Lean Startup Landing Page Test: How do I measure success?
I've set up a landing page for a startup idea as per the instructions here. In other words I have already done the following:
- I've set up a basic landing page quickly explaining my idea
- I've included an email capture on this page
- I've set up an AdWords campaign with a handful of targeted keywords
I understand that the answer will likely depend on the specifics of a given situation, but I have no ballpark whatsoever to base this on. Does anyone have any experience gauging the sucess and/or failure of this kind of landing page test?
3 Answers
The way that I do this is to start with the answer and work with your way back:
- Start by calculating how much an email is worth to you. This can frequently be very subjective and at this stage in the game, you need to make a lot of assumptions. It goes something like this: Let's assume you think you can convert 10% of your respondents to customers and a customer is worth $50. That means each email is worth $5. Configure a Google Analytics Goal with this value.
- Tie your Google Analytics and AdWords accounts together. Once you do this, AdWords will calculate your cost per conversion. Based on your example above, if it takes $100 to get $25 worth of emails (i.e. 5) then you're losing money and you need to decrease your cost per click, increase your conversion rates or both.
Generally speaking, a 15% to 20% conversion rate on your landing page is pretty good but it varies a lot depending on the offer, the industry and the keyword quality. I have some clients that are in the low single digits for conversion rates but the landing page is a direct offer to do business and those conversions are highly lucrative. I have other clients who see better than 40% conversion rates for a free ebook that is a very tight match with the keyword searches.
In your case, if all you're doing is collecting email addresses from people who are expressing interest, your conversion rates should be at the higher end because there is very little risk on their part.
-
Thanks this is useful information; do you have any suggestions on strategies upon which I can base my estimated value of an email on real information? Perhaps I should be looking for industry standards of how many emails might turn into a converted sale, or simply try to get in direct contact with a few of the signups to see if I can sell them on it? – Nick Budden 11 years ago
-
It's really tough to use industry data because there are so many variables involved, most notably the landing page. I've seen case studies where landing page optimization has improved conversion rates by multiple hundreds of percent. Your customer conversions are going to depend on the strength of your offer, its pricing and the effectiveness of your landing page. If I'm just starting off without much data to go on, I'll generally make an assumption on the low end like 1% just to be conservative. – Jon Di Pietro 11 years ago
It's a good question and frankly I don't know what metrics would be a sign of positive feedback.
That said, it seems that the way your current test is set up, you're testing to see if users want to be notified when you launch. Don't you instead want to test if users are willing to pay $50 to access the paid content?
So I'd change the test to "For $50, HypeUnion teaches you club dancing from some of the world's top instructors. Click here to sign-up." Also, instead of saying "for a $50 one-time fee" I'd rather simply say "for $50, HypeUnion ..." No one wants to pay a fee!
And only then, after users click the Sign-up button and validate your value proposition, you tell them it'll be launching in December 2012 and give them the option to be notified of the launch via email. But I think that for the purpose of the test, you shouldn't mention that it's not live until you find out if they're willing to pay!
While you're at it, why not also test for different price points? For instance, 2 weeks at $49, 2 weeks at $39, 2 weeks at $29 and 2 weeks at $99. There's a reason companies price their product in increments of $9 rather than round number and I think you should look into it.
Good luck; sounds like a good concept.
-
Hey @frenchie, you're completely right that what we need is to know if they're willing to spend the $50...not just whether they're willing to register the email. My only issue with asking them to buy the product, and then telling them it's not ready yet after they've tried to order, is that the current (very basic) landing page would presumably convert horribly because it is just an image and an email capture. To properly test if they would be willing to spend their $50 I think I would need to have a landing page that had at least a _reasonable chance_ of converting paid users. – Nick Budden 11 years ago
-
Well then you could tweak it further and have the call-to-action read something like "click here to find out more and sign-up." Overall, if out of 1,000 different page views you can't even get one person to click, then you know you're in trouble. However, if out of 1,000 views you get 5 clicks, then at least you know going in what kind of low baseline to expect. And if after you fancy-up the homepage you go from 5 to 7 sign-ups per 1,000 then you'll have an idea of how effective your homepage really is. I really recommend you try my idea, even if it's only for 10 days. – Frenchie 11 years ago
Nick, very interesting question. I agree with what John has said, but when you circle around and try to evaluate the value of an email, you run into problems again… What really is the email worth? Are you able to achieve the conversion rates you assumed? Industry standards vary so much; it’s virtually impossible to make valid assumptions.
With our startup Compilr, I fundamentally made two extremely inaccurate assumptions. I assumed that we’d be able to achieve a 1% conversion rate and charge roughly $10 per user per month. If you asked anyone in our industry they would say those numbers are very conservative. I even backed this up with a survey data with a large sample size. In the end, we never hit those targets but eventually did make the numbers work but our projections are about 7x lower than we originally anticipated.
From this and other startups that I’m involved in, I’ve concluded that:
- You can never make accurate assumptions, no matter how much experience you have.
- People who say they would pay vs. handing over a dollar are fundamentally two different numbers.
The only way to overcome these inaccuracies in executing a business is by pre-selling the product or creating an MVP that you can sell. You need baseline metrics, before you can build an accurate financial model and conclude that you have a real business.
In your case, you may not have a product ready yet so you may feel uncomfortable selling anything and you don’t want to invest your time until you have any interest. To reduce the anxiety on both sides, why not simplify with an MVP you could pull off in a week? The product could be a page where the user selects what style of dance they want and then present them a list of various YouTube links. Over time as you get more paying customers, you spend more time improving the product and adding new requested features.
