Module 1 - Costs and conversions

How quickly should you expect conversions from YouTube ads?

It depends on a lot of factors, including your audiences, how well your product is received, and whether your landing page converts. Before you look at what your costs and conversions are going to be, look at your whole funnel first. Make sure you're ads ready, and make sure your offer is right, maybe you've already tested it, so that when you press the button and start spending budget, it's actually going to convert into sales or inquiries.

Why shouldn't your first conversion goal be a sale?

Because your first set of conversions is more likely to be getting people onto your site or landing page, which is step one, not the end goal. Your ultimate goal is a sale or a new customer, but the first job of a campaign is to prove that the audience you're serving your ad to is responding to it. If your site converts at 3 percent and you get a thousand visitors, that means the awareness and engagement generated at the top of the funnel has led to 30 people taking some kind of action, and what that action is depends heavily on your offer and how attractive it is.

It's not a case of put an ad up and get a sale straight away. You need to think about taking people on a journey: educating them, generating awareness, and remarketing to them. I see people forgetting the remarketing piece all the time. If someone spends several minutes on your site without purchasing, they're probably interested, so you want them in a remarketing audience, to test and understand whether you can generate an acceptable cost of acquisition from that audience. If you can't, you drop that audience later, but don't rely on one perfect customer seeing an ad and buying straight away as your whole strategy, because if you rely on that, you're going to fail.

What did a real client's conversion actions look like?

On a live customer account for a screwdriver manufacturer that had been running for about three or four months, we tracked a whole range of conversion actions rather than just a sale: contact forms, dealer search, two different click events for finding a dealer, join newsletter, register my tool, request info, catalogue download, an email click, a separate newsletter sign up from another part of the site, phone number clicks, and a second catalogue download. When the campaign started, the aim was simply to get people onto the site and click the request info button. As people started requesting info, we added further conversion actions to teach Google what we wanted people to do on the site, which is set up using Google Tag Manager.

At the time, this customer wasn't even tracking a sale yet, because they were just about to launch their e-commerce site. There were two types of campaign running: a display remarketing campaign and a video campaign, and there was a substantial difference in cost of conversion between the two.

Why is remarketing so important for cost per conversion?

Because on its own, a video campaign can look expensive, but layering remarketing on top can make conversions dramatically cheaper. On this account, the video campaign had around 400,000 impressions and had spent about two thirds of the budget of the display campaign, but it had a much higher cost per thousand impressions and a higher cost per conversion. Running the video campaign alone, without the backup of remarketing, we'd have been getting conversions at around 25 pounds. Once we started remarketing to the people who had watched the video or been on the website, those conversions became around 25 times cheaper, getting conversions at under a pound. That's when the magic happens: combining all the marketing efforts and joining the funnel together properly.

How much budget do you need to start testing?

You can probably start testing with a budget of around 10 to 15 dollars a day. Google gives you the opportunity to scale nicely once you're ready to expand. Once you're scaling, you need to monitor cost carefully, because Google will spend the money you give it access to if you let it. For example, if you set a 50 dollar a day budget and five days later you've only actually spent 75 dollars, you might think you're being conservative. What can happen is that Google builds up a bank of unspent budget, so with 250 dollars allocated over five days and only 75 spent, it could jump in and spend 150 dollars the following day, even though your set daily budget is 50. The real takeaway from this part of the module is that there are ways to set up multiple different types of conversions, not just a single sale.