Home Designing Websites What I Learned from Frying 200 Landing Pages in 12 Months

What I Learned from Frying 200 Landing Pages in 12 Months

by admin

What I Learned from Frying 200 Landing Pages in 12 Months

200 startups

Over the past twelve months, I’ve fried the landlines (project landing pages) of 200 startups. Indie Project Landing Pages, Landing Pages seilaps venture capital-funded, and for-profit corporation branding, and branding from a variety of industries and for a variety of audiences. An average of 25 minutes of frying is more than 3.5 full days of page frying to increase conversions.

What is roasting

Each fry-up is a personalized 20-minute video review of the landing page, where I identified fixes that would turn more visitors into customers.
All of this is based on best practices, test results, and my work as a conversion rate optimization expert.
But what have I learned about high-conversion landing pages? And about running a business? And what can you learn from me? It turns out, quite a lot…

Why did I Launch Roast My Landing Page

Roast My Landing Page started as a side project related to pandemic to allow me to support early-stage startups while generating a second stream of income on top of my freelance marketing work. I ended up with over £70, 000, over 2, 000 email subscribers, and an ever-growing list of insights into the travails and challenges of early-stage businesses and the amazing people who run them.

It’s all about the funders.

And it really came down to the funder. Each person had their own story, passion, and ideas. They were eager to learn how to do justice to their product, service or mailing list and show it to the world in the form of a high conversion rate landing page.
This post is for every founder who has thanked me, challenged me, and (on rare occasions) fought me. It’s been awesome. So, here’s what I learned…

The 9 most common (and easily fixable) things that funders overlook

One goal

50% of funders had one clear goal for their landing page – register, download a lead magnet, or order a demo. The other 50% had multiple calls to action, often with the same priority. This often leads to analysis paralysis and visitor confusion.
How to fix it : focus your landing page on one conversion goal.

Emphasis on UTP

After ordering a roast, Roast My Landing Page customers are asked to fill out a short questionnaire. One of the questions : "What makes your business unique?" Almost every founder was able to elegantly lay out the UTP (unique selling proposition) of their product or business in a form, but only one in five had something like that laid out on their landing page.
One of the key points of your branding : what makes you unique. If a buyer is in the consideration (solution comparison) stage, your UTP is what will make them remember you and decide if you are the best fit.
How to fix : compare your product with your competitors and current practices.

Clear and relevant social proof

Reviews, ratings, awards, or quantitative ratings that confirm that other people are using your product and liking it. Only about 40% of the landing pages had this at the top of the page (what people see when they first get to your page). And of those, only about 50% used testimonials that described the pain and benefits listed elsewhere on the landing page.
How to fix : move your social proof up the page, make sure it’s concise, matches the text in your ad and is relevant to the buyer and the visitor.

Simple language

About 1 of the 8 landing pages I read, I couldn’t understand the first time I read them. It took several attempts to just "understand" these landing pages. Only I had the monetary incentive to try to make sense of what was going on, and the visitors to the landing pages did not.
How to fix it : avoid technical terms and abbreviations, write in plain language. Remember that you are talking to a person, even if they are a B2B buyer. Ask yourself if a 12-year-old can understand your landing page.

Real pain

PAS (pain – excitement – solution) is a common copywriting technique used to increase conversions. Most of the landing pages dealt with the pain they addressed, but only 1 in 15 tried to excite or heighten the visitor’s experience with emotional language and vivid imagery. The ones that did it well created much more powerful landing pages, which prompted me to explore the solution.
How to fix : Excite your visitor by painting a vivid picture of pain, using emotional language, stories, and visuals.

Clear benefits and usecases

Many of the product’s front-ends detail the product’s features, but ignore the benefits and use cases. Studies consistently show higher conversions with benefits-based language. The visitor had to think about exactly how the product would benefit them or solve their problem.
How to fix it : visitors don’t need to think about how and why the product will benefit them. Tell them and show them the benefits and examples of use in language they can understand.

A notable call to action

On many of the landing pages there has been a call to make something , without explicitly stating what exactly Yes, they were going to register, but … how long would it take? What were they agreeing to? How much would it cost? What was the technical set-up involved?
How to fix : add context to your call to action so the visitor knows what to expect.

