Mirror mirror on the wall, which blog search is best of them all?

Last Friday I visited the famous South Park area in San Francisco. It’s a small park south of Market street where a number of cool Web 2.0 startups are located (Twitter’s parent, Obvious Corp, is located in a building on one end of the park). Anyway, I was talking with a number of people and I heard that Technorati’s blog search quality had seen a fairly large improvement lately and was better than Google’s blog search again. It’s hard to test blog search engines unless you know when an event started and can count up the good links and count up the spam.

There’s no way I can — alone — test out the search engine’s quality.

So, thought I’d open it up to all of you. Which engine is now better? Why? Give us some things to test out and reproduce what you’re seeing.

Here, I’ll start. It’s Videoblogging Week this week. So let’s compare both for “videobloggingweek2007.” That’s a top search on Technorati, so it should bias pretty well for Technorati, right? Let’s see!

Technorati for videobloggingweek2007. 147 results.
Google Blog Search for videobloggingweek2007. 76 results.

I didn’t see any obvious spam, did you? I think there might have been one on Google’s results, but it’s hard to tell spam blog, or splog, from actual real blogs anymore.

Also, I didn’t notice any duplicates on either service, did you? How about on the searches you’ve tried? It does look like Technorati has pulled ahead again in this race. The UI on Technorati is certainly ahead of Google’s, especially with the little chart of how many mentions a term has gotten. Search Technorati for EMI, for instance, and you can see that it really spiked today because they announced non-DRM music which got a lot of bloggers all excited.

So, try some searches of your own. Which engine is working best for you?

108 thoughts on “Mirror mirror on the wall, which blog search is best of them all?

  1. Here’s the one thing I LOVE about google blog search that Technorati doesn’t offer:

    The advanced search features offer me the opportunity to search specific time periods.

    So, for example, if I want to read blog posts from a specific day in history (9/11 or Aug. 29, 2005) I can do that easily.

    More to point: when I search blogs, I don’t always only want to know “what people are saying right now” I want to know whatever it is I’m looking for, usually the truth, not the current spin.

  2. Here’s the one thing I LOVE about google blog search that Technorati doesn’t offer:

    The advanced search features offer me the opportunity to search specific time periods.

    So, for example, if I want to read blog posts from a specific day in history (9/11 or Aug. 29, 2005) I can do that easily.

    More to point: when I search blogs, I don’t always only want to know “what people are saying right now” I want to know whatever it is I’m looking for, usually the truth, not the current spin.

  3. Looking at the logs for my blog I’m seeing very few people arrive from blog search engines – almost everyone comes from normal Google.

    Standard Goolge seems to do a very good job on quickly indexing and intergrating blog posts into their search index. I saw my blog posts about Sierra et al. appear very quickly (and high up) on Google, and that brought me a lot of traffic.

  4. Looking at the logs for my blog I’m seeing very few people arrive from blog search engines – almost everyone comes from normal Google.

    Standard Goolge seems to do a very good job on quickly indexing and intergrating blog posts into their search index. I saw my blog posts about Sierra et al. appear very quickly (and high up) on Google, and that brought me a lot of traffic.

  5. Just from my own experience, I’m not really a huge fan of either. Technorati’s index seems adequate most of the time, but for some reason it failed to index a new post for over a week, until I posted again. I think the post might have been duplicated as well.

    As for Google… I have a personal blog hosted on blogspot, and on no occasion has Google blog search ever turned up a single page for it, even when I *know* I’m using search terms that occured in the relevant posts. I have much better luck with a vanilla site search, which I’ve found to work great for searching the archives (whereas new posts have to wait for the googlebot to crawl the site again.)

    Just my personal experiences, perhaps others have had better luck.

  6. Just from my own experience, I’m not really a huge fan of either. Technorati’s index seems adequate most of the time, but for some reason it failed to index a new post for over a week, until I posted again. I think the post might have been duplicated as well.

    As for Google… I have a personal blog hosted on blogspot, and on no occasion has Google blog search ever turned up a single page for it, even when I *know* I’m using search terms that occured in the relevant posts. I have much better luck with a vanilla site search, which I’ve found to work great for searching the archives (whereas new posts have to wait for the googlebot to crawl the site again.)

    Just my personal experiences, perhaps others have had better luck.

  7. There are a lot of ways to compare blog search engines. You could look at which engine returns the most links to a post. You could compare queries to see which engine has the most spam. You could look at which engine returns the most results for some relatively rare phrase like [matt cutts embiggen]. You could look at the speed of indexing individual posts. You could ask which engine had neater features, UI, or flair. You could look at how long each engine takes to return results for a query.

    Every person is going to care about slightly different things. Me, I care less about backlinks and more about recall and speed of getting results back, for example. The great thing is that with a lot of engines competing, every engine is working hard and continues to get better. That’s better for every user.

    My personal advice would be to try several engines and see what you like. Every 2-3 months, try icerocket.com or sphere or t’rati or Google again, just to see which one works best for your search needs.

