# 1 Simple Principle To Set Yourself Up For Success With Paid Advertising

Recently I had someone in my Mastermind asking whether there were any resources for him to find out what’s an average cost per lead (CPL) or average ROI for his industry or something that shows average numbers across different industries.

While there may be something like that, I don’t think it’s a very intelligent to approach to paid advertising under most circumstances. Mainly because every situation is so different, there are so many variables at play, so many ways to build funnels, etc. Trying to generalize like that doesn’t make much sense in my opinion, especially in the beginning when you’re trying to get traction with your paid advertising campaigns.

Instead, what I teach my clients and coaching students is that they should always start by reverse engineering your specific situation or funnel with realistic assumptions, then test against those assumptions to see how close they are to the true actual data you get from a paid advertising campaign.

For example, let’s take this funnel for instance:

• Step 1: Email Optin
• Step 2: Front End Offer – \$50
• Step 3: Upsell – \$200
• Step 4: Downsell – \$100

Example 4 Step Funnel

From a funnel like that, what I would do is come up with some realistic assumptions for what I think would happen from cold traffic (people who have never heard of you and only landed on your website as a result of a paid ad), then work my way backwards.

Let’s assume we can get a 20% conversion on the landing page asking for an email optin.

Then, let’s assume we can get 5% of the people who optin to take the Front End Offer.

From there, let’s assume we can get 20% of the people who take the Front End Offer to take the Upsell.

And for the people that pass on the Upsell, we will assume we can get 10% of those people to take the Downsell.

What that would look like for every 1000 optins is this:

5% of 1000 equals 50 (# of Front End Offer sales), plus 20% of 50 equals 10 (# of Upsells taken), plus 10% of 40 (# of people who passed on Upsell) equals 4 (# of Downsells taken).

Then adding all that up would look like this:

[(50)(\$50)] + [(10)(\$200)] + [(4)(\$100)] = \$2500 + \$2000 + \$400 = \$4900

So based on those numbers, for every 1000 optins we get will yield \$4900 in revenue. Based on that, each optin is worth \$4.90 gross before excluding expenses.

Assuming there’s no other expenses, this would mean we could pay up to \$4.90 per optin. But we all know in order to run a business, you will have expenses.

So my next step is to deduct any business expenses, fulfillment costs, etc to deliver on the revenue being generated. Then re-adjust my value per optin based on my assumptions.

Let’s assume that our business costs are about 40% of our gross revenue. That would mean we would net \$2940 from \$4900 of gross revenue (40% of \$4900 is \$1960, then we need to subtract that from the \$4900).

At a net profit of \$2940, we can pay up to \$2.94 per email optin to break even. The next step now is to go and test our assumptions by setting up a paid advertising campaign to see if the numbers we have will hold true. If not, we can then re-adjust our target cost per lead (CPL) based on the actual real data we get from our paid advertising campaign.

Once we get actual data, it’s likely we will need to look at our optin rate on Step 1 and the conversion rates for all the other steps and go through this exercise again. Only the second time we do it, it will be based on actual data rather than our initial assumptions.

Ultimately, our job is to do everything possible to impact the numbers at each step in a positive way that will allow us to increase what we can pay to get someone to take that first step (in this case it would be our cost per lead). The higher that number becomes, the easier it gets to pay for traffic and ultimately scale a business to the next level.

So next time you go looking for answers in the wrong places, do your best to come up with your own realistic expectations. Then go and execute on something to test your assumptions.

Having your own data to work from is always going to be better than looking for generalizations that’s are highly unlikely to have any relevance to your specific situation.