Ask instead of overthinking

3 out of 5 founders mentioned frustration because they didn’t know what was going on before the roast. They didn’t understand why their landing page wasn’t converting visitors into customers. However, very few actually asked their visitors about it.
How to fix it : sign up for GetSiteControl and add an exit intent survey to your landing page. Find out why they are leaving. Take those reasons into account on your page.

Knowing your stats

Only 2/5 founders could tell me about the current conversion rate of their landing page. In most cases, these people didn’t even know if they had a conversion problem on their page, but they ordered a fry-up anyway. Most of the founders had Google Analytics installed, but they had limited ability to track goals or hadn’t looked at reports in months.
How to fix : set up analytics and event tracking. Find a conversion baseline for future experiments.

7 (slightly more) advanced ideas for improving the effectiveness of your webpages

Approximately 1 out of 4 orders was a fry-up of bandings skylaps where they clearly used a high level of expertise and skill to create a landing page. These businesses still had conversion problems, which were mostly in the following 7 areas.

A niche has not been formed

I’ve seen this over and over again, and I’ve often been asked this. "Should I target a larger or a smaller audience? Of course, the bigger the market, the better?" My advice has always been the same. Create a landing page for as narrow an audience as possible until you find your first customers. Later, create more landing pages.
Landings that focus on multiple "personas" and use cases almost always convert worse. Pick niches, expand later.

Are you suggesting too much or too soon

No one will sign up without understanding their problem, your solution, and your pricing. And very few people pay corporate SaaS without first consulting their account manager. Be mindful of what you offer in your call to action. Too early in the page or purchase cycle, conversion won’t happen.
Write a list of what the visitor needs to know before registering. Make sure this important information comes before your first call to action.

You tell people what you can show them

Many pages used large blocks of text to explain something that could be more clearly and convincingly demonstrated in the form of a simple product snapshot, table, visualization, example, demonstration, illustration, or abstraction.
Show, don’t tell.

You ignore the doubts

When your prospective buyer visits your landing page, they will mentally ask questions. As the number of these unanswered questions increases, the chances of conversion decrease. Find out what those questions are and address them in the text or in the FAQ (frequently asked questions module).
Understand and resolve doubts through user testing, exit intent studies, or surveys.

You need to use better images

About 65% of the weblinks I tested used images from popular collections : photos, icons, and illustrations. While these images were workable and even occasionally relevant, they rarely conveyed the same meaning visually.
Work harder to find or create evocative images for the landing page.

You don’t know your stats

Even people who were extremely smart and knew the effectiveness of their landing page didn’t measure their entire funnel from marketing channel source to account signup. That is, that they may have had a fantastic landing page, but didn’t have a complete measured signup funnel.
Set up funnels in Google Analytics or other tools so that you can track conversions across the entire funnel, not just on the landing page.

You need to test the conversion rate regularly

Lastly. You may already have a high-conversion landing page, but without regular testing, you won’t know anything. No customer has ever had a regular testing schedule. At best, they last checked performance after relaunching their site a few months earlier.
A cycle of constant experimentation increases your customer knowledge and revenue. Keep testing.

What I learned about building a business

How to make money

  • …….. I got about £20, 000 from roasting and another £50, 000 in freelance marketing work from clients who found me through roasting.
  • I spent about £7, 000 on paid production advertising and about £3, 000 on tools and other business expenses.
  • I’ve recently added additional services – rewriting ad copy, rebuilding the landing page, and implementing conversion analytics. About 1 in 4 of my roasting clients ordered the second service.
  • About 1 in 6 of my clients offer me freelance marketing work.

Founders and conversion

  • The most common reason for orders : the funder got someone else’s opinion based on the feeling that their landing page isn’t working or their positioning isn’t right.
  • Most of the founders didn’t know about their conversion rate for their landing page in the moment, even if they had analytics.
  • Almost every funder has used Google Analytics to measure site analytics, most have set goals, but many haven’t actually analyzed their effectiveness.
  • 95% of the roasts were ordered by male funder.
  • SaaS, e-commerce, and B2C applications and tools were the most common types of lenders.


  • About 5 competitors have created almost identical products, clearly inspired by Roast My Landing Page.
  • Many have borrowed text, concepts, calls to action, pricing and marketing ideas directly from my Roast My Landing Pagewebsite.
  • Others were inspired but added their own motifs.
  • Some reached out and asked for my support, feedback, and advice.
  • It turned out that many of these projects closed in a matter of weeks.
  • I made a habit of not focusing on what they were doing and instead put my energy into growing my business. But to my friends I complained.

Collecting reviews

  • Before I fried the banding, I sent the client a form with 7 questions to answer so they could think and focus and give me context.
  • After the roast, I sent out an anonymous Typeform survey to all clients, asking them for one idea for improving the roast, which I used to refine my proposal.
  • About one in four filled out the survey, one in four emailed feedback, and I heard nothing more from one in two.
  • The survey included a 5-star rating. Of those who responded, I received 46 5-star ratings and 4 4-star ratings.
  • The survey also included questions about other areas of their business that founders wanted to fry – this led me to create other services.
  • About 25% of clients responded to my suggestions personally.
  • About 50% of the founders implemented the fixes detailed in the review. Most did so immediately or a few weeks (or even months) later.
  • Just recently, one in four customers asked me to make changes for them.
  • The more I worked in public, the more strangers helped me. People regularly sent in my ideas, corrections, even their own mini-fries several times a week.

How I attracted customers

  • The Roast My Landing Page landing page had various conversion issues.
  • I iterated the landing page about 20 times, tested 3 payment solutions, 4 price ranges and 5 shopping streams. I’m still testing it now.
  • My most effective marketing channels are word of mouth, participation in the Indie Hackers and Productize communities, Twittertargeting, Facebook targeting and then Google Ads, partnerships.
  • The best ROI on paid advertising was about 3.
  • The worst marketing channels were Quora, Reddit, email through Paved.com. Numerous tests on these platforms did not result in significant conversions.
  • The best way to spread the word-of-mouth radio was overperformance I promise to give every roast within 48 hours, but many customers got theirs within an hour of ordering. This made a dizzying impression on the faunders and caused quite a stir.

The tools I used

  • Tools like Loom are great for recording video to video. But when they fail to record an intermediate fry, it’s awful. When it happens 3 times in a row, you think about quitting your job, renting a small reliable car, packing up your stuff and moving to a remote farm away from everything and everyone you know.
  • I have used SPP for order management.
  • Typical form for customer feedback.
  • Google PageSpeed Insights to determine page load time.
  • Autopilot for my mailing list.
  • Ghost for this blog.
  • The analytics tools I used include Hotjar for records and heat maps, Heap for funnels, and Google Analytics for high-level reports and conversion rates.
  • GetSiteControl helps me collect leads through my weekly email newsletter.
  • Hotjar has helped me gather information about exit intentions.
  • I used Stripe and Paypal for payments.

The cost of my product

  • The roast started at £39, went up to £299 and is now £149.
  • I tested several different price points to determine what would make the most money.
  • Shooting roasts is mentally demanding and extremely tiring, so my limit was about 4 a day.
  • I switched to non time based pricing because I got proof that frying increases conversions.
  • I’m still struggling to get the right customers at the right price without stopping customer support.

The processes I have implemented

  • Routine is good, but I struggle with it : weekly reporting cycles and personal retrospectives helped keep me on track.
  • Yes, I did retrospectives, focusing on what I could improve from the previous week.
  • Reporting helped me focus, especially the public posting in my Twitter And detailing my milestones on Indie Hackers.
  • I also have a small community of like-minded people on Slack with whom I can share ideas, I am accountable to them.
  • The roasts themselves are relatively arbitrary in shape. Over time, however, a general structure developed that allowed the emphasis to be on the greatest possibilities.

What’s next?

Not only that. Roast My Landing Page is profitable, but running my business is also paying off in ways I didn’t expect.
I haven’t (yet) been sitting on a beach sipping margaritas, passively making mega bucks, but I enjoy giving actionable feedback at a fair price, seeing my clients’ business grow, networking and continuing to improve my understanding of funders, their business and conversion optimization.

So as not to lose my design finds in tons of bookmarks on my computer, I created a telegram channel Dryn Designe. All the cool and useful stuff (tutorials, articles, links to cool dudes’ accounts) that I find for myself, I’ll post there. Help yourself.

You may also like