  8. There are a lot of ways to compare blog search engines. You could look at which engine returns the most links to a post. You could compare queries to see which engine has the most spam. You could look at which engine returns the most results for some relatively rare phrase like [matt cutts embiggen]. You could look at the speed of indexing individual posts. You could ask which engine had neater features, UI, or flair. You could look at how long each engine takes to return results for a query.

    Every person is going to care about slightly different things. Me, I care less about backlinks and more about recall and speed of getting results back, for example. The great thing is that with a lot of engines competing, every engine is working hard and continues to get better. That’s better for every user.

    My personal advice would be to try several engines and see what you like. Every 2-3 months, try icerocket.com or sphere or t’rati or Google again, just to see which one works best for your search needs.

  9. Hagrin – it’s ok to disagree – it’s the only way we can move forward together.

    Yes, you are right. they have a product called blog search. yes, technorati can compare themselves to it.

    Let me illustrate my point.

    I will put up a blog search tonight – we will call it “Moo Blog Search” – it would be easy for Dave to say “Technorati is bigger than Moo Blog Search” — of course no one uses Moo and so how valid is that?

    Clearly Google main search is being used over the blog search. That’s why it’s not a fair comparison.

  10. Hagrin – it’s ok to disagree – it’s the only way we can move forward together.

    Yes, you are right. they have a product called blog search. yes, technorati can compare themselves to it.

    Let me illustrate my point.

    I will put up a blog search tonight – we will call it “Moo Blog Search” – it would be easy for Dave to say “Technorati is bigger than Moo Blog Search” — of course no one uses Moo and so how valid is that?

    Clearly Google main search is being used over the blog search. That’s why it’s not a fair comparison.

  11. Excellent to see Dave Sifry so active on this thread!

    I have not used the search function in Technorati for a while just because it didn’t really do what I wanted (er, it gave results that I had to wade through). But it sounds like things have improved quite a bit – I’ll go give it another shot.

    Jason Alba
    CEO – JibberJobber.com

  12. Excellent to see Dave Sifry so active on this thread!

    I have not used the search function in Technorati for a while just because it didn’t really do what I wanted (er, it gave results that I had to wade through). But it sounds like things have improved quite a bit – I’ll go give it another shot.

    Jason Alba
    CEO – JibberJobber.com

  13. @Allen Stern -

    I read your article and while I enjoyed it, I politely and respectful disagree entirely.

    Whether or not blogs are being incorporated into the main index or excatly what users are actually using to perform blog searches is irrelevant to this discussion. Google has a tool readily available called “Blog Search” and until Google decommissions the app and just integrates and ranks blogs appropriately in their main index and only their main index, then the only fair comparison doesn’t include the main index.

    Of course, you’re 100% right in what you say – what is a blog, no one actually uses the Google Blog Search, the main index suffices for people’s needs, etc., but the main index really has no place in the comparison.

  14. @Allen Stern -

    I read your article and while I enjoyed it, I politely and respectful disagree entirely.

    Whether or not blogs are being incorporated into the main index or excatly what users are actually using to perform blog searches is irrelevant to this discussion. Google has a tool readily available called “Blog Search” and until Google decommissions the app and just integrates and ranks blogs appropriately in their main index and only their main index, then the only fair comparison doesn’t include the main index.

    Of course, you’re 100% right in what you say – what is a blog, no one actually uses the Google Blog Search, the main index suffices for people’s needs, etc., but the main index really has no place in the comparison.

  15. I have found that ContextWeb’s blog posts often do not get indexed by Technorati for a day or two unless I go into my profile on Technorati and ping. The WordPress blog is all set up to autoping. Not sure what the issue is. – It’s always possible it’s me, though.

    Google seems to take care of itself but I don’t track the pings as closely as Technorati. I still prefer Mr. Sifry’s site. I agree with you Robert about the UI.. also, just feels like more of a community there. That’s what I’m looking for.

    My two cents.

  16. I have found that ContextWeb’s blog posts often do not get indexed by Technorati for a day or two unless I go into my profile on Technorati and ping. The WordPress blog is all set up to autoping. Not sure what the issue is. – It’s always possible it’s me, though.

    Google seems to take care of itself but I don’t track the pings as closely as Technorati. I still prefer Mr. Sifry’s site. I agree with you Robert about the UI.. also, just feels like more of a community there. That’s what I’m looking for.

    My two cents.

  17. Robert – I posted an article (click my name) about why a comparison btw Technorati and Google Blog search is NOT valid. I also include another question you should be asking your visitors in your initial post here.

  18. Robert – I posted an article (click my name) about why a comparison btw Technorati and Google Blog search is NOT valid. I also include another question you should be asking your visitors in your initial post here.

  19. hmm, i think there’s still lots of room for improvement for both.

    IMHO, Technorati is still slow and it often takes a couple of days until the blog posts are up-to-date.

  20. hmm, i think there’s still lots of room for improvement for both.

    IMHO, Technorati is still slow and it often takes a couple of days until the blog posts are up-to-date.

