How Not To Be Afraid Of PPC Optimization With One Basic Formula
I think this post will be very valuable for super busy PPC managers who are also doing a lot of other performance marketing or other marketing tasks.
Now, if you have your search PPC campaigns with generic keywords, or you run a metasearch campaign and you really don’t have the time to dive into each and every keyword or hotel you’re promoting — what do you do? You have to optimize, you know you have to. How are you going to implement as few as possible metrics on your campaigns to know where are you going to bid higher, where lower, and what campaigns you are going to pause?
The idea behind this formula is to make it about predicting what you’re going to earn from every click.
EEPC — Expected Earning Per Click — should be at least equal to the ECPC , the Effective Cost Per Click. When they are equal, we are reaching ROAS (Return on Ad Spend) of 0%, meaning that every dollar of clicks spent is returned back as 1$ of revenue.
You are probably now asking, how can we calculate the EEPC?
Well, here is the formula:
EEPC (Expected Earning Per Click= CVR (Conversion Rate) * EeCT (Earning on eCommerce Transaction)
EEPC = CVR * EeCT
Do you know how you should implement it on a Search PPC or a Social PPC or a Metasearch campaign?
Let’s go one by one:
For a Search PPC campaign:
Let’s say you have one campaign containing two keywords:
A) +Cheap +Hotel +Deals (Broad Match)
B) [Cheap Hotel Deals] (Exact Match)
Now here is the rest of the data that you’ll almost always have,
for A:
Avg. CPC (eCPC) = 3$
CVR (Conversion Rate) = 3%
Avg. EeCT (Average Earning* per eCommerce Transaction) = 90$
Note: I usually refer to “Earning” as the “Net Income” of the transaction. For others it may also mean “Total Revenue,” depending on the business model and the campaign’s objective.
Now you can calculate your EEPC:
EEPC = 0.03 * 90$ = 2.7$
So in this case we get:
2.7$ < 3$
Therefore: EEPC<eCPC
This means that for every click of the keywords, +Cheap +Hotel +Deals, we lose 0.3$ and have a ROAS of -10%.
What needs to happen in order to be at least at a ROAS of 0% ?
Three options:
1. Increase the cost of the product (increase Avg. EeCT)
2. Improve Conversion Rate (increase CVR)
3. Decrease the eCPC
For a PPC manager, usually the only way to reach this equilibrium would be by reducing the Max CPC to the level of eCPC, to reach not more than 2.7$. But it is also important to see the entire map.
Now let’s quickly analyze the second keyword, for performance B; here is the data:
Avg. CPC (eCPC) = 2.5$
CVR (Conversion Rate) = 5%
Avg. EeCT (Average Earning* per eCommerce Transaction) = 90$
EEPC = 0.05 * 90$ = 4.5$
4.5$>2.5$
Therefore:
EEPC > eCPC
That means that the ROAS of this keyword is 80%, and on every click of the keywords, [Cheap Hotel Deals], you will earn 2$.
Now you can speculate about strategy:
1. Maximize profit & ROAS by reducing Max CPC.
2. Increase sales by increasing Max CPC and gaining more traffic.
You are probably asking yourself: What if I don’t want to optimize towards a 0% ROAS but want to have a positive ROAS? Well this is pretty easy, all you need to do in this case is to multiple the left side of the inequality with 1+ROAS, just like that:
EEPC >= eCPC * (1+ROAS)
For example:
EEPC >= eCPC * (1+15%)
EEPC >=1.15eCPC
For Metasearch campaigns or any other type of CPC media campaigns, you will use the same formula. The trick is to know what you are optimizing with the formula. With search PPC, it has been easy because obviously we optimized keywords (in the same way you could optimize Ad Groups — it all depends on your business objective and your available time). So let’s take a look at other types of CPC media and see what to optimize:
Facebook PPC Optimization:
Here we should be careful and here is why. Say you advertise for an OTA or for a hotel chain and have several locations, and you create a separate campaign for each location with the “Website Conversion” campaign objective and a pricing model of “optimized for website conversions” (a correct campaign structure), then for each campaign you will create Ad Sets based on different audience targeting. And finally, you will reach the creative part itself and create the Ads. (Read my 7 tips for facebook ads that will help you avoid wasting advertising dollars)
So now we have three levels of campaign structure which are much more complicated and confusing than the Google Adwords Search PPC structure. The hack here is to understand how Facebook Ads’ optimization engine works: Up to the middle of Q1 of 2015, Facebook Ads would optimize its campaigns by the conversion pixel tracking on the AD level. Meaning that the AD is “learning” how to be served in order to drive the most conversions possible at the best price. In the middle of March 2015, Facebook changed the setting of its campaigns optimization engine and we were asked to choose a conversion pixel for optimization at the Ad Set creation level. It basically means that if we had one Campaign with one Ad Set and one Ad, and we deleted or paused the one ad we had and created a new one, it would start optimizing itself from scratch! Since the update, the new ad would have the privilege of exploiting the learning that the previous ad gained about the audience targeted on the Ad Set level.
So in this case, since the lowest level of optimization engine is implemented by Facebook on the Ads and the highest level is the Ad Set, you should optimize then by Ads.
Metasearch Campaign Optimization:
Let’s say you are buying some metasearch site (e.g., Kayak, Skyscanner) price comparison units. How can you quickly understand where you spent money on a low ROAS?
Here it’s actually easier than with Facebook PPC. If your pricing model is fixed CPC, you get the same formula:
EEPC = CVR * EeCT
And now, based on your inventory control abilities, you should calculate the conversion rate, either by destination, or by hotel name, hotel rating, or any other parameter, based, of course, on your business objectives.
But you probably ask: What if your pricing model is a fixed CPA or what if you are being charged a percentage commission on each transaction?
In this case, you should transform you marginal costs to an eCPC expected value. Here is how :
For a fixed CPA pricing model:
eCPC = Total CPA Costs / Total Click Amount
For a commission based pricing model:
eCPC = Total Commission Costs / Total Click Amount
Important note: if you decide to optimize by destination, you should transform the eCPC of every destination separately. The same goes for every other optimization base (Hotel name, Hotels rating, Country, etc.).
I hope you found this article valuable, and if so, please share with me your “easiest” platform to optimize a campaign on and why?
Thanks for reading.
Like this post? Please click ‘Recommend’ below, and share it with friends. It would mean a world to me.
Peace out!
