Unlearning what Google taught me 🎓🚫
Plus finding a 3x more effective ad creative and many iterations on checkout for the holidays
Over the last 3 months, I’ve been working hard on Wanderly’s holiday season. Last year, before I had an app and started publishing books, I learned a lot about selling gift cards. This year my goal was to double my revenue from last year, which I accomplished, though barely. But I learned a LOT, and much of it was unlearning what Google taught me. Based on what I learned, I have a lot of confidence in what I need to do next.
After adding avatars in the fall, I turned towards building a new checkout flow in time for Black Friday. Instead of trying to reinvent the wheel, I looked at similar sites like Storyworth (subscription + physical book) and other slick checkout experiences (I had recently purchased the updated Remarkable Pro since I had loved the Remarkable 2.0) and tried to learn from what they had built.
I created a step-by-step checkout flow with two major paths: buying for self and buying a gift card. I also included a delayed send feature to enable holiday gifting use cases.
I launched it a week before Black Friday. During this update, I also installed Posthog as an additional analytics tool that I’d heard good things about. I used Google Tag Manager (which I worked on early in my career at Google), which made it almost effortless to add Posthog alongside Google Analytics. Getting actionable data from GA had been too difficult lately, and Posthog had a delightful session replay feature that seemed interesting.
After launch, I gleefully opened up the new session videos, wanting to see how users interacted with checkout, only to see a *major* problem: My landing page was absolutely awful.
I’d been using Wix for about a year, assuming that it’d handle a lot of website optimization for me. As I made edits, my website had been getting slightly worse, but I didn’t think too much of it until I saw user after user scroll through my site and receive fuzzy images that never resolved to their full fidelity; Wix had been serving full-size images in their original file formats, and with the majority of my users on mobile, the full-size image had never loaded. 😱
As a result, very few users made it to the checkout flow. I immediately began hand-coding my marketing page and, within three days, released a new version with fast-loading images. I immediately saw fewer people bouncing off my website, and more people heading to checkout.
In parallel, I’d also been working on my marketing funnel. I knew I needed to create more ad creatives for the holiday season, but I also needed social proof. At the beginning of November, I started a campaign on SocialCat that connected me to multiple social media influencers on Instagram and TikTok.
It was immediately validating. I didn't pay anything, just offered a free book – and dozens of influencers reached out to say they were interested! I carefully selected a varied group of influencers so I could test Wanderly with different kinds of families while also creating some really stellar advertising assets with real kids.
So, with a bunch of social media posts, new ad creatives, a new checkout flow, and a new homepage, what did I learn? Here’s the roundup. 🙂
No one really wants to “Buy Now”
As a PM who worked on many 0 → 1 projects, I have worked on a lot of onboarding flows. Google’s design philosophy is that you try to get people to accomplish their tasks as quickly as possible. And that’s fine if you’re Google: by the time people arrive at the website, they already know what Google is and they’ve also probably already decided to sign up for Google’s (usually free) service.
As Wanderly, no one knows who Wanderly is, and they haven’t decided to sign up. However, 40% of site visitors clicked on the “Buy Now” button at the top of my landing page... many within 10 seconds of landing on my homepage. This was a good sign for my advertising, but my actual conversion rate was incredibly low. I realized my super-streamlined checkout flow wasn’t actually meeting users where they were at.
I took a couple of stabs at trying to address potential user concerns here:
Adding more assurances (e.g. Money back guarantee, mentions of ratings, etc)
Adding more social proof (added TikTok videos from influencers)
Creating a “long” version of checkout that added more value proposition statements along the way (more like a Buzzfeed quiz than a traditional checkout flow)
None of these really seemed to be the “unlock”. I ended the holiday season just over my modest goal. Upon reflection, I think I was too locked into my Google playbook (minimizing steps, assuming intent, and replicating existing flows) instead of leaning into Wanderly’s product strengths and magic; i.e., I was iterating in a local maxima.
