Pleased to present Public Beta for automatic Appraisals

[estibot.com]

I released EstiBot.com for public beta!

Go ahead and have your domains appraised-

Thanks so much for your help and support:
Verbster
ToryC103
DomainSpade
NameClerk.com
Skinny
Showbiz
Dagersh
Megasonic
Dkh
Like5
Rck
Chulium
Rawandy

Cheers!
Josh

 

[Daniel]

Would it not be faster to query the site, apply some regular expression and get the number from there?

 

[estibot.com]

Maybe faster, but that would be violating their usage policies - automated searches distort search engine data and may even lead to getting banned from the search engine. That said, when the API fails, scraping is the backup method that I use, but only in this instance. Often that returns a 403 (forbidden) error, when EstiBot usage volume is high. And when both methods fail -> no backlink data.

I am currently working on a solution. After I finish the fix, EstiBot will be a boy-scout-certified scraping-free service, i.e. all data is gathered according to each service's T&C.

Automated queries that use scraping are bad, because they distort the very keyword data that we domainers use in determining the value of domain names. For instance, automated search queries on Yahoo will lead to distorted OVT data. Say, you query "dark brown titanium widgets" ten thousand times in August and next month's OVT may show a high number of searches for "dark brown titanium widgets", and someone will register darkbrowntitaniumwidgets.com because OVT says it has a lot of searches.... API queries are good, because they do not affect search engine data.

Cheers!
Josh

 

[SlimPickins]

lol...I love the "fondling and groping" bit...

Are you using IE? With Firefox and Safari the buttons are better separated, but thanks for letting me know. I'm sure I can work out something a bit more unobtrusive. Check back soon and let me know what you think.

Thanks everyone for your comments.

Thanks
Josh

hi again josh! well, lets tone it down to piggybacking this time but, yes, the buttons are still riding each other with IE... i checked firefox and, as you said, they are better spaced..... obviously only a COSMETIC concern and NOT to distract you from your more important fine-tuning thats going on but wanted to let ya know..... keep up the great work....denver

 

[estibot.com]

Yes, I admit the IE optimization has made room for other things in my daily efforts...I will look at it, though. In the meantime, order a SEDO appraisal and help me cover the costs

There have been some extra fluctuations with values recently, that's because increased traffic has lead to me exceeding the API quotas and consequently some searches are returning a zero instead of the real result...I'm working on this, too.

Josh

(this post is part of my response to the critique expressed this thread: http://www.namepros.com/364441-what-good-site-find-how-much-post2149901.html)

Let me take this opportunity to update users on recent developments.

I believe in the EstiBot algorithm very strongly and I have in my hands a prototype for v2 which I expect to be pretty damn good. The problems and fluctuations you see are not caused by a fault in the algorithm, but by the lack of reliable resources. This is what you get for making a free service - no money to invest in better resources. I hope one day I have the financial means to make EstiBot as good as I know it can be - the first truly useful automatic domain valuator. As it is right now, it's a prototype, a test version, although it's not doing a bad job as it is, given the limited resources. By popular request (yes, several people have suggested this) I have now added a "donate" button on the site so anyone can help me improve this tool and keep it free.

The appraisal values are just rough estimates and can be wrong especially as it is in testing, but you do get a number of very relevant keyword data at-a-glance. Look at the keyword data and draw your own conclusions. The dollar valuation can help especially to draw your attention to some previously undetected hidden value in some domains. This is all discussed openly on the website. There is also a guide to interpreting the dollar valuation, such as it is.

I use it myself, and regardless of how much I warn against focusing on the appraisal value, it is the first thing I always look at and then decide whether to look further into that domain. I've learned to trust it to a certain extent, and the more I use it the better I can interpret it. The EstiBot algo is so complex that I myself can only try to guess what it'll come up with - apart from the obviously valuable or worthless domains.

It is mostly intended to help valuate low end domains, and regardless of what you say, it more often than not hits the right ballpark in most domain categories. I've tested it against thousands of known sales, and sure it does make mistakes but the majority of results make sense. It is not intended to give a definite value, just a pointer to help the user in his own appraisal, as is also explained on the website.

I'm happy to say that things are moving forward. I'm very pleased to be collaborating with the indispensable NameBio and we are together looking at ways to improve the service.

Meanwhile, I welcome constructive criticism! The other kind I will just ignore at my pleasure

 

[estibot.com]

OK, today I made cool updates to the site (at least I think they're cool)

EstiBot now checks whether the searched domain is for sale at SEDO and also displays the asking price for that domain, and also domains in the other main tld's.

It's interesting to compare the EstiBot valuation to the asking prices..

Cheers!
Josh

 

[briman1970]

OK, today I made cool updates to the site (at least I think they're cool)

EstiBot now checks whether the searched domain is for sale at SEDO and also displays the asking price for that domain, and also domains in the other main tld's.

It's interesting to compare the EstiBot valuation to the asking prices..

Cheers!
Josh

I just keeps getting better Josh! Thanks for the site.

BTW, a domain of mine that was already getting a VERY low valuation has just gotten lower. You may want to check this out Josh. Even crappy LLL.nets are worth $800+!

Domain: wak.net
Keywords (Autodetected) wak
Frequency 2,550,000
Anchor Text 40,500
Title 48,000
Backlinks 2
Alexa Rank Not Ranked
Traffic (uniques) / Day N/A
PPC Ads Score 0/10
Exact Searches/mo 1040
EstiBot.com BETA Valuation : USD 350

 

[estibot.com]

Briman- I actually also noticed this problem when I saw your for-sale thread and punched in wak.net.

Somehow a bug had crawled into my LLL algorithm. This wasn't supposed to happen... I've fixed it now and it should not give less than minimal market value for LLL's. Looks like your wak.net is being valued at, strangely enough, exactly at what you are asking at sedo!! (I promise, I did not tweak wak.net individually as I never do with any domains. That's just what it's saying now according to the general lll.net algorithm!)

And thanks Briman for commenting here and giving me feedback. It always helps a lot to get feedback - really appreciate it.

Josh

 

[estibot.com]

I've found an additional benefit to the new SEDO integration: It helps me find domain bargains. Every so often, a name I input turns out to be for sale in some tld at a reasonable price. I have put in one offer already thanks to this feature, and there is always room for bargaining.

Also, this new improvement should make EstiBot even more useful to the domaining community - domainers get free exposure for their names on sale. More sales take place -> more happy domainers!

Josh

 

[simonj]

Hi Josh, once again thanks for a great tool. A couple of suggestions:

1. Can you get Estibot to check for domains not just at Sedo, but for sale at other sites as well (Bodis.com?). Or do you have an exclusive/affiliate relationship with Sedo?

2. Can you implement a discussion forum on Estibot.com for users (to cultivate an Estibot community), or at least have somewhere people can leave comments? That would be great!

keep it up,
Simon

 

[briman1970]

Briman- I actually also noticed this problem when I saw your for-sale thread and punched in wak.net.

Somehow a bug had crawled into my LLL algorithm. This wasn't supposed to happen... I've fixed it now and it should not give less than minimal market value for LLL's. Looks like your wak.net is being valued at, strangely enough, exactly at what you are asking at sedo!! (I promise, I did not tweak wak.net individually as I never do with any domains. That's just what it's saying now according to the general lll.net algorithm!)

And thanks Briman for commenting here and giving me feedback. It always helps a lot to get feedback - really appreciate it.

Josh

Rep left Josh! Thanks for taking car of that. I'm sure a lot of people will be happier when appraising their LLL.net domains at Estibot now.

BTW, I just lost a sale on Wak.net because a potential buyer got a crappy paid appraisal from GoDaddy at $281! I really wish people would ignore Leapfish and automated registrar appraisals and just go directly to Estibot.

 

[estibot.com]

1. Can you get Estibot to check for domains not just at Sedo, but for sale at other sites as well (Bodis.com?). Or do you have an exclusive/affiliate relationship with Sedo?

I don't have any exclusive relationship with anyone, but SEDO has the best tools for webmasters - only SEDO has an API interface for this purpose. If the other marketplaces provided me with some instructions on how to retrieve that data, I would include as many marketplaces as possible.

2. Can you implement a discussion forum on Estibot.com for users (to cultivate an Estibot community), or at least have somewhere people can leave comments? That would be great!

