- What We Do
- Who We Help
- Case Studies
- About Us
Here at EmberTribe, we know the importance of testing. Especially when it comes to paid search strategies.
Looking for ways to make your paid search campaigns more efficient and profitable?
Check out this week's quick tip video, where we'll break down strategies to help you make the most of your time.
When you're just starting out with paid traffic, it's difficult to know how you should spend your time.
Today we want to show you a very simple framework to help you organize your day and understand what you need to work on, so that you can shorten the time it takes to get results and make your campaigns very profitable.
Okay, so everything we do boils down to testing.
And that's one of the beauties of paid traffic is that you get very fast feedback loops. You know what's gonna work and what's not gonna work very quickly for your business because people are responding to those ads.
We like to break our testing down into two basic categories. One is horizontal testing, and the other is vertical testing.
Let's start with the horizontal testing. What is that?
Well, when you're working on a platform like Facebook, or even Google Display Network, what we're talking about here is really testing across different audiences.
You may have some target buyer personas. Now, if one target buyer persona is maybe a decision maker, maybe we're testing the people who are researching the solutions, so that's a totally different buyer persona.
But if you're a business-to-consumer type of product, maybe we're just testing moms versus dads, that would be a great example. So all people who are prospective buyers of your product or service, but we're testing horizontally, or in a more of a wide approach, to start seeing where we can get traction.
The other piece of testing horizontally, is testing your messaging.
Different value propositions, different ways that you could present your product or service as beneficial. We want to test that across different audiences to see where people are responding.
And the idea here is to cast a wider net to figure out and to spot where you're going to get traction.
Now to get specific, maybe on a platform like Facebook, we're talking about dicing up your audiences between custom audiences and look-alike audiences—going after different interest targets, or different B2B types of targeting, job titles, seniority, or company size.
There's a lot of different ways to dice this up and what we like to do at the beginning of any engagement is to figure out what are the potential options here for us to test. So let's start with horizontal testing, low budget across a bunch of different potential or prospective segments of the audience.
Now the second piece of this is the vertical testing. Vertical testing is all about spotting what's working but then drilling down to both optimize and scale.
We want to find something that works, and then we wanna bring it to the next level. We want to give it more budget, we want to make sure that this is producing more leads or sales for you every day that it's running.
To do vertical types of testing, you want to think about zooming in. We find maybe a campaign that's performing well, and we want to zoom in and figure out what could work better.
So we look at things like click-through rate and we see, okay, which ads are getting the best click-through rate? And then we'll drill down even more into the cost of those clicks. Where is there room to improve? And then we'll look at how these landing pages are converting for this particular campaign. Is there room to improve there?
And you start to see, as you zoom in, where there's blocks in the process of getting people to that buying decision. You want to spot where there's blocks and then come up with tests to unblock those pieces.
If you're having trouble getting click-throughs, maybe you need to try a new ad creative or a new headline. If you're having trouble converting people on the landing page, maybe you need to be more clear with how people can take the next step.
But as you zoom in, uncover the blocks, unblock those blocks, then you start scaling your budget.
Now scaling budgets is a science in and of itself depending on the platform that you're on.
So for Facebook, Facebook's very finicky with how much budget they want to see increase in a certain day or in a certain time period, so you need to do that incrementally.
We like to look at usually 20 to 30% increases over the course of every couple days.
But with search, sometimes you can scale things immediately. If you start with a small budget and you find a certain type of keyword or keyword phrase that's converting very well for you, you can scale those budgets a lot faster without decaying the results or increasing the cost per lead.
With vertical testing, zoom in on the different key performance indicators, KPIs, that contribute to a low cost per lead.
You know after you've spotted something that's working, zoom in on some of those metrics that matter the most. And then figure out how you can optimize to make those better.
Of course, the other piece of this is scaling the budgets. So then raise your budgets incrementally, understand the platform, and the environment that you're in and where it might be more sensitive to sharp increases in budget, and start to build out those campaigns higher, and higher volumes.
Now the whole idea is for these two types of testing to work in concert, to work together. You want to continually be testing horizontally with lower budgets to discover new points of traction. And this'll give you a lot of mileage.
There's a lot to do here, so you wanna run that in parallel. And when you find something that's working double down, start to prune it, scale it, and add budget.
Over time those campaigns that are working, that you've scaled vertically, they're going to peter out, they're going to probably reach their max. Or maybe competitors come in and start raising the cost of running those campaigns.
You always want to be finding new pockets of growth through your horizontal testing, and then when you find one, scale it up through vertical types of testing.