**UPDATED 2018**

I'm often working on building systems to generate leads for businesses, which is a time consuming and tedious task.

It's more often about testing multiple variations of the system, the copy, the ad, the means of delivery etc.

In one of the most recent systems I built out it was a matter of getting leads to a high end product ($1000+) and converting those leads to a free consultation to then sell the package.

The first step in creating this is understanding the specific industry you're creating it for. This takes time to research the local area, understand the business and the funnel and interview the best clients in that industry to identify pain points.

Once we've finished this discovery phase it's time to create several variations of the ad copy and pictures to test initially. I prefer to use tools like AdEspresso to test these variations. At this point we understand who we are targeting and where we are targeting specifically.

The next phase is figuring out where to run the ad in a market with a multitude of different options. I most often choose Facebook as it provides the best and most advanced capabilities and data to target and run ads.

The specific demographic I was building this lead generation system for was looking for people ages 35+, who make a higher level income and more often than not are family based and grounded in their careers. Facebook was the best option for this demographic.

Now that I had my variations, my demographics and my targeting in order it was time to test different variations of the ad using different means. I tested the ads using:

Facebook Lead Ads
Clicks to website
Landing Page Views
Conversions on website

From this initial testing I found (and this is generally the case) that:

Facebook lead ads don't work well and are expensive to run. Sometimes people give the wrong email and phone number due to it auto populating in the form. I had to create a check box to have people certify that their info was correct.
Clicks to website performed very well, was inexpensive and converted the best
Landing page views outperformed clicks to website and encouraged people to actually look at the landing page
Conversions were good, but we didn't have the volume for it to work as effectively in this specific industry.

In summary landing page views and clicks to website works very well for the small business that is focused on bringing in leads, but is not bringing in in excess of 100+ leads a day.

If we did have this higher volume (50+ leads a day) I suspect the conversions would have performed significantly better as Facebook would work to target more people like the ones converting.

Next I found that (and this is probably obvious) mobile wins in terms of ad placement.

Audience network gets clicks, but doesn't convert well.

Instagram itself doesn't always doesn't convert (stories are actually starting to yield some impressive results) well just yet in their ads. Surprisingly enough desktop and right column performed reasonably well. I'd still recommend starting with a mobile approach.

Finally after testing 48 different variations of ad copy, pictures, headlines, call to action buttons and demographic targeting in several different testing rounds I found that the most important pieces were the ad copy, the landing page and the headline.

The key pieces to the ad copy and headline were:

Calling out your intended demographic
Including key pain points that you found during the discovery phase
Humor
The headline must be simple and state what you're doing. It reinforces the ad copy, but does not repeat it.

The key pieces to the landing page were:

Short sentences with bullet points and bold headings
Use scarcity to encourage people to take action
Pop up forms always outperformed forms on a page
Exit intent pop ups improved conversions
Power words like "imagine", "secrets" etc
Social proof from real people

The takeaway form this is that if you want to build something that coverts you have to put in a ton of time testing the different aspects and variations and the different nuances of that specific niche.

One thing to note with all of this is that what works in one market may not work in another. This is why it's so incredibly important to test and to initiate a thorough discovery phase prior to starting any campaign.