Introducing The KickBack Machine

A few months ago, I wrote a bit about Kickstarter, and how they hide (or at the very least, make it very difficult to find) failed projects.

I believe that when it comes to crowdfunding, there’s much to learn from past projects — both successes and failures. And I’m not the only one who feels this way. Early on in The Crowdfunding Bible, Scott Steinberg writes:

Whatever your approach to crowdfunding, your first order of business is to take a hard, analytical look at projects that have succeeded, as well as ventures that have failed. Your goal: To observe and learn how successful projects work, and to understand the subtle nuances and tactics that determine why some triumph while others don’t.

By hiding failed projects, Kickstarter makes it difficult for aspiring crowdfunders to take a hard, analytical look at “ventures that have failed.”

With this in mind, I built a Kickstarter research tool.

It’s called The KickBack Machine, and it’s specifically designed to help aspiring artists, creators, and entrepreneurs do their crowdfunding homework.

For example: Let’s say I’m a hip-hop artist and I want to raise $10,000 for my next album. The KickBack Machine lets me find examples of past hip-hop projects that tried to raise approximate $10,000.

Here’s a short intro video I made to explain the tool:

The site has limitations. Notably, it only knows about Kickstarter projects that ended after mid-June 2012, and you can currently only filter by category and project goal. But it’s a start, and it should be at least somewhat useful in its current form.

If you’re an artist, creator, or entrepreneur who’s considering Kickstarter as a crowdfunding service, I’d encourage you to check out The KickBack Machine. I’d love to hear your suggestions for how it could be a more useful tool.

Kickstarter: it’s a social thing

Felix Salmon, pointing to Jeanne Pi’s recent Kickstarter analysis:

[Y]our chances of raising $10,000 on Kickstarter are just 9% if you have 10 Facebook friends, rising to 20% if you have 100 friends, and 40% if you have 1,000 friends.

Clearly, your existing network matters. As Seth Godin says “Kickstarter is the last step, not the first one.”

(via Bert)

Some Kickstarter follow-up

Last week, I wrote a bit about a design decision I’d noticed on the crowdfunding site Kickstarter. Basically, Kickstarter projects that fail to meet their funding goals are:

1. Hidden from search engines by way of “noindex” robot meta tags
2. Not included in Kickstarter’s own Discover interface

The post got a fair bit of attention. It made it to the front page of Hacker News, was republished by Gizmodo, and was linked by GigaOM, TechCrunch, the NYT’s Bits Blog, and others.

It also generated a fair amount of smart discussion. For example, Ben Brooks raised the excellent point that “successfully funded” doesn’t necessarily mean successful:

[A] bigger point that Misener didn’t touch on is the projects that were funded that still failed.

This is not uncommon, but it’s rarely (if ever) talked about. It’s going to take a major failure for a change to be made. Imagine if the runaway success project for the Pebble watch fails to ever see the light of day. Kickstarter can rebound from that, but they need to have much better communication than they currently have with backers. Right now all I ever hear from Kickstarter is what new projects they think I should back.

What about the projects that are massively behind schedule? What responsibility should Kickstarter have to backers over those failures?

I heard from several different people about this, many of whom who had backed fruitless projects. Alan Taylor, for example, wrote me to say:

I paid $150 through Kickstarter to Zion Eyez last year and they never delivered their product and I have no recourse. Its vapor hardware and they keep pushing the date out when they will make these video glasses. I do not believe they are making anything but money.

A quick stroll through the comments section on the Zion Eyez Kickstarter page suggests that Alan’s not alone.

As for Kickstarter hiding projects that fail to meet funding goals, co-founder Yancey Strickler commented on my post earlier this week, explaining why Kickstarter projects are de-indexed from search engines:

This was implemented about a year into Kickstarter’s life after tons of requests from former project creators. Because Kickstarter projects index very highly in search, creators were seeing their unsuccessfully funded projects ranking extremely high — in some cases as the #1 result — for their name. That obviously sucked, so we made the decision to de-index them.

He also explained why unsuccessful projects aren’t easy to browse to through the site’s main interface:

it’s because it would be a poor user experience (there’s no action that anyone could take) and it would expose the creators of unsuccessfully funded projects to unnecessary criticism from the web (those projects would be prime for trolling).

I think it’s worth pointing out that Kickstarter displays lots of unactionable projects very prominently on its site — project for which the deadline has passed and contribution is no longer a possibility — but this is limited to successfully funded projects.

Many people (Yancey included) suggested that there’s not much to learn from projects that fall short of their funding goals. Maybe that’s true. Personally, I’m less interested in what individual Kickstarter failures can tell us, and much more interested to learn from failures in the aggregate. There are tons of interesting questions I’d love to be able to answer, like:

• Which Kickstarter categories have the best track record for success?
• What’s the percentage of successful/unsuccessful projects, or average money raised, category-by-category?
• Are there useful insights about campaign length, reward levels, or other attributes that would only show up by comparing successes and failures?

And I’m sure there are lots of other useful bits buried in the aggregate failures.

Earlier this week, I heard a radio interview that confirmed my belief that there’s value in understanding failure. It was Nora Young’s interview with Henry Petroski, the author of To Forgive Design: Understanding Failure.

It’s a book about plane crashes and crane accidents, but perhaps its larger point is appropos to online crowdfunding as well. In the interview with Nora, he talks about the engineering mindset:

They want, above all else, to succeed. And how do you succeed? Well, you succeed by anticipating how you could fail… so failure is really the starting point. You get to success by thinking about failure.

