Ask YC: How do we announce the alpha launch?

http://onista.wordpress.com/2008/08/03/how-do-we-announce-the-alpha-launch/

by prashantdesale • 18 years ago

27 26 18 years ago

onista.wordpress.com

swombat2• 18 years ago

How about starting by getting some target users using it and seeing whether they spread the word and recommend it to their friends?

The rest is all very nice, but a bit OTT for an alpha launch. Not putting down your work at all, but anything that hasn't got real users yet is probably still a long way from being worthy of being officially "launched".

If you media-blitz and overhype it and it's not ready, you'll end up pulling a Cuil and losing most of your target audience on day 1.

Once you have a few hundred normal users who are telling you your product kicks ass, then think about marketing. Until then, focus on getting your product right.

wheels• 18 years ago

I sometimes read things like this and wonder if I should honestly take them seriously or not.

Cuil did great on the PR front. Sure, their product wasn't all that wonderful, but getting that much attention even if lackluster is something most startups can only dream of.

We're in a similar situation to the poster, but since we're B2B our "testing" basically means working with two customers. We're planning on launching with as much hooplah as we can raise when our demo site goes live in the near future.

There's a lot of power in a launch; it's your chance to get it right and get some attention. At no other point early in the startup will you have as much momentum to getting PR attention. If you can pull it off reasonably it's a great chance to catapult your startup a good ways down the road.

swombat2• 18 years ago

I totally agree. There is a lot of power in the launch. That's why you should only use it once you have something that can actually impress then tens of thousands of people who'll come and have a look.

You say Cuil did a great job on the marketing. Yes, they did. However, most people who went and tried Cuil had a bad experience. That compares very unfavourably with Google, which is unflinchingly brilliant. Result? Sure, Cuil did great by getting (say) a million people to try them on the first day. But of that million, maybe a thousand will actually use them more than once.

It's not really worth going through all that PR work if your stick rate is only 0.1%. Unless, of course, the whole thing was about investors / buyers, not about users (which I believe is the case).

prashantdesale10• 18 years ago

Thanks, The plan (and more) that I am outlining on blog is something we are planning to follow in stages. not in one day. So it will be as you are suggesting.

I have had it testing from some eBay sellers already, but certainly having it testing from couple of hundred users will help a lot.

nostrademons2• 18 years ago

I'd do the early phases like this:

1.) Release to family members, trusted friends & people that have a stake in the project. Get their feedback.

2.) Release on news.YC. Get their feedback.

3.) Find some influential bloggers in your field. Don't release to them; instead, send them a personal e-mail asking for their feedback.

4.) Go to forums in your field, pick out some dedicated early users that are really into the space. Send them personal e-mails and get their feedback.

5.) Release to the blogs and forums you identified in 3.) and 4.) Pay close attention to feedback and iterate.

6.) Release on Reddit/Digg/the rest of your list.

I've been a part of a startup that tried the PR blitz thing and it failed miserably, because we didn't serve any useful purpose. Gotta make it basically good before people will pay attention.

There've also been several times when I've released on news.YC to a very ho-hum reception and then killed the project, or otherwise took it in very different directions. That's probably not the response you want to hear, but it's much better to kill a product without a market early than to keep improving it speculatively.

prashantdesale10• 18 years ago

Sounds like a good plan and yes this is what we shall do. Iterating is the best approach as we get some time to fix the issues before more users know about it.

Any idea on how to find influential bloggers? (Besides Googling) Like is there any blog searching place that shows me ranks of blogs (based on traffic) and and have those distributed in categories like (Technology, Business, E-commerce, blah...)

nostrademons2• 18 years ago

I'm not really one to ask, since it's not like I've had great luck finding influential bloggers, and I usually leave the marketing to others anyway. I'd check out your competitors, see who's commenting or writing about them, and then see if they have blogs or homepages though.

pibefision• 18 years ago

First, fix your subscription news form.

"There has been an error submitting your email address. Please contact us at contact@onista.com."

prashantdesale10• 18 years ago

Thanks for pointing it out. Fixed. Moved to new server and missed recreating this table. This table was not part of the other schema.

matt1• 18 years ago

Hey, I Googled 'Onista' and found your recent Vimeo video: http://www.vimeo.com/1376104 as well as some introductions in various forums. Looks like you're building up the hype well.

