Hacker News new | comments | show | ask | jobs | submit login
Every Step Costs You 20% of Users (gaborcselle.com)
71 points by gaborcselle on Oct 23, 2012 | hide | past | web | favorite | 32 comments

My favorite example annecdote: when signing up to pay for Dropbox, I got pretty far along the tunnel. I think after giving Credit Card details, but I'm not sure. At that point, I was presented with a choice: "Pay monthly ($10 per month)" or "Pay Yearly ($8 per month, paid in advance".

Being faced with this question, I decided to think about it, then left the page and didn't sign up. I did end up signing up a while later, so maybe this doesn't mean anything, but I could definitely imagine not remembering and ending up never signing up.

And all they had to do to was add another option: "Sign up for monthly billing ($10) per month. Don't worry, you can later convert to a year and start saving from that point".

I think Freshplum is a YC startup trying to solve exactly this problem for online retailers / subscription services: https://freshplum.com/

I wonder what the averages are for users of business software. Lead to trial. Trial to customer. Paying customer to active user. It would be interested just to get an average of how many dormant paying customers there are for web-based subscription apps. I know I am a dormant customer for at lease 3.

This is a "How long is a piece of string" sort of question. Want just rough SWAG numbers?

Lead to trial: 5% (high touch sales, high integration costs for trial) through 20% (self serve trial)

Trial to customer: 1~2% (online self serve SaaS), 30~50% (online self-serve SaaS with CC captured upfront), 80%+ (high touch, high integration coats, good sales team aggressively working new accounts)

Dormant customers: Very dependent on users and business practices of company. At AR it is about 15% or so. It would be higher but I sweep accounts once a year and self-terminate anyone I can't get in touch with.

As long as you're throwing out numbers, what's the overall impact of capturing CC up front on the whole funnel vs. when trial ends? Obviously you'll get more trial -> customer conversions if you capture CC upfront, at the cost of people in the top of the funnel, but in general, which is better when you step back to overall lead -> customer rates?

I bought your lifecycle email course, btw, and I'm just getting started working through it. I'm especially interested in this question if you assume that the lifecycle email approach is being maximized during the trial (in either scenario).

When you say CC captured up front, do you mean just requiring a credit card to activate your account, or actually automatically charging when the trial is over?

usually when you take CC up front you also automatically charge at the end of the trial. It's a good way to reduce tire kickers in your app, but can result in lost users at the top of the funnel.

In addition to losing 20% or so users at each stage of the funnel, the bigger picture is that most startups closely follow Zipf's Law, e.g.:

* Hacker News has (hypothetically) 300,000 unique monthly users.

* Of those, 30,000 have an account.

* Of those, 3,000 submit at least one comment or story over the course of the month.

* Of those, about 300 are regular contributors (at least 5 comments and/or story submissions per week).

I don't know what the real stats for HN are, but I doubt that is very far off.

wat. Those are variables that shouldn't even be on the same axis.

What you can do is to code those variables as 'levels of involvement'. Then maybe you may have a zipf's law distribution.

How are you measuring the app store pageview -> install step?

People that don't have DrawChat yet get invited with an SMS that contains a link to a page on our server. That page redirects you to the App Store. So we know how many people see our App Store listing.

On the other side, Apple provides you daily download numbers, which you then compare to the "first app open" counts on Flurry, Localytics, or your analytics provider of choice.

We can differentiate people who came to DrawChat organically vs. through invites by seeing if they had an invite waiting after sign up.

That makes sense. Just wanted to clarify that you weren't able to measure your organic app store pageview -> install conversion. I thought I had missed some hidden feature in iTunes connect, which is definitely possible.

My thought exactly. Is that something Google Play provides developers?

Is having a forced Facebook login necessarily a bad thing? Sure, the sign-up rate might only be 50%; but if it's inherently a social app, it'll pay of with a richer user experience once the user is 'signed-in with his graph'. Of course there are many apps that simply abuse the social grap for viral invites only - that sucks. But there are also apps that are just more fun - if they can overlay / provide access to the social graph. At the end: Every app attracts the users it deserves; those who appreciate it and those who don't. As an app developer, if my app is worth the trust of a social graph, I'd rather have the users that appreciate it.

Good point, Facebook login might cost you in the middle of the funnel but help you at the end.

You can likely mitigate the drop off at the signup stage quite drastically by first creating value, then asking users to sign up.

Signing is a conscious decision on part of the user to accept the possibility that your app will spam their wall with achievements and updates. If you ask them to take the risk before you show them why they should, it's natural to expect such a large drop off; I've personally deleted several apps because of it.

On the other hand, having an sign-in-less "Explore" type experience may actually hurt your signup conversions. Users get a feel for the app, decide that they'll sign up later, close the app, and never come again.

I might be hallucinating but I think Instagram used to have an "Explore" screen with interesting pics pre-signup, and in the latest versions they got rid of that.

You may certainly be correct, but part of your viral coefficient is likely driven by continued usage of your application. As such, I guess there are two ways you could go about it:

1. Spam user's social network on sign up; 2. Send out periodic blasts about progress/offers as the user continues using the app.

The first method will likely net you terrible reviews and negative sentiment, while for the second, an exploratory mode wouldn't really hurt you. The key would be designing exactly when and how the sign up should be. For example, maybe after 5 days or 5 app opens (whichever is more), the app loses functionality until you connect. It'd be different for every app. For games (or gamified aspects), you could just incentivize sharing.

On the iOS platform, you want to be very careful about timing when you ask for the push notification permission. which you'll need it for scheme 2.

One newbie mistake on iOS is to ask for notification permissions right after install. This is very similar to your point about signing up: You have to build some trust with the user first, then ask to notify them.

I meant socially, not for push notifications. Since, after all , the goal is to engage more users. You don't need push notifications here at all.

"Only 90% of of the people who download your app will actually open it - I'm guessing that people are stockpiling apps for the impending apocalypse."Hah!

Or attention-span-slippage caused by slow app downloads. Happens to me all the time.

Or load time/time to engagement. Sometimes I'll just dump an app if it throws up registration screens within 20 seconds of opening it.

Excellent point.

More like appocalypse.

My best guess is that folks hear about an app from the media or friends, download it, but then forget to use it or never hit the intended use case.

Both theories resonate with me, empirical evidence suggests that your's is the more likely.

I would trust a vampire chicken.

I find it amusing that the title claims that the cost is 20% on every step but the article itself says that the value varies and is not unique for every business. What's the point of choosing 20% in an arbitrary way like this?

Following this line of thought we could rename the article to "Every Step Costs You 50% of Users" if we chose that case as the example.

Quoting it as 20% kept the title concise while sacrificing very little correctness.

At Kera we're only limiting users to people with google accounts so it's interesting for me to read:

"completion rates as low as 50% with forced Facebook logins, while another one allows signing up with different methods (username/password, Facebook, LinkedIn) claims completion rates of 90%"

Do you have more information on this?

At Kera we're only limiting users to people with google accounts.

You mentioned "completion rates as low as 50% with forced Facebook logins, while another one allows signing up with different methods (username/password, Facebook, LinkedIn) claims completion rates of 90%"

Do you have more information on this?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact