| 4:49 pm on Dec 30, 2004 (gmt 0)|
Now this is a joy to use.
Got good results with anything I threw at it. Knew that Manhattan is in New York, NY; brought up the most likely result for Miami (FL), and gave "Did you mean" options for OK, AZ, TX, etc., though it did miss the Manhattan, KS possibility.
What I liked most (besides the easily scannable results), was that I didn't have to jump through hoops when entering a search phrase -- no quotes, plus or minus signs, booleans -- there's quite an intelligent parser underneath everything.
This is my flavor of local search.
And yep, you'll be able to print pizza coupons :)
| 5:00 pm on Dec 30, 2004 (gmt 0)|
One of the primary battlegrounds in local search will be won on interface design (much to tedster's pleasure, I'm sure).
It's a slick interface, that's for sure.
| 5:12 pm on Dec 30, 2004 (gmt 0)|
Playing around with it a bit more. It's smart enough to bring up a list of Caribbean and Jamaican restaurants for "jerk chicken," Chinese restaurants for "kung pao chicken," and Mexican restaurants for "chicken mole."
Not bad, though it struck out on "curried chicken."
| 5:13 pm on Dec 30, 2004 (gmt 0)|
I was impressed as well by my "dim sum" query.
| 8:01 pm on Dec 30, 2004 (gmt 0)|
Here is my search result: "No matches were found for honest attorney in or near San Francisco, CA."
Proves that honest attorneys are hard to come by! :)
| 9:00 pm on Dec 30, 2004 (gmt 0)|
Worked great for some searches, and ....uhhh..... not as well for others.
It is fast and easy though!
Also instructive, I was kind of amazed at my reaction to the results.
| 9:17 pm on Dec 30, 2004 (gmt 0)|
this is one of the best local search / SE's I have seen that lived up to what local search should be and the next growth area and results look fairly good and i think this will grow via word of mouth the way google did
| 8:39 pm on Dec 31, 2004 (gmt 0)|
I really don't think the primary challenges in local search are on the user side. The big challenge is to build a system that makes it easy for both SME's and the companies serving their PPC needs to build, launch and optimize their campaigns cost-effectively and within the bounds of each SME's budget.
Today, no such system exists, and neither Google nor Overture's systems lend themselves to the SME market. If you've ever tried to systematize launch and management of a <$500/mo ppc campaign, you'll have noticed that it's currently impossible to both automate the process entirely *and* be truly efficient.
Anyone care to share their thoughts?
| 4:30 pm on Jan 11, 2005 (gmt 0)|
Pretty neat. I actually did a typo by complete accident - and it still knew what I was looking for. I was impressed by their distance estimation for my area. Haven't seen another engine nail my specific city that well in that respect.
Some bad results in some instances though - a search for something in my zip code brought up three "sponsored listings" within a few miles of me, but the "regular results" started with stores 15 miles away and didn't even list the local ones.
I think they could also use geotargeting a little more to help try and figure out user intent.
The UI doesn't work quite correctly. It kept repeating old searches even though I was clearly typing in new ones and actually clicking on search. Aggravated me to the point that I stopped playing with it.
| 4:34 am on Jan 13, 2005 (gmt 0)|
I've had to bite my tongue on this thread, as I have strong feelings, a client relationship, and personal contacts here,
but on a seperate but related note, how about this one. [TrueLocal.com]
| 3:59 pm on Jan 13, 2005 (gmt 0)|
Seems like that needs more work.
I tried several searches.
One had a closest match that was in NY state, I'm in MN. It completely missed the local businesses.
Another had a lot of local businesses, but their websites links almost all went to the websites for their international parent companies instead of the local operations local own websites.
A third led to an initial page that included only a couple real local matches. You had to click thru to another page to get the rest of the real local matches.
Maybe I didn't understand how to work it.
| 4:12 pm on Jan 13, 2005 (gmt 0)|
Both of these local search sites appear to be going in the right direction -- easy to use and somewhat intelligent. They're both in demo or beta, so it's difficult to assess exactly how good the final products might be, but both look like they'll get the job done.
