- newslepear
- Posts
- 🍐#89: Pricing page framework, learnings from Resend, and open text field vs select options for self-reported attribution
🍐#89: Pricing page framework, learnings from Resend, and open text field vs select options for self-reported attribution
Jakub Czakon
November 10, 2024
Hey,
A pear was really anxious, felt threatened, under attack. So they went to a doctor. Diagnosis? Pearanoia🍐.
This week on the agenda:
Open text field vs select options for self-reported attribution
DNA of a great pricing page from Elena Verna
Learnings from Resend by Dani Passos
+ a few bonus links at the end
Total pearusing time: 5min
Before we start a word from this week’s sponsor:
Hyper-relevant, cookie-less developer ads
Reach developers on developer docs/sites like Flask, Jupyter, Linux Mint, or daily.dev.
Serve hyper-relevant ads based on the content of a particular docs page or article. No cookies involved.
Dev tool companies like Snyk, Datadog, or Mongo use it.
There is a special offer for newsletter subscribers. You get 10% off any new campaign, and 15% off a $5k+ campaign that starts in November.
Developer marketing insights
1. DNA of a great pricing page from Elena Verna
Elena Verna is a well-known growth advisor. I love her memes on LinkedIn, her thoughts in the newsletters. And I love this pricing page skeleton.
Simple but powerful. I bet a few buckets of gold in experimentation $ went into arriving at this. A few interesting things:
No header “placeholder” copy like “Pricing” or “Pricing that works for every team”. It doesn’t add much and it moves the table down. BTW, I do get the argument that reinforcing what your tool does can work better.
Logos above the main table to build social proof. This is something I never see, I haven’t tried but would love to test it out. I see how this can boost conversions for the undecided.
+ lots of well-thought-out bits like 3 plans (individual/team/organization), For whom each plan is designed, “Everything in +” pattern to show what extra you get, CTA moved up, and more.
In the article where this image is from she goes way deeper, does pricing teardowns of some awesome companies, and more.
Read the whole thing here (you need to subscribe to the newsletter, but you’d want to do that anyway ;).
2. Open text field vs select options for self-reported attribution
As you probably know by now, I like to use self-reported attribution to figure out where people are coming from and what channels are working. And I believe in keeping the “How did you hear about us?” question required and open text as opposed to options that people can select. Especially early on.
Big benefits of keeping it open text include:
You don’t bias what the users answer
You get to learn about channels you didn’t know about
You don’t need to deal with randomizing answers, tweaking options you display overtime etc
You get a piece of qualitative insights on the user journey sometimes (just sort from the longest answers and filter out junk).
Yes, a big % of answers is “aasd” but I guarantee you that a big % of the first option selected (randomized or not) carries as much information as “aasd”. It just looks better.
“But you can’t do quantitative analysis on free text self-reported attribution” I hear sometimes.
Of course, you can. You just need to categorize the outputs. You do that by hand, with a rule-based method, or machine learning model.
But probably the best option today is to use a prompt to categorize the answers. Here is an example of one that worked for me.
I have a required field on the signup form on my website. It asks the question "How did you find out about us?". Users can input anything.
In the next message, I will provide you with the responses.
What I want you to do is: read the responses and categorize them.
Return the results in the table with the following columns: Original input, Assigned Category.
So is there no situation where you should use the select options?
It is just like with surveys. At first, you want to keep it free-form to get a wide set of options. Explore. Over time as you become more certain in the options people choose you can move to option selection. Make quantitative analysis easier.
But with selections + other (which people rarely fill btw), you lose that exploration. You will not learn about that new influencer, or obscure meetup, where people talked about your product. And with prompt-based categorization, you can have the best of both worlds. So why not just do that?
3. Learnings from Resend by Dani Passos
Dani Passos is a devrel lead at Shopify and had some interesting thoughts on the Resend growth trajectory. Figured I’d share.
My takeaways:
You want to create high emotions. The worst that can happen to you is not someone calling you out. It’s crickets. In this case, if you play your cards right, a big influencer pointing to your downtime, when you cheekily share your vs competitor's page during their downtime, can end up with many folks checking your product out. Ideally, you’d just get love all around, and Resends gets plenty of that, but either way emotions >> crickets.
When someone talks crap about your company take the high road. Be a Ted Lasso of Twitter and reply to levelsio with powerful company backstory. But do that with kindness. If you want to mic drop like Zeno you can recommend a book that your adversary wrote. Way to go.
Need more developer marketing insights?
1. Work with me 🍐
"Thanks so much for your time and all the thoughtful feedback coming from the workshop.
I feel like we are in a much better place to start our website rebuild. "
If you want my help I do Workshops (60-minute session on whatever you want), Teardowns (audit+suggestions for your homepage, messaging, ads etc), and longer-term Advising.
2. Bonus links to check out
3. Join our Slack community
"Been here 20 min and already folks are sharing great advice."
2000+ dev tool CMOs, heads of growth, product marketers, and other practitioners talking shop.
Reply