  21. Dave
    Its my Netvibes page that I used. If you type in ZapTXT in their blog search module, you’ll see the results stack up for all default blog search engines that come with the blog search module they offer.

  22. Dave
    Its my Netvibes page that I used. If you type in ZapTXT in their blog search module, you’ll see the results stack up for all default blog search engines that come with the blog search module they offer.

  23. Technorati has gotten much better. even before the recent updates I got a lot less spam from Technorati when compared to Google.

    Monitoring a persistent search string or keyword on the Netvibes blog search engine is a very effective way to let the blog search engines go tete-a-tete. Since our users use a host of blog search engines to monitor RSS feeds via IM/SMS/Email, this allows us to constantly monitor how well each engine is doing. IceRocket often surprises me with finds that the others haven’t caught onto yet but they don’t always find all the results.

    On to your test: We had 3 mentions in the blogosphere in the last 12 hours. If you do a sweep across all blog search engines using Netvibes, you’ll see that:
    - Technorati caught 2 of 3
    - Google also caught 2 of 3 (not the same 2 as Technorati)
    - Sphere caught 2 of 3
    - IceRocket caught 1
    This could change as the indexing catches up in the next few hours but thats what it looks like right now.

  24. Technorati has gotten much better. even before the recent updates I got a lot less spam from Technorati when compared to Google.

    Monitoring a persistent search string or keyword on the Netvibes blog search engine is a very effective way to let the blog search engines go tete-a-tete. Since our users use a host of blog search engines to monitor RSS feeds via IM/SMS/Email, this allows us to constantly monitor how well each engine is doing. IceRocket often surprises me with finds that the others haven’t caught onto yet but they don’t always find all the results.

    On to your test: We had 3 mentions in the blogosphere in the last 12 hours. If you do a sweep across all blog search engines using Netvibes, you’ll see that:
    - Technorati caught 2 of 3
    - Google also caught 2 of 3 (not the same 2 as Technorati)
    - Sphere caught 2 of 3
    - IceRocket caught 1
    This could change as the indexing catches up in the next few hours but thats what it looks like right now.

  25. When looking at who is best I think one must look at two things.

    * Quality of Results (i.e. SPAM is huge problem)
    * Quantity of Results (i.e. Up to date, coverage)

    To me the quantity of results don’t mean a whole lot especially if I have thousands of results returned and a good majority of it are SPAM. So quality is more important, than quantity in my book.

    For example, I frequently (once per week) like to do blog searches to find out what is happening in real estate for a particular area (search: “real estate austin”).

    http://technorati.com/search/real+estate+austin

    For this search, on the first page the 3rd, 5th – 10th are all SPAM. There are 10 search results on the first page, so 70% of the results are SPAM. I the 1st, 2nd, and 4th result contain the words “real” , “estate”, and “austin” but have nothing to do with what I am looking for.

    The same search on Google Blogs don’t contain any Spam on the first page.
    http://blogsearch.google.com/blogsearch?hl=en&ie=UTF-8&oe=UTF-8&tab=wb&q=real+estate+austin&btnG=Search+Blogs

    The first result is a blog post about Hyde Park real estate in Austin. Highly relevant to what I am looking for.

    Some more competitive areas are ripe with SPAM. Check out the following:
    http://technorati.com/search/phentermine
    http://technorati.com/search/ringtones

    Given that a recent Microsoft research study mentioned that Blogspot contained more than 70% Spam Blogs I am not surprised by the results that I am seeing. Google seems to be doing a much better job at filtering SPAM than Technorati.

  26. When looking at who is best I think one must look at two things.

    * Quality of Results (i.e. SPAM is huge problem)
    * Quantity of Results (i.e. Up to date, coverage)

    To me the quantity of results don’t mean a whole lot especially if I have thousands of results returned and a good majority of it are SPAM. So quality is more important, than quantity in my book.

    For example, I frequently (once per week) like to do blog searches to find out what is happening in real estate for a particular area (search: “real estate austin”).

    http://technorati.com/search/real+estate+austin

    For this search, on the first page the 3rd, 5th – 10th are all SPAM. There are 10 search results on the first page, so 70% of the results are SPAM. I the 1st, 2nd, and 4th result contain the words “real” , “estate”, and “austin” but have nothing to do with what I am looking for.

    The same search on Google Blogs don’t contain any Spam on the first page.
    http://blogsearch.google.com/blogsearch?hl=en&ie=UTF-8&oe=UTF-8&tab=wb&q=real+estate+austin&btnG=Search+Blogs

    The first result is a blog post about Hyde Park real estate in Austin. Highly relevant to what I am looking for.

    Some more competitive areas are ripe with SPAM. Check out the following:
    http://technorati.com/search/phentermine
    http://technorati.com/search/ringtones

    Given that a recent Microsoft research study mentioned that Blogspot contained more than 70% Spam Blogs I am not surprised by the results that I am seeing. Google seems to be doing a much better job at filtering SPAM than Technorati.

  27. Pingback: Blog Titans

Comments are closed.