It's striking how visually alike they are in layout. It will be interesting to see how they both develop, how their usability improves, and what differences there will be to distinguish them (choice is good).
| 4:45 pm on Jan 13, 2005 (gmt 0)|
Wow. It's weird to be on the other side of the fence. I can hear Tim [webmasterworld.com] laughing at me now.
That's what you get for harassing me all for of these years, Jake...
So, this is my baby. I was hoping it wouldn't get posted this early, but so be it.
TL is coming along. IMHO, it's light years ahead of where we started, but still needs work. Obviously, mapping isn't there yet. Also, to be completely honest, our synonym matching algorithm isn't in place yet. So basic queries that should work don't quite work well yet.
Where we exceed is our data. I believe we've got the best data source out there for local search. It's unmatched in the space, and I'm very excited about what we'll be doing with it. Example: we shine in queries involving franchises. Try a city with the query "fast food" and see what I mean.
I apprecate the feedback from the both of you - ken_b's frustrations are dead on, and they're issues we're dealing with. Jim makes an excellent point about the interface. I can tell you that we're not quite where we want to be; a new interface is in development.
Our distance calculation algorithm has some issues, which are being resolved as well.
I've learned a ton about consumer facing products so far - little things, like how people use the search box, is fascinating. I can tell you that a not-insignificant amount of people get confused by the dual-search box. Most of the time they'll figure it out and feed it a second query, but that's not good for the user. We're working on issues like that.
Keep your eye on it, and throw me any feedback if you have suggestions. We're planning on a big launch at SES NYC.
| 4:48 pm on Jan 13, 2005 (gmt 0)|
I've been a beta tester for TrueLocal since about early October, so I thought I'd chime in with a few of my thoughts.
TL has come a long way since I first started testing it. The layout and usability has definitely had massive improvement since the beginning. The relevancy of the searches also gets better literally with each passing week and it also gets increasingly "smarter". Searches for off the wall phrases that would return no results in Novemeber, return them now.
It has several things about it that I like as a consumer:
It clusters franchise locations, so if I type in pizza, the first 6 results aren't different locations of the same chain. But, I can also easily view all of a chains locations if I choose to.
I like the logos being included - makes it easy to scan for major chains/logos I recognize. They recently added icons to the listings without logos that identifies the "type" of business they are - I like this because when I do a search for a generic term that brings up different business types, I can see at a glance via the icons which listings are in the category I am thinking of in my head.
The interface is easy to use, clean, the ads are obvious even to regular surfers and don't "clutter" the results.
Course, it's beta and still missing a few things I'd like to see it come out with - mainly maps (as I stated in another thread, I rate maps high on my must have list for a local engine), possibly some coupons, it's distance estimation to improve for my local area specifically (I've been reading a lot about local search though and realize this is probably one of the top challenges for local engines).
I'd also like to see it's ability to match based on generic keywords and recognize misspellings, synonyms and the like improve.
As a beta tester, I've been able to witness their improvement over the last few months and with the speed that they have been making positive changes, I expect to see them continue to do so.
| 6:16 pm on Jan 13, 2005 (gmt 0)|
It's looking great Jake - big strides have been made since the last time I checked in.
In another thread, Chicago mentioned IYP's trying to emulate search. I think TrueLocal is doing a really good job with this.
It looks like a search engine, it feels like a search engine, it delivers results like a search engine, but it ain't a search engine.
My understanding of how TL works (Jake, correct me if I'm wrong) is that they have a massive data set ordered by SIC Code / Search Categories. This directory is invisible (at least initially) to the user, and is instead accessible via the search interface.
So instead of drilling down to find the stuff you're looking for, your search takes you immediately to the granular, merchant level - then you have the option of drilling back up to the topic level (e.g. restaurants).
The obvious benefit of this is that you get the ease of user experience of search, without the risk of ever getting non local results to your queries.
