Ask HN: Review my Craigslist killer

Perhaps a way to break into the market would be to find an esoteric genre of items to trade with a very devoted user group - these people generate buzz for free. Then, when you have a group of users, slowly expand into more mainstream categories that have some relation to the original base. For example, you could start with car stereos and speakers, then slowly expand into other car parts, then into everything. I realize this may require shipping and isn't CL-like, so it may not be what you are looking for.

I think the most important thing for me that Craigslist is missing is user profiles of some sort. When I am going to drive 30 miles to some guy's house to pick up a widget, I would feel much more comfortable if other users said he was on the level and dependable.

Feature-wise, I'm unclear on what your site offers that Craigslist doesn't, other than a more web 2.0 interface. If anything, I think a simpler, quicker interface would be better for something like this.

What makes Craigslist "just work" is the simple, bloat-free interface. It makes it easy to view large quantities of data on a single page, and doesn't make it a "chore" to browse through the site for stuff.

Your UI is really way too bloated. Cut down on all the "web 2.0" look, focus on what your customers need rather than making sure your design screams "I'm a Web 2.0 Startup!"

Cut down on the graphics, the crazy whitespace.

Just my two cents.

Marketing in small towns is going to be difficult. A lot of small towns lags behind more urban places in internet adoption. Their population are also generally older and less tech savvy than cities - so you should plan for this group (simple interfaces, etc...).

You also get small towns that are relatively near each other: if a person wants to buy something $200 + he will happily drive through to the next town. So if you implement a type of map and list nearby towns it may be good.

A lot of selling over here happens with small regional papers – a regional paper will cover one or two towns and will be filled with regional news (stuff like school performance, local politics, etc...). Those papers also contain classified advertisements. If you can maybe link up with one or two of them to put those advertisements online – or provide a system in which they can put advertisements on your system (e.g. show on their website and redirect to yours when a user clicks on the add) it would be great.

Most of those papers are usually a one or two man operation without much of a website – you should check if you can get a system in which you supply a CMS type system with your advertisement system to them. You provide the website framework and advertisement system and they fill it with content & advertisements.

I think you need to think a lot more about your urls - make them friendly for people and search engines, so instead of:

http://www.menkle.com/BrowseCategory.aspx?c=3

why not:

http://www.menkle.com/categories/electronics

and instead of:

http://www.menkle.com/ShowClassified.aspx?id=847

why not:

http://www.menkle.com/items/dvd/american_pie_2 or

http://www.menkle.com/washington/georgetown/dvd/american_pie...

some well thought out url rewriting will go a long way + will give the benefit of obscuring the fact you're using asp, which might help with security.

As other would have already commented, it seems that the simple interface is one of hte biggest drivers for CL. You would find the same if you dig through tons of forbes/media articles. I think that is not the case. That is just post-facto insight - something which gets generated a lot in the media.

The primary driver of business for CL afaik is data and the people who already use CL. If I want to sell something I go to CL because it offers me the most amount of buyers. If I want to buy something vice-versa. It is a self-fulfilling loop.

The only way to beat CL is to figure out how do you overtake CL in its primary driver of business. It will not be easy - I dont know even if it is possible. But what I do know is that anything else (neat UI, AJAX) is just inconsequential.

Here's a start: don't use javascript for link navigation. Try browsing Craigslist with javascript disabled and you'll see that the site is fully functional. Then try browsing your site the same way and see how far you get.

Craigslist quickly presents a lot of useful information in an accessable manner. It's successful in part because everyone and everything can traverse their site.

Craigslist on lynx:

front page: http://i606.photobucket.com/albums/tt145/lynxcraigslist/cl.p...

sf bay area: http://i606.photobucket.com/albums/tt145/lynxcraigslist/cl2....

software jobs: http://i606.photobucket.com/albums/tt145/lynxcraigslist/cl3....

specific job posting: http://i606.photobucket.com/albums/tt145/lynxcraigslist/cl4....

Although not particularly pleasing to look at, nothing stopped me from getting right to the information. Now consider menkle on lynx:

http://i606.photobucket.com/albums/tt145/lynxcraigslist/menk...

Unfortunately this is as far as we (and most search engines) are going to get because your links depend on javascript for navigation.

Install Noscript in firefox and try your site out. When you do, you'll see that your site is completely unusable if the person doesn't have javascript.(can't change categories)

So either remove it, or add a javascript check that shows different content when it doesn't detect javascript

It has to be clean and professional to give me some confidence.

And let me tell you, I really really hate CL.

