Leaky (YC S11) is Hipmunk for Car Insurance
techcrunch.com
techcrunch.com
1. Hide the About Us page, or make it less specific (instead of "four guys", call it a "team of entrepreneurs". A grainy cell phone picture of the team playing video games in lawn chairs with a messy coffee table doesn't lend any more confidence to the average consumer.
2. Get rid of the "design credit goes to Square until we can hire a designer". The average consumer will wonder why you haven't hired a designer yet and will be turned off by whatever answer you give them.
3. This one is more for the nerds than the average consumer (though the average person might spot that something is wrong even if they can't pinpoint it): that iPad on the homepage is badly photoshopped (it's extra wide and the Safari toolbar doesn't extend the width of the screen).
4. The word leaky has a negative connotation. Your product offers a positive outcome (saving money on car insurance), so you need to come up with a more positive name for your business.
Branding is so important, especially in the early stages of a company. A bad branding can kill even the best company.
Despite that fact, there are nearly no articles about branding for startups.
1) You should state up front what materials I should gather before I begin the process
2) The back button doesn't work during the data-entry process. I fucked up putting in one of my cars data and I couldn't go back
3) The name "leaky" and requiring PII?
4) If you require so much PII, you need some really good disclosure as to your data retention, opt-out, etc during the process.
We went through the AngelPad incubator during the Winter 2011 class. TechCrunch covered us in a little blurb at the time. Since then we've received VC funding and have built out our team a little bit.
There's no publicly available pricing API for insurance carriers yet, so I'm curious as to how Leaky comes up with their numbers.
Get the user started on the conversion funnel while you explain to them what the service does - I'm pretty certain (based on my lead generation experience) that this will end in a far superior conversion rate for you.
edit: To find this page I went to the bottom of the home page, tried to click on the Pennsylvania state outline, clicked the Pennsylvania text link, looked for my car on the most stolen cars lists, found "Car insurance quotes get started" at the bottom http://coverhound.com/step1
What appears to be their full featured future product looks EXTREMELY interesting. It appears that coverhound will almost work like an automated insurance agent, for example, give them your info once, and they'll check rates for you on a regular and recurring basis. THAT product seems extremely valuable to me and would be doing the world a great favor.
The change leaky (or coverhound) has of succeeding, is if they do what mint did. provide a way to work with insurance companies, even if there's no 'publicly available pricing API'. There will be the magic that is missing in this market.
but, of course, once you have your service mandated by the gov (as is here in cali) you will make it harder and harder for the public to be informed about prices of your service. That's the gotcha with car insurance. no company benefits being transparent about rates.
1)Even though I know you guys are from YC (and not for example the Russian mob). I still bailed when you asked for my driver's license # and SSN. It's not enough for you to just say "trust us".
2) I checked it out from the couch quickly on my iPhone. The experience was not that good. The form drop-downs were slow loading and left me staring at empty pick lists. I had to scroll side to side to read the label on each field.
3) I was prompted twice to enter my email as an early adopter (once for each car I was entering presumably). This did not instill me with confidence about supplying personal info.
Hope that's info you guys can use. Good luck!
EDIT: in fact, for a subscription based service, you could figure out the optimal pricing quite easily. If the annual price of your subscription were less than the typical insurance savings by having it monitored, you'd have a hell of a business model.
If so, you need to encrypt it first. Storing sensitive data in the user's session is akin to writing it to disk in plaintext.
http://www.leaky.com/ vs. https://squareup.com/
EDIT: I realize they say "until we can hire a designer of our own." — but don't they have the Start Fund + YC + SV Angel money?
With their permission?
It's not like a new car company starts with someone else's body shape until they can design their own car or copies their TV ad for a few months. I can appreciate that Square might've OKed this or look kindly on a temporary copy from a startup, but for about the same level of effort they could've differentiated their site a fair bit more (even just a different hue for the background).
I'm not sure if I'm missing something here - looking at code/designs and tweaking them to learn is one thing, but is basing a funded company on something so blatant now considered OK? I'd like to know if this is the thought pattern for people who are designers or programmers.