It's looking great, and I'm excited to see how it progresses.
| 6:19 pm on Jan 13, 2005 (gmt 0)|
Well I can sure understand how a project getting out earlier than anticipated could affect initial impressions. Good luck with your project.
Actually, I was a bit worried when I posted. Makes me a bit nervous posting publically about what might be another members project. I'm not sure how I'd react if my own site popped up on an open board for discussion when I wasn't really ready for that yet.
| 6:40 pm on Jan 13, 2005 (gmt 0)|
>>Makes me a bit nervous posting publically about what might be another members project
Jake and I are friends. I informed him of the possibility and recieved his permission, well prior to posting. The post was about TL, not Jake, but he has been publicly acknowledging his work in other posts.
And as you can see, Jake is more than a gentleman about percieved strengths and weaknesses of TL. The url is in Jakes profile and Jake is a prolific member of WebmasterWorld and soon to be a prolific member of the local search arena and forum.
TL is doing a fantastic job, and they are a team to be reckoned with. I will leave whatever specifics regarding the project to Jake to highlight. But it is worth noting that TL is a brain child of Tim Nye. Tim, is a member here at WebmasterWorld and like Jake, is a brilliant individual who has had a tremendous success in the local arena through his data company GeoSign.
Just as we look at Interchange, it is important to highlight other pure local search engines that extend beyond G and Y. Jake is not the only one. There is Lawrence, who has been in this space for some time who is now doing tremendous things in the local search space. He has been a valuable member of this forum.
The local search community is small. The local search community on WebmasterWorld is even smaller. We tend to know eachother and in many instances cooperate.
I will continue to post about innovation in this space, TL is an example of such. And when necessary, I will gain permission, as I did in this case.
| 6:56 pm on Jan 13, 2005 (gmt 0)|
|Actually, I was a bit worried when I posted. |
Don't be! Constructive criticism from knowledgeable folks in the local search space in an extremely valuable tool we can use to make the engine better. 
Businesses in other (non-internet) sectors should be so lucky to receive the kind of feedback that is present via this board and within this industry.
|but it ain't a search engine |
Actually, it is. We do full text indexing of websites associated with local business names.
1. I think that's the most salesy thing I've ever said in my life. ;-)
| 10:06 pm on Jan 13, 2005 (gmt 0)|
Just tried a few searches, very happy with the results. One thing I couldn't get used to is putting in my location first, then what I was searching for. For some reason it seems more natural to type in "shoes Dover NH" as opposed to "Dover, NH shoes". Probably just me:)
| 10:35 pm on Jan 13, 2005 (gmt 0)|
Good point roscoe. Actually, I'm with ya. When using TL and YL, I often have to fix it because I also type it in using the same order without thinking.
| 5:01 am on Jan 14, 2005 (gmt 0)|
|Actually, I was a bit worried when I posted. Makes me a bit nervous posting publically about what might be another members project. |
Same here. I wasn't sure whose project it was, but I kind of got that inkling and wound up being very diplomatic -- writing a couple of sentences without actually saying anything.
But now that the cat is out of the bag, I can still say that LD and TL are both going in the right direction. This is local search the way it should be -- actually finding what you need nearby.
They both have some problems to overcome, much to do with the source information, over which they don't have much control. For instance, "computer store" or "computer retailer" can pop up listings for local offices of very large firms. I don't think I can drop by Intel's New York City office for a handful of IDE jumpers.
Other than that, Local Direct wins hands down for knowing where you want to search. NYC, New York City, Manhattan -- it understands them all. (Okay, True Local understands Manhattan, but doesn't give New York as an option.) And LD's single search box is definitely a plus, much more natual languare oriented, a la Jeeves.
Though again, one's a demo, the other is in beta. They've both taken the IYPs one step further, and both (at least to me) outshine the local search offerings of Google, Yahoo, etc. It will be very interesting to see how this all plays out and what other folks might bring to the table.
Whichever way it turns out, I'm going to benefit.