Live data from Hacker News

Ask HN: Question about CTR, conversions, and validation

news.ycombinator.com

11–14 of 14 posts

Re: Ask HN: Question about CTR, conversions, and validation

#11
There's a lot of relevant information we don't know (targeting, platform, demographics, competition, payment etc.).

If your 6 signups are paying anything, I would continue iterating ads and working on the product.

If those 6 signups are unpaid, and just people who signed up after viewing the site, I would stop spending and reassess the product and marketing. I'd expect only a small fraction of those people to ever pay anything, with many never returning to the site at all.

Also I wouldn't focus too much on the CTR or CVR - you are getting dirt cheap (likely untargeted) visitors ($0.019/each) so you should expect low rates. What matters is your ROAS (return on ad spend). You should be tracking that obsessively.

Re: Ask HN: Question about CTR, conversions, and validation

#12
It depends completely on the medium. We had a ratio of 3% of users on our app paying, so out of 1000 users, 30 would buy something. (Advertised as a recipe app, but we funneled them to ingredients)

But when we did ads, it was practically 0 out of 1000.

I wouldn't pay too much attention to CTR. CTR can mean a range of things. The best way to have a high CTR is pornography, and people who optimize solely for CTR often end up with ads that are borderline porn. The next best are trypophobia triggering images, or something like a celebrity promising a miracle weight loss. That doesn't mean they'll buy, their lizard brain just clicks the button.

One way to think of it is like fishing. If you go to the ocean, you have a very large net and basically don't need to target anyone. The majority won't care about your product, but you'll have a huge, scalable market to advertise to. Leave it for over 3 months or so, they'll come.

Or you can do targeted fishing in a small pond (aka sales). You won't catch very much fish, but you'll end up with a very high conversion rate and most of the people you get are already interested. In my experience with sales, about 1 in 2 will respond to an email if it's targeted enough and if you follow up, so it's basically a 50% CTR?

You don't necessarily want to do sales, not for courses, so you might opt for something in the middle, like using influencers (aka KOL).

Mass marketing is effective if you can get it from multiple angles. There's a saying that someone has to see the product 8 times before they click. So they go to your Udemy, it shows up thrice on Facebook, a friend talks about it, they see it on a billboard, then on a car, and then their spouse mentions it, and finally they buy. This works better for shampoo brands and such.

Re: Ask HN: Question about CTR, conversions, and validation

#14
First of all I want to say thank you so much to all of you. Your responses are really helpful and I gained a lot of insight from them, appreciate it! You guys are right the focus should be on the payments not just the signups. So far non of the users who signed up purchased the courses. I tried to reach out to the ones that signed up but none responded. I'm going to talk to more users and get more feedback as you guys suggested. As for the kind of ads, they are targeted to a specific group by location, age, interests, and search queries which we believe will be our target customers. We are doing the ads through Google.
Post reply on HN