Is Square within the YC stable, is that it?
Should I be doing this with my side projects, to piggyback on that success?
(Keep in mind that I'm assuming in this case that they don't have permission.)
"Why would they need permission?"
That's what I've been replying to.
The Tesla Roadster, IIRC, uses a Lotus body. But they actually licensed it from Lotus.
From a decision about magazine covers: "None of the individual elements [...] ordinary lines, typefaces, and colors — qualifies for copyright protection. But the distinctive arrangement and layout of those elements is entitled to protection as a graphic work."
http://scholar.google.com/scholar_case?case=3745368693852483...
The "X for Y" only works if X is sufficiently common. Surely better to define in simple terms what the site actually does.
- The homepage is hit or miss, the pictures really conveys what youare about but there are a lot of words clouding the message
- I thought the interface was very responsive and intuitive through the initial steps
- I didn't like how the 'options for coverage' was buried below the fold, everything looks cheap when they offer $50k/100K liability, you really have to compare apples to apples on that page if you want to tell me how much I am saving
- I only was shown a quote from 1 insurance company, that's no big deal for me right now but obviously that limits the usefulness, I got a prompt saying 'we're not available in your area' and took my email so at least you won't lose customers who see what I'm seeing right now
- [BUG] The 6 month premium shown in the left hand column with all the insurance companies is different than the 6 month premium shown in the 'buy now' box.
Name
DOB
Address
Make, model, year of car
Drivers License
Optional Social Security
Sex
Marital Status
Student/military affiliation
Optional Employer
Optional Professional/Alumni AssociationsThere does seem to be a bit of chicken and egg barrier though. I may be willing to give out this information to allstate.com or statefarm.com or progressive.com but I am not sure I or others will be as willing to give it out to leaky.com.
Maybe they are awesome, and maybe it's just me, but if you want me to give this information out on the spur of the moment, I guess I would like a site that somehow exudes a bit more trust.
That they are YC backed encouraged me to click and try it out. Otherwise, I would probably not have clicked at all.
(But YC backing isn't really enough for me to eagerly give out all that information, not until pg and yc is providing ebay like yc-clickers-insurance on their portfolio. :) )
Regarding the average user, I suspect that insurance companies have names like Prudential, All State, State Farm, and symbols like the Rock of Gibraltar, big old granite columned buildings, and stuffy images for a reason.
And um, leaky isn't doing it for me.
ymmv
Otoh, how IS leaky different from esurance? I don't understand that from their website either.
Oh wait, DOB and SS# and driver's license?
Um, NO - there is no need for this kind of info to get a general quote - I've had it with big insurance harvesting personal info. There really ought to be a law.
Car insurance quote should be three questions:
1. What zip code do you live in
2. What year were you born
3. What do you drive
(and obviously if you've had accidents recently)Insurance tables shouldn't work like they do. What they are trying to figure out from all your personal info is if you own or rent, where you work, how many times you have moved recently, etc. That's a bunch of bull.
A smart thing to do here might be to figure out a way to do progressive quotes/queries; take the DOB+zip+car information just to start, then let the user decide whether the variance in the results (which will be the best-case ones) merit coughing up more info.
A "soft" quote based on general information.
A "hard" quote based on very specific information.
They'd never convince me they cannot do this, especially with all the extremely misleading commercials on television.
A soft quote would be just fine for me and probably many others.
I very much like this concept though, its much like Confused.com in the UK as the techcrunch article pointed out. The challenge is getting enough carriers to actively participate so there's a high quality customer experience. I would guess Leaky is screen scraping the carrier sites. A bit easier in CA since you can't use credit scoring to price auto insurance here.
I spent not less than 15 minutes filling out all the information on your site and the site back with NOTHING. The quotation page kept spinning and spinning.
My experience with regular insurance company websites has been a lot better and that's why they have my money.
I gave my(and my spouse's) personal info and time because this is a YC venture. Disappointing.