I've thought of that, but I just thought the discussion is here at NamePros. But if I do get requests (like yours) I will definitely do that.

Thanks-
Josh

 

[Brujah]

Maybe contact someone directly for a discount on API usage with KeywordDiscovery or WordTracker as part of their affiliate programs? This is such a great tool. I hope you find the right monetization so that you don't have to worry about the costs.

 

[estibot.com]

Maybe contact someone directly for a discount on API usage with KeywordDiscovery or WordTracker as part of their affiliate programs? This is such a great tool. I hope you find the right monetization so that you don't have to worry about the costs.

Thanks Brujah, I really hope so too.

Wordtracker and KeywordDiscovery are way too expensive. I have negotiated with them but they will not give me much discount even if I promise free advertising on a busy site with highly targeted traffic... Wordtracker at least is willing to give some discount, but not enough.

I'm telling you, if I could afford to access WordTracker and ditch the out-of-date Overture, you'd be getting so much better appraisals. Much, much better.

I'm willing to pay most costs out of pocket, but only to a certain extent.

We'll see. If I do find the right monetization, I will use the profits to improve the site. So much more I could do with a bit more money...

 

[briman1970]

KeywordDiscovery are way too expensive.

I'm telling you, if I could afford to access WordTracker and ditch the out-of-date Overture, you'd be getting so much better appraisals. Much, much better.

I hope you can somehow afford to get Keyword Discovery into your algorithm eventually. They are expensive because they rock so hard. Btw, how expensive are they?

 

[Brujah]

Thanks Brujah, I really hope so too.

Wordtracker and KeywordDiscovery are way too expensive. I have negotiated with them but they will not give me much discount even if I promise free advertising on a busy site with highly targeted traffic... Wordtracker at least is willing to give some discount, but not enough.

I'm telling you, if I could afford to access WordTracker and ditch the out-of-date Overture, you'd be getting so much better appraisals. Much, much better.

I'm willing to pay most costs out of pocket, but only to a certain extent.

We'll see. If I do find the right monetization, I will use the profits to improve the site. So much more I could do with a bit more money...

Consider some of my previous posts, regarding allowing some users to pay a small amount for queries, and cached results maybe at a lower cost. Maybe a more enterprise level too, using a model similiar to DropPal.com should someone want to purchase a list of results they can receive by email or check in for later.

The FREE service will probably work great for most, but some might appreciate the extras if they were available. As people use the service more and more, I really think cached results will be beneficial. I'm sure there's a fair amount of overlap as people check the same dropping domains, or domains for sale on forums, etc..

 

[accent]

Hi Josh-
Estibot has become an important part of my research. One issue is looming large, it has been discussed before, but I think it needs more attention.

The presence of a registered .net is still bloating the results of an LLLL.com. Witness:

Domain: ymkr.com
Keywords (Autodetected) ymkr
Frequency 712
Anchor Text 2,180
Title 28
Backlinks 1
Alexa Rank Not Ranked
Traffic (uniques) / Day N/A
PPC Ads Score 0/10
Related Searches/mo 0
http://EstiBot.com BETA Valuation : USD 400

The name has .net and .jp forwarding to a Japanese site, and .cn + .com.cn taken but not resolving. The name is worth very little, from the data, unless you want to hold out for the Japanese site to buy it, which would be cybersquating.

This situation has been prevalent in the LLLL.coms since you added the .net checking and often badly skews the values. I could simply buy the .net and force up the value of my .coms, a situation that becomes more likely as Estibot catches on. Yes, the fact that other TLDs are taken often shows that there is value in a name, but the effect on LLLL.coms is way out of proportion. Usually the .net is just parked.

At least give us an "ignore other TLDs" button.

Thanks. This is bugging me because I would like to do a listing of my domains with Estibot values and many of them are LLLL.coms.

 

[_FLIP_]

Nice tool. Thanks.

 

[estibot.com]

The presence of a registered .net is still bloating the results of an LLLL.com.

