Ask HN: Would You Ever Pay For Search? | Hacker News Reader

Ask HN: Would You Ever Pay For Search?

19 pointsRyanHolliday31 comments

My frustration recently has been the massive numbers of scraper sites polluting my SERPs. So, an idea I've had (that no doubt has been thought of before) is a search engine with editor curation & user voting and reviews to determine site rankings.

But in keeping with Jason Cohen's post, http://blog.asmartbear.com/customer-validation.html , "When ten people say they'll give you money if you build this thing, that's the only validation that counts."

So: would you ever pay $1-$2 a month for search, or is Google et al cutting it just fine for you?

(my first HN post, hoping I'm not breaking any rules I'm not aware of here)

View on HN

I'd happily pay one or two dollars a month for search that was significantly better than google. But making a better search engine than google is... highly nontrivial.

Even if you get a few million users paying a dollar a month before you stop sucking, your total revenue is barely equivalent to google's orange juice budget. They have thousands of engineers working on making search better plus a pretty good working search engine plus huge infrastructure. What do you have?

edit: Oh yes, and if you do manage to make search better in a way that google hasn't thought of, your best bet is to sell it to google.

Yeah, there's definitely a major technological hurdle. I was more curious if anyone is actually willing to pay for search. So far, it's been running on the ad-supported model.

Also, I don't have any illusions about killing Google at this point. You don't need to kill Google to make a niche search business that produces a profit, was more my thinking.

> a major technological hurdle

I don't think it's a technological hurdle at all, scraping websites is fairly standard nowadays. The challenge is a logical one. You just need to mentally think up of a better logical algorithm than Page Rank.

Page Rank is a very robust heuristic with many perhaps unwitting parallels to social psychology. How people gain status in groups and social norms are created is similar to how websites rank. Perhaps you could catch up on social psychology theory or other seemingly unrelated fields to see if any sparks fly about how you could apply it to the web.

> for search that was significantly better than google

Why aren't you using Duck Duck Go? [ http://duckduckgo.com/ ]

It's 10 times faster than Google to find stuff if you're a programmer, you can just press J, K like it was VIM and just focus your eyes on the green box to scan like 100 webpages instantly.

It's even faster if you don't use Chromium but a browser with a sane and configurable BACK button (ex. Opera ex. BACKSPACE) instead of moveyour-hand + ALT + ARROW nonsense.

Set homepage to DuckDuckGo and just press backspace with pinky to get back to entering a new search. IMO even if you want to end up at a google service like Google News it's faster, ex. "!news wikileaks"

Really? I don't feel this way at all.

As a long time user of google and someone who has given DDG a significant 'chance' I am still using google as my default search.

Part of the reason is information density; I make a search on google and just instantly get more on the screen, and probably because I am so used to google i absorb that information much more quickly. I look at DDG and my brain hurts just a tiny bit trying to parse it all.. not sure how to explain it exactly. But as someone who really is into the idea of a search engine made for power users / programmers.. ddg just wasn't doing it for me.

Re keyboard shortcuts: this is not a selling point for me, nor should it be for any power user. It is trivial to map backspace to the back button in any leading browser. I personally use 'b' for back, 'j' for down, 'k' for up.

Well it's a completely different cognitive heuristic. Duck Duck Go the heuristic is continuous rapid scanning. For Google: send, receive, digest.

I believe the brain tends to only evaluate around 7+/-2 ( http://en.wikipedia.org/wiki/The_Magical_Number_Seven,_Plus_... ) chunks at a time for each comparison anyway so I prefer the minimalism of Duck Duck Go. Which displays 7-9 results in my visual field at the default zoom level. But I agree my original claims were perhaps a bit hyperbolic and there is definitely a role for consumer preference in the search engine market.

I think perhaps we can agree Google is perhaps a bit too rigid in only offering it's single namesake search prompt and should provide further variations of its core product to cater to a variety of heuristics for visual processing of information.

No I wouldn't.

1) Because there are so many free providers out there.

2) If the free providers don't find something, I don't know what I'm missing out on, so there is no incentive to pay. If I know that I'm missing out on something I probably have enough information to know how to find it without paying anyway.

What you are doing is asking people to pay for you to help them find something when as far as they can tell they have already found it. You can't demonstrate what value you provide to the buyer without giving all that value away.

All that said, I don't think that makes it a bad idea, just not one that you can charge the user for. If you build a system where user feedback effects search rankings, and you can convince enough people to use it and actually give you feedback on sites, and if you got enough users I think there would be other ways of generating revenue (probably selling to Google as someone else suggested would be your best bet, assuming of course they didn't just build it themselves)

[Edit: Google have tried something similar to this before - http://googlesystem.blogspot.com/2008/11/google-searchwiki-l... but it seems to have disappeared now so I'm guessing it was an experiment that didn't work]

I think your pricing is off by a couple orders of magnitude if you can nail search in a particular domain. (Court cases, for example.)

Right. You could use other non google alternatives and create your own specialty search, charging for better, faster results like realtime updates for court cases. Indextank is one such engine, and beats google on being realtime and fully configurable (take into account things like user votes etc). You could set up indextank to índex court filings.

Only very niche search sites (legal, medical, marketing, ...) can possibly hope to charge people enough to be profitable. It needs to fill a need that Google does not fulfill at all, not just be a little better.