The design is pleasing on the eyes, thanks square!
I'd also really appreciate it if you added "why do we ask this" flyouts next to each field. For instance, when you ask about home ownership, you don't actually provide homeowner's insurance, so you should make it clear why: "Some car insurance companies give you a discount if you own your own home."
For some reason, the text "(including your spouse)" seemed like an odd bit of reminder text. Do people forget this often? In any case, the flyout reminders could help here too, since you could add more detail: "This includes anyone who will drive any of your vehicles---you, your spouse, your kids, etc"
What a bizzare name choice. Looks like a good product though, but isn't this sort of thing fairly old? Or is it just new for car insurance? These sort of price comparison websites have been the staple of UK Prime Time TV advertising for the last 3 years.
There's no sort of progress meter, so I don't know how long I have to wait, and after the N/A quotes there's still a spinner running, so I'm unsure as to whether I should wait, or what.
Idea: great. Execution: really sub-par.
These sort of sites aren't a perfect solution by any means, but a large % of people in the UK now use them to buy car insurance, and so presumably Leaky is on the road (see what I did there...) to being a household name by doing the same thing in the US.
Of course in the States you guys have that old bloke and his gecko to compete with... and he has pretty deep pockets ;)
It's the only reason my mind can fathom.
When I worked briefly for an UK-based insurance company, I realized they had a strange policy: non-fault claims raise your premium more than at-fault claims. Why? Because statistically, they've found that people who get into a non-fault claim typically end up in a serious at-fault claim within the next two years. Creepy huh?
If there is no API available for the US carriers, I'm very curious as to how internal information like this is being collected.
If you find an insurance that you are happy with, you'll likely stick with it for a couple years at least, and there is no reason to visit this website during these years.
Given the above, I don't see how they can ever make a ton of money.
If x% is 1%, and y% is 1%, that leads to about $10 million per year. Not too bad, if they can get those sorts of numbers for x and y.
One question would be, how long will people use them for, once they find a good insurance company? I've stuck with one insurance company for the last 10 years. I suspect that x will start low, and then grow as the company becomes known, and then shrink as people stop using it.
I'll be eagerly waiting for this service to make it to Florida. Good luck guys.
Do you have plans already for filling the top of the funnel here? Its going to be really hard to compete with established businesses buying traffic in this market.
Is there any chance you guys are going to break into the commercial insurance space? I'd love to talk to you!
I guess if you have never heard of orbitz you most likely are not in the market for american car insurance.
On a side note everyone talks about how global the internet is but I wonder how many "internet based services" outside of media portals are actually global...
I'd think a site that makes 40 million a year off lead generation would be able to offer anything like this as well.
If you're smart, you'll visit every six months and see whether someone else is cheaper. At the very least you should visit every time you get a new car.
If they're smart they'll send you a reminder email every six months to get you to try again every time your insurance is up for renewal. Heck, they could recalculate everything for you (based on the new set of assumptions) and email you a set of quotes -- that's one semi-spammy email I wouldn't mind getting at all.
However, I can't get the site to work. Maybe it's getting hammered, maybe it doesn't work yet, maybe there's something iffy about my input (I did put in a fake licence number because I was too lazy to find my real one). All I'm getting is a bunch of "N/A"s. Never mind, I renewed my car insurance last month anyway, so they've got four months to fix it.
Eventually that would just drive down the affiliate fees. The reason that they're (presumably) high is that they've factored in the lifetime value of a customer. If you systematically lower that, then the referral fees will follow.
Still, lifetime value, really? I admit that I've been letting my car insurance roll over from year to year because I'm too lazy to go looking for new quotes (though this site, once it's working, will significantly lower that barrier) but next time I get a new car I'll certainly shop around to see who'll give me a good rate.
The trick is convincing people that what you're providing is different from the standard lead gen sites. You've done a good job in the description, but people will always be wary. Drop me a line if you want to talk shop
Question: anyone know if someone has a similar model in the health insurance space? It's sorely needed.