Accentnepal- thank you so much for this feedback. I agree with you, this part of the algo is suboptimal. Many times the fact that certain other TLD's are taken is a pretty good indicator of increased value. When I added this feature I started getting a better accuracy overall. However, again the problem with a universal algorithm is that a feature that is good for one group of domains may skew the results for another group of domains. I'm in the process of creating sub-algorithms to treat certain type of domains (groups/subgroups) differentially. I've already done this with LLL's, CCC's and also some of you may have noticed that GEO domains are now being detected and are assigned bonuses depending on type of GEO keyword. The detection is not perfect though.

However, I realize that this tld taken/not taken formula is imperfect and I will make this my first priority now that I have your feedback. I'll look into it and see if I can make it better. Remember though that the calculations are not from the top of my head - they are all based on statistical testing, and I've only implemented those functions in the algo that actually do have significance in known domain sale prices. That's why I can't go and change things from the top of my head, because while a certain change might improve the appraisal for some domains, it might at the same time lessen the accuracy for the majority of domains....I have to test each change before implementing it.

Re keyword discovery - do you think it's better than Wordtracker? For sure it's a hell of a lot more expensive.

Once the keyword research financing issue is solved I may be able to release the bulk tool. I have finished a functional prototype, but using it costs money. Even the current system costs me a small amount of money per each domain appraised - API access is not free. Each time a domain is appraised, my credit card is charged. I would probably have to charge for the bulk feature, at least until the ad revenue starts to cover it (if ever).

 

[accent]

Thanks Josh.

I notice Estibot being mentioned more often as the tool improves.

 

[estibot.com]

Hi everyone, there have been some problems lately. You may have noticed some funny results.

This is because:
1) yahoo search output has changed. I have now accomodated for that.
2) yahoo api is causing me some connection problems
3) Overture is down. This will cause inaccuracies. I am working on an alternate solution.
4) .org WhoIs is acting up. I have corrected the algorithm to account for the unreliability of .org whois

On the positive side, I've added a new feature: EstiBot will now detect if your domain has especially high commercial value, and the valuation will reflect that. EstiBot will also notify of increased commercial value on the results page.

For instance, try laptop.com or pda.com

Thanks!
Josh

 

[estibot.com]

Long time no update here.

I've updated the current algorithm to reflect some findings in my new research. I will not be updating the current version much anymore, but I'm focusing my efforts on releasing v2, which will be a complete overhaul of the system.

V2 is going to have a lot of features, and as I finish programming some of the features, I will be updating the current pages with some of the features, to give a taste of what's in store for V2.

I've now added a Bulk Overture tool and a Domain Extractor (they both extract domains from any files, text, web pages, anything, and the bulk overture will also display the overture search popularity for each domain found).

Enjoy!
Josh

 

[estibot.com]

Very pleased to announce....by popular demand:

the long-awaited bulk tool!

I'm releasing the Alpha version of the bulk appraisal tool for testing. Please provide me with feedback, help me improve it.

Knock yourselves out: Get up to 100 domains appraised in one go

http://estibot.com/batch.php

Cheers!
Josh

 

[barcode]

Very pleased to announce....by popular demand:

the long-awaited bulk tool!

I'm releasing the Alpha version of the bulk appraisal tool for testing. Please provide me with feedback, help me improve it.

Knock yourselves out: Get up to 100 domains appraised in one go

http://estibot.com/batch.php

Cheers!
Josh

Time to make this a paid service

BUG - there is a difference between the valuations reported by the bulk tool versus the single domain search!

 

[estibot.com]

BUG - there is a difference between the valuations reported by the bulk tool versus the single domain search!

Thanks for testing!

Not a bug, they are two different algorithms and the values are calculated differently.

Maybe I should explain this a bit:

The single search algorithm is based on different data than the bulk tool. This is due to the efficiency requirements of the bulk tool. The bulk tool is fast; only the essentials are figured in. Still it's pretty good, better than the single search for the keywords that it can detect, for those that it cannot, I recommend running the single search with user-defined keywords.

I may keep the single search and bulk tool algorithms separate. I think it may be useful - due to the efficiency requirements, the bulk tool has its limitations. The single search would complement the bulk tool and enable the user to manually run the domains for which he/she wants traffic data, backlink data and so forth.

It's all under development and I'm open to suggestions.

Josh

 

[sashas]

Is estibot down? The anchor/frequency fields etc. all show zero results (and this for a one word name that has 60k OVT and 445 million Google!)