanis, likely due to the publicity within the SEo community that the engine has received. Pull the feature off for now, and wait until the buzz dies down.
Remove the Link Submission Option
As the Mahalo Publishers’ Guide indicates, it is possible for site publishers to submit their links for inclusion in Mahalo search results. The potential for “DMOZ Submission Syndrome” exists with such a concept; publishers submit their sites and then fill web design discussion boards asking why their sites aren’t included.
Allow editors to add sites as they see fit, if they can be trusted, and let publishers worry about building quality sites.
Get Popular Quickly, then Introduce the Revenue Stream(s)
Mahalo’s largest obstacle will be getting the word out to the general public and letting them know why Mahalo provides a quality unique alternative. However, Calacanis has already shown that he can at least market to one segment of the public (the SEO community) with a site that clearly isn’t even finished and may well have pulled off a brilliant tactical move in the process; SEOs dismissing an incomplete idea will pay less attention to it when it has been completed, if any, and this gives Calacanis a certain comfort zone to grow the engine before spam become a potential issue.
I don’t suspect that mass marketing will be especially difficult, especially when one considers CBS and Ted Leonsis are two of the major investors; a few radio ads, some TV spots, maybe mix in some ads on the CBS network of websites, use the AOL subscriber base as a marketing list, and Mahalo has a potentially very powerful marketing campaign ready to go.
It is within the realm of possibility that Mahalo will be successful. It clearly has some work to do, but it appears that Calacanis and his team are well aware of it. I wouldn’t be surprised to see SEOs talking about Mahalo in a different light within 18-24 months. I’m not convinced as of yet that Mahalo will reach the level of Google/MSN/Yahoo!, but stranger things have happened; does anyone else here remember when Google was a little dog in the yard?
June 2nd, 2007 at 11:45 am
Interesting take Adam. I know you’re not the Jason detractor the rest of us are and that helps to add balance to all that’s being said. I’ll be the first to admit I don’t like Jason. I actually agree with a lot of what he’s said about seo, though I do disagree with some key points. My dislike is more the methods he employs to spread his message. It comes across as very hypocritical to me. But that dislike does create a bias in me about Mahalo and I would prefer to see it fail. I know in my review and in everything I’ll say that I am looking to find things wrong with the site.
The main thing for me as to not liking the site is I just don’t see any hand-crafted results. It could be coincidence, but every search I performed pointed me to very obvious sites like Wikipedia and IMDB. The kind of sites many would go to first.
It seemed to me that using Mahalo would only lead to the most shallow information online. In fairness to them they might be looking to improve the links and are just working now to get pages up as fast as possible. Something tells me they aren’t, but there’s no way to know at this point.
The majority of searches performed will always lead to a non-page with the Google results and ads. That strikes me as very MFA. If I were to create a new site with a search box that pulled Google results and ads and nothing else everyone would call it an MFA site. I don’tsee why Mahalo is any different just because they dress it up a little more.
Of course 6 months from now there may be another means of monetization on the site.
The site is obviously in alpha at the moment and could very well be improved in a few months or a year. For now I can’t see any reason to use the site and something tells me I won’t find any reason to use it in six months or a year, but the only way to know that is to see what it’s like in six months or a year.
June 2nd, 2007 at 5:35 pm
That’s precisely what I’m ultimately driving at, Steven. There’s no way to tell exactly what could happen, although the signs I see in my own eyes lead to a certain cautious optimism.
Mind you, I wouldn’t be surprised to see a crash and burn on this, either, but my hunch is that the point Mahalo is at right now is a stopgap point, not a final destination. Let’s judge it after it gets to the point where it thinks it wants to be.