That's the big question that none of us probably have an answer to. Like PG said about auctions: "a startup that wants to do this has to expend more effort on their strategy for cracking the monopoly than on how their auction site will work" and like he said about dating sites "if you want to do a dating startup, don't focus on the novel take on dating that you're going to offer. That's the easy half. Focus on novel ways to get around the chicken and egg problem."

Building a better CL isn't such a hard thing. The hard part is overcoming the momentum that CL has or in your case (since you want to go for people in rural areas) figuring out how to get people using it before there's a reason to use it (i.e. other people using it).

So, the chances that your site will succeed aren't wonderful. It's not that it isn't a worthy site, it's that there's a high level of random chance involved in whether a site like this succeeds. Success for a web office suite will be based on features, polish and the like. Success for your site will be (mostly) determined by whether other people use it since the majority of the value will be the user contributed content and not your code.

But, in terms of the code:

* Better search. CL only matches whole words as far as I can tell and it's just not great. Do it better. Get people to what they want more easily.

* Logins. I don't like having to create logins and until you're big don't put that hurdle in front of people. When your site has a million users, they will see the value of an account. Right now, an account is another reason not to use your site. Why should I register for a site with no one on it? However, if I can just post without registering, well, I might. Maybe offer the option of either ala CL.

- Your site does not work with Javascript turned off. If you're trying to get the small towns and such, you need to be careful on this.

- At first entrance of your site, it looks like a typical domain parked page with some ads. Make it more clear your site is about small towns and neighbourhoods. Display your top cities up front, or at the very least use some simple ip location to grab where I am and display relevant content.

- You want to make me register to contact someone. Not going to happen

- All the AJAX you are using is totally unnecessary.

- Stop calling yourself a Craigslist killer and focus on a single niche, in a few small towns and do it well. Dominate those towns and that niche, then expand.

That, along with everyone elses comments should start you off .. good luck

Its little things like that that make CL such a winner. CL tells users that it cares more about them than profiting or collecting data from them.

Also, there's no listings in central Texas.

It does look a little bit (like someone suggested) like a parked domain. I think its the overabundance of whitespace and soft pastel colors. Do something different and unique. I suppose it can even be ugly, just make it a different kind of ugly than Craigslist and domain squatter pages.

There are a lot of things people would like to do on Craigslist but can't (your question above). Many of them are unfortunately viewed as illegal or immoral by various cultural groups, municipalities and governments. Perhaps you should host your site offshore and tap into new markets, offer new features like anonymous cash, etc.? I wouldn't do it, but it'd be interesting.

* Best Catagories: I am not sure there is anything in particular you can focus on. If there are catagories that Ebay and Craigslist forbid, that are easy to handle, then you might advertise to buyers / sellers for that catagory, but the only such catagories that come to mind are firearms and used underwear. You can probably do the firearms, as long as you try to only assist in-state transactions, and you prominently warn parties that transactions are archived on the site forever. Maybe you could allow someone with a FFL to pay for a special account where they could do interstate firearms posts. Offering pets for free is another one.

* What I wish I could do on Craigslist: 1) Make a good search engine for historical pricing data, so I can see if something is a good deal, and price my own offerings. This is hard on CL because there is no way to tell if a posting ever sold, or if the transaction occurred at the advertised price. 2) Archive all posts in a searchable manner. This is good for pricing, but also for tracking down stolen shit. 3) Don't stop people from selling, trading, or giving away pets. 4) Swap meet - see below 5) Never completely delete posts. You can make them "go gray" YC-style with a few flaggings, but if the poster wants to remove it, or the transaction is done, it should go into some permanent archive. Access to the permanent archive can be sold at a subscription of $10 a year or something if you need to keep out spammers and etc.

* What I would change about your site to make it better: 1) Use a simple, no java script, craigslist like or google like interface. Look at plentyoffish if you want something a little more featured than CL. This site (yc) is also good. 2) In addition to searching by zip or region, have a "special region" called "mail order" for people buying or selling through the mail. 3) Offer a way to do an auction. Avoid the sniping crap by extending the auction for an hour every time someone makes another bid (i.e., make it a real auction)

Here's a way to try to get jumpstarted: hook up with some small flea market to do a "county wide garage sale". Charge $5 to pile your shit in a parking lot to sell it, free admission, the $5 charge is waived if you post all your stuff for sale on merkle 30 days before the swap meet.

I would love to consolidate all my meeting of potential craigslist buyers, and all the driving to check out items that turn out not to be as advertised, into one day and one location. I could tell people I'm interested in the item, and they should bring it to a certain place at a certain time, and then I could do all the haggling and get my business done in one hour and save gas.

I like your idea. The most important thing is to make the interface cleaner, don't use ajax or javascript or anything else that a web designer would think was cool.