Behind the Paywall: An Experiment to Try & Save My Travel Blog

If there’s one thing I’ve learned about myself on my 30+ orbits ’round the sun, it’s that I have a hard time quitting at stuff. Once I decide to do something, I do it – and I usually succeed at it, if by nothing other than brute force and stubbornness. If failure starts to look like it might happen, I tend to put my head down and try harder, until that failure is inevitable and undeniable.

All this to say, I’m not giving up on my travel blog yet, though my most recent recap certainly suggested it as a possibility within the next year.

One of my goals for this year was to try some sort of paid model for accessing my content. In my recap, I thought that might be a Substack – I’ve already tried and failed at Patreon –, but after some research, testing, and signing up for a few fellow bloggers’ Substacks, I pivoted to try and set up a paywall instead.

This post is a living report of that experiment, sharing the tech I’m using, the different settings I’m using, and my results. I hope it’s helpful if you’re wondering how a paywall might work for your blog; if enough of us use something like this, people will begin to understand that content is no longer free in a world where Google doesn’t also drive free traffic to support our sites in other ways (ads and affiliates).

If you have any questions after reading, I’d love to answer them in the comments!

My Tech Stack

So to briefly cover the tech I used to make this work, I have to admit it’s a bit of a Frankenstein’s monster:

• I’m using Leaky Paywalls (free) for the basic paywall technology including trigger controls and appearance
• I’m using Profile Press (free) to manage subscription options within the paywall
• I’m using Content Control (free) to manage the conditional appearance of blocks to disable ads for readers who pay for that
• I have some hacky GA4 code written by ChatGPT (free) to try and track how often the paywall is triggering; it’s counting the number of sessions that have 5+ pageviews, which is more or less equivalent to the number of times the paywall triggers. (Okay, this isn’t entirely accurate, but I can’t for the life of me figure out how to get anything to trigger on the paywall itself, even using code provided by Leaky Paywall themselves!)

So I’ve managed to set up the paywall for free… But as you might expect, hacking together a free/paywall with a bunch of free plugins has had a bit of a learning curve. Every time I change one thing, something else seems to shift in the tech stack. There have been more than a few days when people couldn’t sign up at all – and I didn’t know until someone told me!

Initial Setup

Here was my initial setup/settings for the paywall:

• Everyone hits the wall after reading 5 Alaska articles (so on the 6th article); there’s no wall on any other destination content.
• Once they hit the paywall, they can’t see any more Alaska articles for 30 days. (Readers do not know this.)
• I set up my free/paywall with the following options:
  • Register for free to continue reading (with ads)
  • Pay for ad-free access, either a $5 monthly or $50 lifetime fee
  • The lifetime fee includes a code for a $50 product I sell

Within the first few days, I made a few other changes.

1. I switched it to be ALL articles (not just Alaska)
2. I dropped it to 4 free articles, so the wall triggers on the 5th article.

I made the first change to reduce confusion, and the second one to try and prompt the wall to trigger more often to start and get some meaningful data. I honestly didn’t know how many people even hit five articles on my site in a month; now I do – it’s about 5% of people on an average day.

As my friend Jeremy hypothesized when I was brainstorming this with him, the wall takes a little time to snowball. Maybe someone clicks an article from social media or an email one day, then comes back a week later and reads a few more, then comes back in another few days and finally hits the wall. That means it takes time to figure out the baseline of people who will sign up each day; as of two weeks out, that felt pretty well established, which is when I wrote this first update.

After the First Month (mid-October to mid-November 2024)

The paywall was up and running for a full day as of October 17th; I collected the following data through November 15th. (There’s a discrepancy in my GA4 data on the 16th I can’t figure out, so I’m throwing out that data.) Here are the basics:

• In that time, I had 44,148 sessions 😭 (That’s 69% down from the same period in 2023)
• The “paywall trigger” event was recorded 2,353 times.
• 145 people signed up through the paywall:
  • 137 people signed up for a free account
  • 8 people signed up for a paid account (7 lifetime / 1 monthly)

The paywall was triggered in 5.4% of sessions.
Once the paywall was triggered, 6.1% of people signed up.

Here are a few other ways of slicing the data…

5.8% of people converted through the “freewall” portion, which is a nice conversion rate, in my opinion.

I reached out to Mediavine’s Grow team as their “Exclusive Content” tool is the same as a freewall (a wall where people don’t have to pay money, they just give their email to keep reading). They do not collect or will not provide data on conversion rates. That seems like a big omission on their end since people then have no data to optimize from, but okay. (I have strong opinions about Grow that you are welcome to ask about any time!)