No. Not unless you manage to come up with a very different type of search engine. What you're talking about goes back somewhat to the earlier Yahoo days - more a portal. More recently, Netscape tried a similar thing, but it also didn't work so well. About.com is still doing things something like this as well.

I've been working with search technology for a while now, and while crappy search is easy, good search is hard. There will always be more people trying to game the system than you have people trying to stop them, so the battle to provide good results will never end.

I think 99% of the time that I search, I find what I'm looking for within 1 or 2 queries. I'm really pleased with google at this point in time. The SPAM/SERP pages just get visually filtered by my brain.

In short, I cannot envision ever paying for general search. Unless it was in a very specialized vertical industry, like court cases, and my current search needs weren't being met (which they currently are, in all fields I use search for).

The amount of data that's created daily is so huge for even Google to index. I am really skeptical how any search engine is able to solve the relevancy of the search results to the keywords and the integrity of the search results given that there will always be blackhat SEO methods.

Given that you are able to solve the above, how to store the terabyte of index generated daily?

Assuming you solve that, next come the user experience, how to make it easy for users to adopt your search. How can they filter the search results? How to data visualize so that users can better understand it, etc.

Now, does $1-$2 make sense?

I don't know that the $1-$2 price point is really the important thing. I have no idea if a search engine would make a profit at that price point.

There's no product as of now. I have no idea about storing the terabytes of data, or data visualization, or filtering. I'm not qualified to solve most of those problems. I'm just curious if anyone feels that this kind of search would be worth paying for.

You might be able to charge for search in a specialized sub-domain of search, particularly if you can find a niche where the quality of search results could make/cost the user money.

For generalized search though, I think it's gonna be hard to find many users who will pay. Looking at Google Zeitgeist as an indicator of what most people search for, it look to me like most searches are for fairly trivial information. Most searches don't require the best results, just a number of good results. Google and Bing both do a good job providing good results.. so what's your selling point going to be?

I would be a paying search engine subscriber, even at higher price points if the quality of the results was satisfactory and the following points were met:

(a) everything is over SSL, (b) there are no advertisements, (c) there is a strict pro-user privacy policy without keeping logs and user profiling.

I think the saying "If you are not paying for the product, you are the product" has a strong element of truth in it in most cases.

Google could easily provide that. See http://encrypted.google.com/

I don't mind paying for it if I get sensible results with ranking based on verified users reviews and opinions included. For example, if I searching for taxi service, I'd like to know more about the service based on feedback from verified users before I select it.

You just reinvented DMOZ?

http://www.dmoz.org/

DMOZ doesn't really have user voting, though. But it's definitely a reasonable comparison.

My guess is that some people might pay for some kinds of searches. But your engine would have to be really fast and/or summarise results in a way Google doesn't.

Can you give examples of the scraper sites you are talking about? What terms are you searching for?

This site. http://efreedom.com is winding me up at the moment. It's scraping from stackoverflow and Google has it ahead in the serps for a fair number of search queries I've run.

I wouldn't mind so much, but a) the site isn't as user friendly as stackoverflow and b) obviously isn't updated with answers as fast as stackoverflow is.

Obviously I should just be searching stackoverflow directly, but I'm afraid that habit of going to google first is a difficult one to break.

Doesn't Google provide a way to blacklist some domains, and not return results from those sites?

efreedom is annoying and they are not the only site with http://stackoverflow.com/ data. I don't even want to link to the other sites.

If you are using chrome, you might be interested in this extension: https://chrome.google.com/webstore/detail/gledhololmniapejef...

There might be such extensions for other browsers, I don't really know of any.

Looking at the traffic in Alexa, it looks like stackoverflow is doing just fine. The UI is very similar on the two sites, except efreedom is much more slimmed down it seems. Hopefully this site is getting people answers better than if it didn't exist. I checked and stackoverflow does a a cc-wiki data dump in xml so I doubt this site is html scraping, it's probably using the xml data.

Mahalo comes to mind (there's some useful content, but there's also a massive amount of things scraped and barely cited or not cited at all). It may be unfair to call eHow a scraper site, but it usually ranks high for the terms and is very light on actual content, and again with the barely cited or not cited. Beyond that, I run into a lot of seemingly one-off sites that are geared toward a specific keyword combination, have no real content, but somehow manage to rank.

It's easy enough to ignore the big, repeat offender sites, but not so much for the one-off sites, and avoidable or not they're still there cluttering my SERPs.

Right, so it's a fairly complicated problem. I don't think you can approach this as a business directly. You first have to just focus on solving the problem, probably for a niche. If you succeed, then you might have a chance of making money off it.

Jason Cohen's advice is probably not aimed at people addressing such fundamental problems. Imagine Page and Brin going around asking people if they'd pay for better search results. People would laugh at them.

But why would you bother charging for it, when you are likely to be able to monetize it well if you got any traction.

Ad supported is a workable model, I agree.

The question if you would pay for it is directly because of Jason Cohen's post. If it's not worth paying for at some level, it's probably not a problem that's really screaming for a solution.

I'd pay for Google, for example. The fact that I don't have to pay for it is nice, but I'd cough up every month for it if it was a premium service, because it does something I absolutely need.

I would pay 10x that for a site in every way like google except punctuation didn't get ignored.

I would easily pay a couple hundred bucks a year for something significantly better than google in anyway.

On a slightly unrelated note, how much will you pay for a service which takes your search query and then gives customized analysis of top five sites ranked by relevance and the quality of content.

Reply on news.ycombinator.com