This is an internal memo recently created by a partner to our AdWords team regarding AdWords and pay-per-click advertising.
We’re issuing this as a blog as it is of general interest to anyone who wants to improve ROI with their ad budget no matter the platform being used, and it provides some insight into Worry Free’s philosophy regarding managing ads.
The purpose of managing a PPC campaign for a client is to generate as many leads (or direct sales) for the client at as low an ad spend per sale as possible.
Another way of saying this is that we create viable ad campaigns. A viable campaign is one that is sustainable into the foreseeable future. The client is getting good value for their ad spend and knows it.
Note that Google’s purpose is to get the client to spend money on ads while providing the best possible user experience for searchers. They do this well and this is what drives their soaring stock price.
However, this means that Google is interested in slightly different metrics than we are.
Therefore, do not expect that Google suggestions will automatically be in alignment with our purpose, but may in fact run contrary to it.
The most important metric
One of the first and most important things to understand about a new client is –
What is an acceptable cost per lead (or sale)?
For some clients, this will be quite high. For example, a commercial solar panel installation can generate hundreds of thousands of dollars for a client and tens of thousands of profit.
Therefore, you have a lot more leeway in getting that lead. Of course, the length of the sales funnel may be long – but in the above case, if you’re getting a lead for $250-500, that’s pretty good ROI, so long as the leads are closing and becoming sales.
On the other hand, if a lead is only worth $100 total in sales, representing $20 of profit for the client, in order to make the campaign viable for the client, you’ve got to be getting sales for roughly $4-6 in clicks – pretty tough, which is why we rarely recommend that these types of clients do PPC through Google. Facebook can be effective for this type of sale, which is usually B2C, but even then, the same basic rules apply.
The first goal of any campaign is to rapidly split test the components of the campaign until you have something that produces demonstrable leads and sales at a viable cost per lead. My recommendation is to start by promoting the item or service (if there are several) that you feel gives the greatest leeway to burn clicks (because of the high value of the sale) and focus in on that until you establish your first viable campaign.
From there, expand the campaign, looking for other similar avenues to produce the same result.
No client will stick around for long if we aren’t able to get that product of viable cost per sale or lead – so when a month goes by with an astronomical cost per lead, it’s time to make a rapid and effective change to the account.
We are very much less concerned with generating clicks and fully using a budget than we are with getting leads at a viable cost.
Clients can get antsy if they see their full budget isn’t used for the first month or two, but if they are genned into our methodology, and if we demonstrate that the ROI we’re getting makes it worthwhile to continue, they’ll be fine. It’s better than spending a bunch of money for no result – which is what many of our competitors do.
Google isn’t perfect, and neither is your audience…
If your campaign is getting loads of clicks and no leads, it is a virtual guarantee that:
a) Ads are triggering for irrelevant searches.
This can be caused by:
- Homonymic searches (Person searches for “bat,” meaning a small furry mammal that they want to build a little house for, but Google shows them your ad for a baseball bat)
- Searches for products that have wide use in vastly different fields. (Person needs batting for audio insulation, is shown a batting cage)
- Ads are triggering on broad search terms and burning money (You have special autographed baseball bat memorabilia, ads are being clicked by people looking for bats to use in their backyard game.)
- Informational searches are triggering ads for products or services. (“Who was the first baseball player to use a wooden bat?” Google pops up your autographed bat collection)
- The quality of your item puts it in a different class than the majority of searches. (You are selling 24kt gold bats, people just want to play wiffle ball)
- Ads are triggering for related items that are in a different category (You have bat earrings, ads are displayed for sporting equipment searches)
b) People are clicking them anyway.
You’d think people would see a picture of a baseball bat and realize it wasn’t a small furry mammal, but this is dead wrong. People just click anyway. Then bounce, wasting the client’s money.
But it’s still your fault
Google is just a collection of negatively charged particles flying through silicon at the speed of light.
Most of the people using it aren’t even aware that the top search results for their queries are ads, and a scary number of people in our society are functionally illiterate, search while driving down the highway, are drunk or high, or some combination of all of the above.
So even though Google isn’t perfect and your audience is easily distracted, it’s still the client’s cash that you’re spending, and you need to get results in spite of these issues.
The first thing to know is that no amount of landing page changes will remedy the above issues. Changing a landing page when you’ve got the wrong people clicking your ads is like using a space heater in a room where the AC is too cold. Check the above factors thoroughly before moving onto landing page changes, or before assuming that something is wrong with the client’s offering.
To remedy, your search terms must be very specific. Start with small ad groups of closely related terms until you are quite sure that at least you’ve found the audience you’re looking for, and they’re finding you. Google gives you a number of tools to see whether you are on target. You can look at the demographics of the people clicking your ads, the actual search terms that trigger ads, and look at what pages they visit after they hit your site.
You can also look to see what exact sites or apps people are visiting just before they click your ad. If you see suspiciously high clicks coming from a certain app or youtube channel that aren’t resulting in sales, check it out by clicking the link that Google provides. There could be an issue as described above or the app itself could be set up in a funky way that makes it hard for people to take action without clicking an ad they don’t really care about. You can exclude such channels and apps individually or by category.
You may need to ditch some high CTR terms if people just aren’t getting it and you aren’t getting sales for a volume of clicks. Better to get a smaller volume of clicks with better producing keywords than more clicks with less result.
Another strategy is to prequalify your clicks. If the price of your item is higher because it is specialized use, name the price in the ad itself.
If your item or service is specialized to a unique industry, use industry nomenclature in the ad wording.
If using remarketing or display ads, make sure the image clearly conveys the use of the product AND the headline prequalifies clicks as above.
This is not a complete list of everything that can go wrong with a PPC campaign – just a list of commonly found items. If the issue isn’t listed above, it doesn’t mean there isn’t an issue – it just means we haven’t found it yet and need to keep looking.
The key thing to know is that we are driving for a viable (meaning sustainable into the future) cost per lead, and that we lose clients if we don’t achieve this rapidly. Anything less than a viable campaign that is producing profit for our clients first and foremost must be recognized as something to be remedied, then action must be taken to fix it.
This means hard work up front, but once things are working, you’ve created an automatic money machine for the client, and you will have created a worry free ad campaign for the client.