Agreed.

Kickstarter hides failure

(TL;DR: Kickstarter does not want you to see failed projects. Failed Kickstarter campaign pages include robot meta tags to keep search engines from indexing them. Plus, Kickstarter’s front page and “Discover” interface never show failed projects. Ever.)

Update @ 17:30 CET: In response to some great feedback over at Hacker News, I’ve made a few edits below to clarify a few points. First, to emphasise that failed results do show up in Kickstarter’s own search results. Second, to clarify that I don’t think there’s anything nefarious or ill-intentioned going on here. Just that Kickstarter has made an interesting design decision when it comes to how it displays (or doesn’t display) “failed” projects.

—

Only show success

Here’s a fun crowdfunding experiment: visit Kickstarter’s main page, click around, and without using the search box, try to find a project that isn’t either:

a) Successful

b) In progress

You can’t.

Spend more than a few minutes poking around, and you’ll realize that Kickstarter’s front page and Discover pages are clearly built to highlight projects that are currently seeking funding, or have already been successfully funded.

From a business perspective, this makes total sense. Kickstarter’s business model is built on taking a 5% cut of successful campaigns. Showing failures isn’t in their interest.

First, failed projects aren’t actionable. No one can back a project that’s already missed its funding goal.

Second, failed projects look bad. If you’re trying to convince the world that anyone can crowdfund anything, it doesn’t help to remind people that 56% of Kickstarter projects fail to meet their funding goal.

When I first noticed that Kickstarter’s web interface wasn’t showing me any failures, I wanted to be sure. To confirm my suspicions, I wrote a scraper (using the excellent Scrapy framework) designed to browse through Kickstarter’s Discover pages, extracting project details from every single campaign page it could find.

The result: 27,399 projects[1. Of the 27,399 projects I scraped, 23,059 were successful, and 4,340 were in progress. Given that the New York Times recently reported that there are about 50,000 projects on Kickstarter, and given Kickstarter’s widely cited 44% overall success rate, 23,059 successful projects suggests that my scraper got all the successful projects.]. Every single project my scraper could find was either successful or in progress.

This means that if you’re a human being (instead of a scraper) you could browse Kickstarter’s Discovery section for days, weeks, or months. You could look at more than 27,000 projects. And you’d never come across a failure.

Hide failed projects from Google

To be clear, Kickstarter doesn’t pull failed projects off their site.

Links to failed Kickstarter projects still work. For example, I can still link to Instaprint, a project that failed to meet its funding goal on April 29. Or, if you know the name of a failed project, you can search for it using Kickstarter’s search engine. Here’s a search for Instaprint. The project also shows up in the search results for “instagram” and “photo booth“.

But here’s the thing: search for Instaprint on Google, or Bing, or DuckDuckGo, and the Kickstarter project page is nowhere in the results.

Why?

In the header section of every single failed Kickstarter project I could find, I found this robots meta tag:

<meta name="robots" content="noindex"/>

This tag, which shows up on failed Kickstarter projects, but not on successful or in progress projects, tells search engines to ignore the page.

Kickstarter doesn’t pull failures off their site. They just make it difficult to find them through third-party search engines.

So what?

I don’t think Kickstarter is doing anything nefarious or ill-intentioned here. It makes perfect business sense for them to keep their main page and discovery mechanism free of failed projects.

I point this out because Kickstarter has made an interesting design decision.

Mostly, I find Kickstarter’s approach an interesting counterpoint to other types of online transaction platforms. eBay, for example, displays auctions that weren’t successful (you can even search for them). When an item is out of stock on Amazon, you can still search for it.

Kickstarter’s “hide failures” tactic is also interesting when compared to other crowdfunding sites. For example, Indiegogo seems to list unsuccessful projects[2. Yes, I understand that the definition of “successful” is different on Indiegogo than Kickstarter. Some projects that don’t meet their their funding goals still receive money.] alongside successful ones.

But mostly, I think this matters for entrepreneurs who are planning to use Kickstarter as a funding platform. Recently, I talked to Scott Steinberg, who wrote The Crowdfunding Bible, and he talked about why research is so important for entrepreneurs.

He told me that if you’re going to use a crowdfunding service like Kickstarter, it’s important to figure out what’s worked for others in the past, but also to figure out what hasn’t worked for others in the past.

If you hide failure, it’s hard to learn from others’ mistakes.

Comcast, use-based pricing, and Canada

In the U.S., Comcast is experimenting with “use-based pricing” (which sounds an awful lot like usage-based billing). The NYT reports:

In the test markets, 300 gigabytes will be the monthly allotment of data for a basic broadband Internet customer. From there, Comcast will try charging a premium for additional data. In one of the tests, customers will be able to buy more data access in what Comcast calls “increments” or “blocks”: for instance, $10 for 50 gigabytes.

One of the article’s writers, Brian Stelter, was interviewed on last week’s On the Media. During the interview, he did a little futurecasting:

I can imagine in the future having to choose both how fast the connection I want, as well as how much data I want to be able to use.

Whoa. Hang on! Internet plans where you pay according to both speed and volume? That sounds awf… oh, wait a second. That’s business as usual in Canada.

Bell Internet packages:

Telus: compare plans:

TekSavvy – Residential Internet: