There I was, with a coffee, looking the 5 split tests I have been running. And like so many people before me, I was appalled.Test #1: 7% conversionTest #2: 3% conversionTest #3: 6.8% conversionI kept hearing about these 30-40% conversion rates from the ‘gurus’. But was it really possible with cold traffic from Facebook? My cost per acquisition for a simple email address was close to $15 – an impossibly high number by what I’d been told is possible.I tested every single element on the landing page. The headline, button, opt in form, body copy, with/without testimonials, adding an image, and dozens of other things.To top it all off, it was in the investing niche and it was incredibly hard to get good numbers. People were VERY skeptical.Do a little hop, skip, and a jump, to the present and we’re hitting $2.50 CPAs, 40% conversion rates, and sub $1 clicks in the same niche!How to turn a failing Facebook campaign into a winner with two major tweaksA lot of blood, sweat, and tears went into this and I feel like I am giving away my secret sauce. However, if it wasn’t for all of you sharing your ideas I wouldn’t be where I am today.Rule #1 – don’t test the landing pageA lot of people will tell you to test your landing page. Every course I took, or blog post I read, was about testing the elements of your landing page. It simply isn’t true.“Don’t worry about the traffic” they’d say. “Focus on testing your landing page!”.Rubbish.Don’t get me wrong. Having a good offer on your landing page is crucial, but it only has to be ‘good’. It doesn’t have to be exceptional. The most important elements are the audience, and how congruent your offer is to that audience.I should really revise the title of rule #1, and call it “Test your audience, not your landing page”.And that’s exactly what we did with our offer. We tested 3 audiences and our front-end conversions jumped from 7% to 25% once we tested audience number 3.Check it out.BEFORE: 7.5% conversionAFTER: 24.8% conversionThat whole increase happened just by changing the audience! I am not going to give away details about this particular campaign, because it is ongoing and we don’t want to give our client’s competition any head starts.I will, however, explain how you can do it in every industry. And that brings me to rule no. #2 with Facebook ads:Rule #2 – overlay your target market with an existing audienceThis is my most powerful discovery yet. I am sure people much smarter than me got their first, but they certainly haven’t shared it with me.The formula is simple and the execution is even easier. First, find the ‘guru’ of the industry. This could be a magazine, a book, or a TV series.Find out what the most common question is, and then answer it. Here is an example:Download 7 [guru] Strategies to [benefit].So let’s say your target market is moms who want to go to college, you’d say something like “Download 7 [guru] Strategies To Get College Funding – Plus: How to Do It With Kids”So when it comes to the audience, simply pick ‘people who have shown an interest in [guru]’ AND ‘moms who have no higher education’.Marry those two audiences together in your offer. It is very important to give them something to download. If you don’t do that your conversion rate will suffer.By giving our audience a download we were able to increase our front end conversion from 21% to 40%. Check out the screen shot below:A whopping 40.2% conversion rate from cold traffic on an audience of 1 million.CPA went from $11.78 to just $2.41. Some of our competition are paying $5 per click, never mind $2.41 per subscriber.How you can do this todayThe easiest way to do it is to think of a T.V. series, and play off that credibility. Make sure you are not infringing on their trademark and get to work.Let’s take gardening for example.Download 8 ‘HGTV Mom Only’ Strategies To Renovate Your Garden On A Zero BudgetThen simply overlay the audience with people who like HGTV and are moms. Be sure to go to HGTV’s website and checkout their policy on using their content. Without a shadow of a doubt, you’d be able to get away with it if you give them full credit.From there you can start cultivating that email list, which is a whole other ball game.There are a couple of things you should be watching out for:Make sure the audience is passionate. Going after something that lacks a passionate following will be very apparent in the results. Drop it, and test another audience. I had to test 3 before I got to my perfect one and I suspect I got lucky. In the future I would have to go through way more audiences than that.When you don’t know where to start: If you don’t know what audience to go after you are in trouble. But there is an easy fix. Simply add a custom audience pixel to your website. Once you have 200-300 people on that audience, go to the Facebook ads manager and look at “Audience Insights”. It will tell you which pages your audience likes, their interests, and all sorts of commonalities. Start there and you’ll be fine.So there you have it. Simply find your audience ‘guru’ and then find a ‘benefit’ they all want, marry the two, and you have your masterpiece.Only once you get respectable conversion rates from your audience should you start testing your landing page. And even then, I would spend more time testing my offer than landing page elements anyways.I would spend all my time finding other audiences and commonalities among my target market. You can craft 3-4 different offers and run them periodically.What do you think? How have you improved your front-end conversions with Facebook traffic?