As alluded to in my last post, my biggest focus at the moment is to work on products whose need has been validated. This might mean buying a business with an existing set of customers, as I went into detail in the last post. But it could also mean thoroughly validating the need for a product before building the product.
How does one validate the product need without a product? There are a number of ways to do this, but one of the easiest and most fun ways is the “landing page method”. This is what you do:
- Build a landing page
- Direct paid traffic
- Measure sign-ups or pre-orders.
The “landing page method” is best explained in this brilliant post by Justin Mares. Justin started a bone broth company, Kettle & Fire, by employing this methodology to validate the product need. First, he built a basic landing page, with a logo, some good copy and an “Order Now” button. He also collected payment for the order via Paypal. Since he did not actually have a product yet, after payment, he would send them an email saying the product was out of stock and offer them the choice of a refund or a discounted future purchase.
Then, he directed Bing Ads at the landing page. The results were stunning: $50 in ads translated to ~8k visitors, and 30% of those visitors even clicked the Order Now button. Some people went all the way and even paid for the product that they had only heard of 2 mins ago!
He ended up generating $500 of actual revenue from a marketing budget of $50 and a barebones landing page. Clearly, there was a market for a product.
An even bigger example of a company that started from a landing page is billion-dollar startup Flexport. As he describes on this podcast, before building the product, the founder Ryan directed Google Ads to a landing page. He had surprisingly good success with it. FoxConn (manufacturers of the iPhone), Saudi Aramco and a bunch of other massive companies signed up.
The landing page method is a great way to validate product needs.
Failing the landing page test doesn’t mean there isn’t a need for the product, but succeeding a landing page test almost certainly means there is.
This sprint, I embarked on running a landing page test for one of the ideas floating in my head.
All sorts of classes, especially yoga and other workout classes, are moving to Zoom due to COVID. Zoom doesn’t make it easy to collect payments from attendees. Enter Pay2Zoom. An event page for your class that collects payments on your behalf, and sends a Zoom link only to paying attendees.
I built a very simple landing page using Unicorn Platform in one afternoon. When you click the "Get It" button, you are taken to a sign-up page where you enter your email. After the email is collected, the user is informed that they'll be notified when the product is available.
Building the landing page was the easy (and fun!) part. I easily spent way more time on user research in order to craft the copy, and come up with a list of keywords for the ads that direct to this landing page.
I crafted up some ads for Google Adwords.
I immediately ran into a few issues:
- Keywords related to a payment service called Xoom overlapped significantly with the keywords I was targeting. All those clicks that were looking for Xoom-related topics were basically useless. I quickly set these search terms as negative keywords.
- I did not put in any geographical limitations on where my ads were shown. A lot of the sign-ups I initially got were from India or Nigeria or the UK, where English is also the lingua franca. Given that the product had to do with payments, and the substantial variation in rules surrounding payments, I definitely preferred to focus the product in the US market, at least initially. However, I changed the ads to be U.S. only about halfway into the experiment.
- Unsurprisingly, the keywords that were most effective were "zoom payment" and "zoom payments". But these keywords are ambiguous, and overlapped significantly with user intent re: how to pay for one of Zoom's many subscription plans. I’ve only used Google Adwords in small ways previously so I found a course (KeywordsAnywhere and Google Keyword Planner were useful tools as well) and watched some relevant lessons to get a better understanding how to find keywords, how to set up negative keywords, how to use "search terms" to expand my list of keywords. Overall, however, all this didn't move the needle much, as we'll see below.
- Namecheap (pay2zoom.com): $9
- Mailchimp: $10/month
- Unicorn Platform: $8/month
- Ultimate Google Ads training: $10
- Google Adwords: $200
We finally get to the exciting part. How'd it go? I ran ads over the course of a week with a total budget of about $200, which resulted in 13 conversions. The vast majority of conversions were from outside the US.
So... not great! ~$15 per conversion might be viable long-term financially. But product viability vs. product excitement are two different things. Unfortunately, I had to conclude that there wasn’t sufficient demand to invest in building the product.
I was only going to spend time on this project if was going to be a smashing success. And this wasn’t it. But hey, better to fail early than fail late.
Onward to more landing page tests! Long live the Landing Page Method!
After running this experiment about a month ago, I heard about zmurl, a project pretty much identical to Pay2Zoom created by a friend of a friend. Over a Zoom call, I learnt that they had been processing a decent chunk of traffic, all of which was organic since they launched on TechCrunch. From their experience over the last couple of months, there was a ton of demand for support+payment tools related to recurring online events.
The success of zmurl reiterates the point made earlier:
Failing the landing page test doesn’t mean there isn’t a need for the product.
Is the Landing Page method still a good tool for validating need? Yes. There's a high false negative rate but the true positive rate is also very high.
Ultimately, the point of product need validation isn't to perfectly confirm whether there's a need. Rather, it is to quickly evaluate if it's worth investing more time into the product.