Engineering For Profit

Profitable campaigns must be engineered

Campaigns require a lot of engineering to go from tadpoles wasting your time, to fishies who can get some clicks, to the sharks who hunt for your profit relentlessly and successfully. (Ok. Not a great analogy. But it probably works well enough).

Each time you try a new campaign, you will have to be ready to modify that campaign so that it yields profit. This is how you create profit. And some campaigns will make it, while some will fail or need to be reengineered.

The point here is this: You're going to have to create a lot of very focused (one user need, one market, one landing page) campaigns, make lots of modifications, and try them out in order to find out where the profit can be harvested from within your SEM marketplace. It will help if you organize the steps along the way and develop a structured method for getting your campaigns through it - or dropping them before they cost you much money.

First: Define your "gauntlet" of campaign milestones

To become a machine that is cranking out profitable campaigns, and then scaling them as far as that profit can be scaled, you need a well defined process. A process for creating and managing (i.e., changing) campaigns so that they show their profitability - and you can decide what to do with them in a fast and systematic fashion. As we stated (joked?) earlier: Nobody has ever created a search campaign that was profitable the first time it ran. It always requires "engineering".

You certainly can (and probably should) define your own "milestones" for campaigns on their journey to maximum profitability. Here are our recommendations. This will help you create profit more quickly and be ready to scale that profit. Only the first two of these really don't call for any judgment.

1. Successfully get your campaign to run (correct all errors and policy violations)
2. Get your campaign to create a volume of impressions (hundreds are minimum)
3. Achieve a volume of clicks (dozens per day) - with an acceptable bounce rate (under 50%)
4. Produce a minimum of ten conversions (per day or week)
5. Produce a minimum of ten conversions (per day or week) THAT CREATE PROFIT
6. Create more profitable conversions across locations, devices, schedule, keywords

What are these "milestones"?

They are well defined standards that you use to identify each campaign you are working on. They cut out unnecessary discussions about where a campaign is at and what to do next. Suppose you have a campaign at stage 4 (just got a set of conversions). Everyone knows the job now is to try to make that set of conversions profitable. Of course there will be excitement at how to do that. And those are tough judgments. But you don't have to have long discussions about "where it's at".

Another example: Let's talk about a campaign at stage 3 (getting clicks but a big bounce rate). It is pointless to proceed with "profit engineering" if your bounce rate is 80%. So your campaign is at "stage 3" until you can get that down to say, 40%. If you can't, you probably don't want to put any more time into trying to get conversions with that campaign.

While these six milestones may seem tedious, they not not need to be. In fact, they can provide you with a very systematic method of getting to profit while minimizing the expense from those campaigns that do not turn out to generate profit or take a long time to do so. No point in "scaling" anything until you get those conversions profitable. Even a small number of them.

Small campaign changes can yield large changes in the results

When we talk about changing anything in a campaign (in an attempt to improve performance), it always reminds us of golf (a horrible game played by masochists - in case you're not familiar). And any golfer who’s suddenly decided to change some “small” part of their swing (usually mid-backswing) can attest to the shockingly large effect that can have on the results of their swing. (Always carry a few extra golf balls). 

Golf jokes aside, that’s a lot like changing things in search marketing campaigns. Seemingly small changes can yield LARGE differences in your results. The challenge is to try lots of combinations and learn what works.  Here's some examples.

• An ad that doesn’t work at 9am might work perfectly at 9pm.
• An ad that never gets clicked on a desktop might yield a large number of clicks on mobile.
• An ad that works well in Nashville might not work at all in Memphis.

Try lots of combinations

Please do not think that the people looking for your business at 9am behave the same as the ones who look at 9pm. Please don’t think that the word “value” means the same thing to your customer as the word “cheap” (even though Google might treat them as "close variants"). 

Maybe your customers want you to show the price. Maybe they don’t.  Maybe your customers like to do mobile searches at 3am? On the weekend? Maybe they type in their zip code with their search. Maybe your website is too slow on the phone. Maybe your people don’t answer the calls from mobile visitors fast enough. There are a lot of potential combinations of elements you can put in a Google Ads search campaign and you really have to try them out to see what will work.

“Settings” are a critical element

There is a tendency to run all devices or all hours together. You need to resist that. Try very specific combinations. We recommend never mixing mobile and desktop. Mobile searches are different from desktop searches. Some things are bought on mobile devices. Some aren’t. Some businesses want phone calls. Some do not benefit from phone calls. Don’t assume that mobile and desktop searches behave the same.

The point is, you are much better off starting with specific devices, hours, and bids and seeing how well they work. If you want to run both mobile and desktop, create separate campaigns (something CampaignBuilder makes very easy).

Advanced Ads features don’t help – before profitability

Google Ads has dozens of advanced features like: sitelinks, automated bidding, audiences, and remarketing lists can improve performance. These features can enhance performance for campaigns that are already performing well. Here's what they can do if you apply them at the wrong stage of campaign development.

• If you create four different ads and want to understand which one really works well, it isn’t going to help to have Google “automatically” turn off the ads that get fewer clicks before you can see which ones will actually get more customers (conversions).
• If you have ten keywords in your campaign and you want to understand which ones really work, it isn’t going to help If you are using “automated bidding” to dynamically change the bidding before you can see which ones will actually get more customers (conversions).
• If you are trying to understand which keyword/ad combinations work well, it isn’t going to help if you use sitelinks to take your visitors to one of 4 different landing pages.