People love Wanderly illustrations
I tried many different ad creatives (and as a pro tip, I paid a Fiverr contractor to create some Canva templates that served as the basis of many of my creatives—it was 100% worth it!). The common thread between my top-performing ads was that Wanderly illustrations were front-and-center. Between the two ad creatives below, I lowered my cost per conversion by threefold.
Social media influencers were useful for assets, but not ads
I also ran influencer advertisements on Instagram and TikTok, and neither did well compared to the ads above. My TikTok ad with @sheshe_n_zz got a lot of engagement (55K views and over 400 likes!), but very few people left TikTok to engage with checkout. In retrospect, I’m not surprised.
On Instagram, I ran ads on 3 different reels. While none of them did well, one did perform better than the others. I expected the one with good lighting + product showcase to get a lot of leads, but it turns out that the heartfelt testimonial between a grandma and grandchild (that didn’t even showcase the app or books!) that led to more conversions.
Ultimately, I embedded the video and stills on my homepage and in email campaigns and pinned them to my Instagram feed. Next year, I’ll still work with influencers but be more targeted about how I use the content.
Not capturing emails is like cutting yourself off at the knees
It took me halfway through the holiday season to realize I was bringing a lot of people to my site and then losing them forever. Google had trained me that email marketing was gross. Even though I have observed countless direct-to-consumer companies ask for my email address, my lingering feelings from 13 years at Google prevented me from prioritizing it for Wanderly.
I finally had an aha moment in early December when I realized I had no way of reaching back out to any customer who had expressed interest. So I tried a few things to get email addresses, like a pop-up offering a X% discount or doing an abandoned cart follow-up. I still have to invest more in my email marketing strategy to leverage these contacts, but at least I have a way to reach back out to potential customers as I continue to improve Wanderly.
Promo codes drove interest but didn’t meaningfully change conversions
One of my big questions entering this holiday season was, “Is the price point right?”. I tried discount codes from 10-30% off, and none of them really seemed to move the needle. My promo code strategy was also haphazard; I changed discounts multiple times during the season, I sent out codes to different communities, and I had the email marketing promos on my homepage. I ended up spending too much time reconciling my multi-pronged approach.
However, showing a discount in my ad copy *did* make a difference in ad click-through rate. Next year, I might just apply a single discount upfront to all transactions during the holiday season, so I can get the boost from advertising but not have to worry about logistics.
I misunderstood my gifting use case
My original hypothesis was that during the holiday season, Wanderly would be most compelling as a digital gift from relatives to a family they didn’t live with. It’s possible that’s still the case, and I didn’t reach that audience this season. What became clear from the questions in my checkout flow was that ~60% of the people I reached with my ads were parents searching for a holiday gift for their own child.
If you like Wanderly, you want a lot of it
Of the sales I did make, at least half were from users purchasing multiple items. In some cases, a user would want to buy a family subscription for themselves but also want multiple books. In other cases, a user would want to buy gift cards but for multiple recipients. I’d set my own bar at converting a single user to buy a single book, and instead, I should have embraced the enthusiastic buyer persona.
Sending a note forward in time to myself, next holiday season I want to:
Start working with influencers in early October. Their content was useful, but it took a while for book production + content production, and I want it all ready before Black Friday.
Create more ad creatives that highlight Wanderly’s illustrations.
Adjust the “cart” mechanics to allow for multiple books & multiple gift cards.
Create a thoughtful cart abandonment strategy.
Hard code a discount into checkout for the holiday season.
Using user studies, determine how to best sell Wanderly and understand what prevents final conversion. Maybe it’s showcasing the magic earlier, wanting something physical under the tree, seeing a proof of a book, or changing the product altogether.
Ultimately, I put product development on hold so I could learn as much as possible during the holiday season. It was a hard trade-off because there have been things I’ve been wanting to build for months, but I knew the learning ROI would be best over the holiday season. Starting today, I’m moving back into product development mode… now armed with all the things I’ve learned this gifting season.
Wishing you all a happy, healthy, and prosperous 2025! Thanks for sticking with me :)
So helpful and detailed, thank you for sharing!