That also means I have no point of comparison for how my freewall performs compared to others.

0.3% of people converted to a paid account once the paywall triggered. This is low compared to industry standards – but it’s not common to have a hybrid wall.

Most websites either have a freewall or a paywall – not both. As mentioned, there’s no good data on freewall conversion rates, but paywall conversion rates typically range from 0.8-2.0%, so I’m well below that. I think that’s unsurprising as some portion of people are using the free option who might otherwise pay (but I think far more people would abandon my site outright if I didn’t have a free option!)

Just 0.34% of sessions resulted in a signup; 0.02% of those sessions resulted in a paid signup. While this doesn’t sound great, it’s actually not too bad.

If I had 100,000 sessions and my 0.02% paid conversion rate, that would be 20 paid people per month. At $50 for lifetime access, that’s an extra $1,000 each month. Not too bad for something that’s now completely passive!

Additionally, if I had 100,000 sessions and the same free conversion rate (0.22%), that would be another 220 people* on my mailing list each month. There are all kinds of ways to sell to those people to help drive more revenue from my site/mailing list.

*I’ve actually been tracking how many people are not already on my mailing list, and it’s about 30% new people, so more realistically, this number is around 60-70 new subscribers per month.

At this point, I’m happy with my setup and plan to continue running my paywall. I have a plan for offering a discount on Black Friday, and a few experiments I might run to try and optimize paid conversions better.

Month #2 (mid-Nov to mid-Dec 2024)

Hey hey! I’m back with an update; I thought I might try to make it easier to track my metrics and update this post going forward, so I thought a table would help:

	Total Events	Event %	Free Signups	Free %	Paid Signups	Paid %	Total Signups	Total %
Month 1	2353	5.4%	137	5.8%	8	0.3%	145	6.2%
Month 2	2712	7.5%	117	4.3%	4	0.1%	121	4.5%

As you can see, I had more events in Month 2 than in Month 1, which lends some credibility to Jeremy’s theory that it takes time for the free/paywall to ramp up to trigger for returning visitors. However, both my raw and my percentage of signups (both free and paid) were down pretty significantly in Month 2 vs Month 1. Does this mean that returning visitors are less likely to sign up, even though they’ve come back to my site again?

I have a few ideas for why that may be (Holiday shopping takes up their budget? People get annoyed and bounce?), but it makes the next month even more interesting: is this a trend over time that signup rates go down, or is it something specific to this time of the year? Will it improve as more people (hopefully) start visiting my site in the new year?

Only time will tell! Stay tuned for future reports ☺️ Also, next month, I’ll share some tips on how I’ve tried to improve my free/paywall conversion rates – and how well they’ve worked!

Month #3 (mid-Dec 2024 to mid-Jan 2025)

It’s now been a full business quarter of having the paywall on my site; it’s certainly been an interesting experiment so far.

Here are the updated stats:

	Total Events	Event %	Free Signups	Free %	Paid Signups	Paid %	Total Signups	Total %
Month 1	2353	5.4%	137	5.8%	8	0.3%	145	6.2%
Month 2	2712	7.5%	117	4.3%	4	0.1%	121	4.5%
Month 3	3433	6.5%	112	3.3%	6	0.2%	118	3.4%

As you can see the paywall triggered more this month but I had fewer signups again. I suppose this makes sense as some people are just not going to ever signup, and they can see 4 free articles (and then the paywall) once every 30 days. I don’t have a ton of insight into how many people that might be, but I still don’t feel like the dropping rates are a problem per se.

Not included in the numbers above are two important stats:

• I’ve earned $720 in total from having the paywall on my site. Sure, that’s not a ton. But it’s a few hundred dollars a month I wasn’t earning before.
• I’ve gotten 141 new *free* subscribers from the paywall. Again, not a ton in the grand scheme of things, but that’s 141 new people who I can try to upsell to the lifetime membership or other products I sell.

On the whole, I’m satisfied with the results of my experiment so far, and plan to leave it running – I’m also planning to roll it out on some of my other sites with slightly different settings (more free articles, lower lifetime cost, etc.).

I’m hoping to record a webinar soon explaining how I set this up step by step, so that you can do the same. If you want to know when that’s released – it will be free! – sign up for my mailing list!

Have any questions about my paywall experiment? Am I doing math weird or wrong? Let me know in the comments below!