My 2c after watching the video: I don't understand the instant negotiation. Is there a limit on the number of times I can submit a bid? If not, can't I just start really low and then increment by $1 or $2 until I meet the minimum? If there is a limit, what's stopping me from setting up multiple accounts so that I can discover the minimum, then bid that amount on my legit account? Since this seems to be a major differentiation point, I hope you've ensured there is no easy way to cheat the system ;)

Also, how do people pay each other? If you're relying on people to exchange money privately, how are you going to handle disputes?

One more thing about the video: your accent. I'd try to get someone without an accent, probably a woman. I think this would appeal more to your target audience.

How many people do you have working on this besides you?

As far as the Onista site goes, you've done a fantastic job with it. There seems to do and explore and the interface is intuitive and clean.

Wish you the best of luck in the coming weeks!

Matt

prashantdesale10• 18 years ago

Hey Matt,

Yeah, that video will be linked from our index page to show demo. Regarding accent, yes that is one thing we need to fix. I and my partners all are Indians (moved to US almost years back) and hence have heavy accent. I am trying to find somebody who does not have accent and willing to help us do the narration.

About negotiation process, buyer can attempt as many times as he wants to negotiate the price, but if it is accepted then it will be locked for that particular product for that particular buyer. We have written pretty smart algorithm for negotiation and it will be impossible for buyer to know the minimum because it keeps on changing based on demand, requests, popularity, and lot of other factors.

So far we are 3 working on this parttime. We all have fulltime jobs and are developing onista on weekends and evenings.

prashantdesale10• 18 years ago

Missed addressing the payment question.

On Onista, users can pay each other the same way they do on eBay.

Paypal, Credit Card, Money Order, Cashier's Check, Personal Check

Harkins• 18 years ago

You don't. Alpha tests are, by definition, internal-only.

cperciva2• 18 years ago

Alpha tests are, by definition, internal-only.

There are several different definitions. One definition I like is that alpha testing is when you know there are still missing features, while beta testing is when the planned features exist and you're just trying to make sure that they all work properly -- in which case you could very easily have public alpha testing.

prashantdesale10• 18 years ago

We have done lot of testing on our application using some external users as well and fixed close to 2k bugs to make it more stable. As you are suggesting, we are planning public Alpha.

drandall• 18 years ago

If you do announce something, don't call it a 'launch' call it an 'alpha.' When you do a beta, call it a beta. Don't use the word launch until you're ready, b/c any news is going to say "But you launched in August! What's new?" and all you can say is "We're out of beta" or something like that.

prashantdesale10• 18 years ago

Very good advice. You are absolutely right. I did not think of that.

Thank you very much.

agotterer• 18 years ago

That actually happened to use with PriceAdvance. We "launched" our beta and then officially launched a few months later and a number of write ups mentioned we had already previously launched.

Also if your going to submit to techcrunch, make sure they are the first. They like to break news. You have a better chance of being covered if they get dibs.

abstractbill• 18 years ago

You list "Create Wikipedia page for Onista". I wouldn't advise it - you'll just get the page deleted immediately for not being notable.

huhtenberg• 18 years ago

> 8. Create Wikipedia page for Onista.

Wikipedia is not a promotion vehicle and using it as such is one of the most blatant and despised violations of WP editing policies (which you should familiarize yourselves with). Your article will be speedy-deleted; moreover, it will be very hard to restore the article later on even if your project does become notable. Keep this in mind.

prashantdesale10• 18 years ago

Thanks a lot. I was not aware of it.

demandred• 18 years ago

Just release already.

This from 28 February 08 (and blog posts dating back about 9 to 10 months as well):

We have been working on our Social Marketplace “Onista” for quite sometime and every week (and every month) we think we are almost ready for launch “only if we fix these remaining little issues”. To be honest, there is no such thing called “little issues”. There are bugs in product (tons of bugs) and we must fix those before launching.

Ask YC: How do we announce the alpha launch?

Related Stories