Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: How to estimate (best case) early user growth?

news.ycombinator.com

Post reply on HN
Print view View original

11–19 of 19 posts

Re: Ask HN: How to estimate (best case) early user growth?

#11

Post by idlewords »

smanek wrote: post #7
tom wrote: post #4

Have you launched something yet? Paid service? Free? Freemium? Without launching and looking at real, initial numbers just assume you'll be like 99% of the startups out there and get none. That's what your potential jnvestors will do until they see some real numbers. Seriously, get something real, prove you're on to something, and THEN talk numbers with investors. You've got a bigtime uphill battle without at least o…

Completely free. I agree with you - we won't be raising money till after we launch and (hopefully) gain some traction. But one of our founders is handling the business side of things and really wants to get at least some rough numbers now. Given his considerable business experience (and the lack of downside to letting him create his model), it sounds like a reasonable request. Rather than have him pull the numbers ou…

Not having a model is better than using a model based on imaginary data, which is what you'll have until you launch your site. As other people here have pointed out, traffic for new sites is very bursty and depends in large part on who links to you when.

Re: Ask HN: How to estimate (best case) early user growth?

#12

Post by nico »

Whatever estimate you come up with, will be BS, and if investors know their way around, they'll know this. This is true not only for your user base, but your sales and revenue projections as well. Probably the most accurate number you can get are expenses, and even that won't be very good.

But all of the above doesn't mean you shouldn't do your homework. So you still have to do some research, learn how other startups grew, estimate your market and market penetration in time. What investors really care about is the reasoning behind estimating those numbers, rather than the numbers themselves.

Here's a good series of posts with useful tips for pitching a VC:

- http://bothsidesofthetable.com/pitching-a-vc/

Good luck!

Re: Ask HN: How to estimate (best case) early user growth?

#13

Post by MWarneford »

I agree with the general sentiment that traction beats a prediction, however, I think its valuable to understand the relationship between the key metrics before you start.

Thats not to say that your model will match reality, but its certainly valuable to understand the dynamics involved.

I'm working on a series of blog posts that develop a virtual world business model. However, the principles also translate to freemium & social business too.

I recently wrote a post (with google spreadsheet) illustrating how the addressable market size (network carrying capacity) effects the viral growth of your application.

The theory goes that as the network becomes saturated the viral invites, paid acquisition, etc starts to fall on players who have already evaluated your service. Over time the success of your marketing and viral actives decline.

Check out the post at: http://dubitplatform.com/blog/2009/7/30/calculate-how-the-ad...

The google spreadsheet linked in the post illustrates how I've set about modeling viral growth, paid acquisition, ARPU, churn, and so on.

The model applies to any viral application - but its worth noting that products relying on advertisements can expect a significantly lower ARPU, knocking into your paid user acquisition.

You may also find this post interesting: http://dubitplatform.com/blog/2009/7/26/calculate-how-much-y...

It goes through the relationship between ARPU, LTV, and how your viral coefficient effects how much you can afford to spend acquiring a new player.

I think paid acquisition is a under utilized strategy within social networks - Zynga aggressively bought FBK users.

Hope it helps!

Re: Ask HN: How to estimate (best case) early user growth?

#14

Post by petervandijck »

smanek wrote: post #7
tom wrote: post #4

Have you launched something yet? Paid service? Free? Freemium? Without launching and looking at real, initial numbers just assume you'll be like 99% of the startups out there and get none. That's what your potential jnvestors will do until they see some real numbers. Seriously, get something real, prove you're on to something, and THEN talk numbers with investors. You've got a bigtime uphill battle without at least o…

Completely free. I agree with you - we won't be raising money till after we launch and (hopefully) gain some traction. But one of our founders is handling the business side of things and really wants to get at least some rough numbers now. Given his considerable business experience (and the lack of downside to letting him create his model), it sounds like a reasonable request. Rather than have him pull the numbers ou…

"Given his considerable business experience (and the lack of downside to letting him create his model)"

Warning signs! Translation: "He doesn't know what he's doing" (or he wouldn't be making predictions on made-up numbers), "and I'm letting him waste some time" (ie, "the lack of downside").

Not sounding good :)

Re: Ask HN: How to estimate (best case) early user growth?

#15

Post by petervandijck »

smanek wrote: post #9
bilbo0s wrote: post #8

Earlier quoted context omitted.

I agree with @tom's sentiment here. Though I would like to add something, just as a note to you. If you have 'investors' that are interested in those estimates, you should probably rethink the wisdom of doing business with them. The portion of your post that indicated that potential investors were interested in the numbers threw a red flag up for me.

I think it's just informal interest. One of our founders has a lot of VC friends/ex-colleagues and he just wants to be ready for our launch in a few weeks. We won't be raising money that early, but it seems like he can talk to some people and get us on their radar earlier than normal (so they know us (and we know them) for when we are ready). I'm willing to defer to the business-founder with experience on this matter…

If he wants to make up numbers now instead of waiting a few weeks and getting real (ie. dissapointing) numbers, that's bullshit radar material.

Re: Ask HN: How to estimate (best case) early user growth?

#16

Post by joez »

As lots of people already said, your user growth projection is probably not going to be very exact. ARPU can be pretty close. Expenses can be really close.

Depending on your product, you might be able to set up some initial targets and just run with those. I.e. if your marketing plan is to start at colleges (or one particular college that you have strong affiliation with), you can calculate a growth for that slice of your market. Since you are pretty set on creating a model, this will probably help you get closer. I.e. your user acquisition went from a geometric curve to more lumpy. Kinks in your curve are a good indication you gave it some thought. I.e. is there a point at time where you switch to a more aggressive growth by buying users?

If you have some goals in mind of when you want to hit profitability, you can backout a user growth rate. So you have ARPU and a pretty good idea of what your expenses will be, just divide expenses by ARPU and you have a user target. So if you then have a goal of hitting break even in 13 months, you know that you will need a x factor of geometric growth. You can make this more complicated by changing your ARPU (I don't know your business model but probably ARPU goes up as you hit critical mass?).

Again, probably still a BS hockey stick but hope this helps!

Re: Ask HN: How to estimate (best case) early user growth?

#17

Post by petervandijck »

smanek wrote: post #3
lsb wrote: post #2

What are your numbers so far?

We are prelaunch, so no numbers yet. We are doing a small 'soft launch' in a few weeks (I'll mention it here when the time comes) - but the business guys could use a hand estimating 'best case' numbers (some potential partners, investors, etc. are interested). Incidentally it would also help me figure out if my back-of-napkin load-testing numbers are in the right ballpark. I know this is a bit premature for me, but t…

OK, let's get some numbers out for a typical "not selling anything for money" startup:

- First 1000 users means nothing whatsoever. - 10,000 users means nothing. - If 100,000 free users take 2 years, you should shut down. - When you double every, say, 3,4 months above 10,000 users, you're doing great. - Active users (used service in past month) are what matter, not signed-up users.

If it's a for-pay service, ignore above, then you can just calculate the $$

Re: Ask HN: How to estimate (best case) early user growth?

#18

Post by lowkey »

idlewords wrote: post #11
smanek wrote: post #7

Earlier quoted context omitted.

Completely free. I agree with you - we won't be raising money till after we launch and (hopefully) gain some traction. But one of our founders is handling the business side of things and really wants to get at least some rough numbers now. Given his considerable business experience (and the lack of downside to letting him create his model), it sounds like a reasonable request. Rather than have him pull the numbers ou…

Not having a model is better than using a model based on imaginary data, which is what you'll have until you launch your site. As other people here have pointed out, traffic for new sites is very bursty and depends in large part on who links to you when.

I call bullshit. The purpose of making a forecast is not to predict the future, it is to identify errors in the logic of your assumptions. The experienced business guy is simply putting together a model based on obviously fictional data to help understand the financial dynamics of the business - probably to see if there even IS a real business.

It's nice rhetoric to proclaim that since user adoption figures cannot be predicted with accuracy, they should not be predicted at all. Even without accurate forecasts, it is trivial to calculate the required breakeven users and examine scenarios where that breakeven is achieved in a short,medium, or long ramp period. Other valuable metrics are also more easily discerned from a model than by pure guess.

It sounds like me that that business guy is simply doing his job - and it sounds like he knows his job better than many readers here.

Re: Ask HN: How to estimate (best case) early user growth?

#19

Post by idlewords »

lowkey wrote: post #18
idlewords wrote: post #11

Earlier quoted context omitted.

Not having a model is better than using a model based on imaginary data, which is what you'll have until you launch your site. As other people here have pointed out, traffic for new sites is very bursty and depends in large part on who links to you when.

I call bullshit. The purpose of making a forecast is not to predict the future, it is to identify errors in the logic of your assumptions. The experienced business guy is simply putting together a model based on obviously fictional data to help understand the financial dynamics of the business - probably to see if there even IS a real business. It's nice rhetoric to proclaim that since user adoption figures cannot be…

You don't need to badger your devs for anticipated traffic figures in order to calculate those kinds of scenarios. It's one thing to calculate various thresholds and set up hypotheses that you'll test later, once you've collected real data. It's another thing to make numbers up out of whole cloth so you can fool yourself into thinking you understand more than you do about your business.
Post reply on HN