Module 5 - Looking longer term

What are you actually trying to achieve on YouTube in the long run?

Everything we do longer term with YouTube ads is to try and achieve what I would call a satisfactory target cost of acquisition. If you have followed the other parts of the course you will know you need to have made notes, designed your financial plan and set some targets to hit. Once you start hitting those costs of acquisition, we then look to lower that cost, or at least keep it within the same range.

As an example, if you are achieving a cost of acquisition of 28 and your target is 30, that probably says there is some headroom in there.

Why should you not be satisfied when Google hits your target?

Because Google will happily keep charging you your target price when there are cheaper conversions available. You are telling Google, I want a lead, I want a sale, I want this to happen for this cost. If you put into your plan at the start that your target is 30 dollars and you are happy with 30 dollars, and Google starts delivering at 28, yes, that is great, you are achieving what you wanted. But there is nothing to say there are not conversions in there at 10 dollars.

Google will happily continue to serve you conversions at your target. Our job is to test the limits of Google, find where there are cheaper conversions to be had, and lower that cost as quickly as possible within our budget.

It is like walking into a shop. There is an item on the shelf, you say right, I will buy that for 50 dollars, you pick it up and you pay for it, and the shopkeeper knows he or she would have sold it to you for 20 bucks. Google is no different. They are going to sell you conversions as expensively as they can.

That is not to say costs never creep up. If you start spending more and more it may creep up, and it probably will, depending on how many impressions are available, how big your market is, how attractive your product is, all of those things. But the sequence is: set up, get ads running, achieve some conversions, drive that to an acceptable level, then drive the cost down. So that when we do scale, we know we are not wasting budget.

How fast should you scale a YouTube campaign?

Slowly, by 10 to 15 per cent every three to five days. When you come to upping costs, Google will typically tell you that you are able to get more impressions, or that you are missing out on conversions, and then ask whether you want to add to your budget. I would say, and quite a lot of people in the ads industry would say, do not necessarily just click the button. Google will say you are currently spending 35 bucks a day, if you double it we will do X, if you treble it we will do Y.

I would much rather see you add 10 to 15 per cent to your budget on a regular basis. Remember Google is in business to take as much of your money as possible. They want to see you press that button.

It seems you achieve better long term results by upping your budget slowly and regularly. Add 10 or 15 per cent to a campaign maybe every three, four or five days. Do not do it every day. You need to let the campaign settle and grow a little. It is a way of saying to Google, that is okay, are you going to continue to give me what I am getting if I put a little bit more in? I will give you a little bit more, Google, but you give me what you are promising.

What patterns should you be watching for?

Patterns are everything in Google, so monitor weekly and bi-weekly and look at how campaigns behave over time. A campaign might start off here, grow a bit, grow a bit more, fall off a bit. You tweak something, it grows again, it does well, and then suddenly it drops off. If you see that across several campaigns you will be able to learn how Google is treating your account.

There is no one size fits all. Google does have patterns it works to, and there are different patterns within different niches, depending on how many competitors you are up against, whether there are enough impressions to go around, how favourable your bid is compared to somebody else, and how well your ad is engaged with. Like in traditional PPC, they are measuring whether people are getting a good experience when they click the ad, and how long they spend on site. All of that goes back into your ad rank, and the same applies to YouTube.

If another advertiser comes in, or your ad has been up for a while, Google might be saying, right, okay, come on guys, you have shown this ad to enough people, or you have shown it to the same group of people too often. Google does not want people seeing the same thing again and again. It wants fresh content. If I was publishing a magazine and the same ads were in the mag every week, people would get bored. Google is the same.

What do you do when campaigns start going stale?

Reinvigorate the account by launching new ads. Based on the size of your audiences and your budget, you will see a point at which campaigns drop off, and that is the moment to refresh.

If you do not want to sit there watching patterns, it is always a great idea to refresh all your campaigns on a monthly basis. That avoids the account getting too stale. Launch your first set of ads, let them run, see which ones are successful, duplicate those, make a split test, try a different audience, and do that monthly. You will have more success than if you let an ad run ad infinitum. Let it run for six, eight or ten weeks and Google starts to demote it just because it has been seen before.

So you will need to keep going in: refresh the content, run different ads, try different audiences, change the bids.

What happens when competitors move into your space?

Your placements start to underperform, and you may need to raise your bid. Bear in mind you are up against other people bidding for the same placements you are trying to use. They are not necessarily your competitors, and they might not even be in your industry. They are just trying to take that space and use it as their own.

If they are bidding more, you will see some of your placements starting to underperform. You can look and say, this month we got some great placements, how did they do this month? Right, they are not as good. You might want to up the bid again by 10 or 15 per cent, let it learn, and see if it achieves anything more. All of this is driven by not overspending. The most important thing here is not overspending.

Why should you not have all your eggs in one basket?

Because if a single big campaign falls off a cliff, the whole account goes with it. I have seen many advertisers come in with one or two campaigns, maybe two or three, and they think they are winners. The issue is that if one of those campaigns falls off a cliff, they go from one day getting loads of conversions to the next day getting very, very few. They panic, go back to square one, and then they are really trying to piece an account back together.

What you want to do is run multiple campaigns. I would rather you ran many smaller campaigns than one or two really big ones. That is a really good bit of advice that you need to stick to.

Why does lifetime value change what you can afford to pay?

Because if you know a customer comes back, you can afford to acquire them at a higher cost per acquisition than you first thought. You need to understand the process people go through in your business. If we can acquire a customer for a certain cost, what is the lifetime value of that customer? You may well find that at the front end or the top of the funnel it is worthwhile acquiring customers at a lower profit margin or a higher cost per acquisition than you might have expected.

That brings us on to remarketing. We talk about YouTube first in terms of awareness, and YouTube first in terms of lead generation and top of the funnel. But you need the back end remarketing loops running: the omni-channel approach, other social networks, not treating this as YouTube only. If you do not build that back end, you are fundamentally missing the majority of the point of why you are doing this.

If you get 100 people on your website and three of them take action, there are 97 people that you paid for that you should be remarketing to. You should be getting a good percentage of them to come back, because they have been warmed up to an extent and you paid for that. If you do not take advantage of that, you are wasting probably 20, 30 or 40 per cent of your ad budget, and you may as well not bother.

Can you speed the whole process up?

You can, but it will not be as efficient. I do get asked that a lot. Do not run before you can walk. There are people who come in wanting to set up hundreds and hundreds of pounds or dollars of campaigns on day one. I would much rather see you split that into smaller campaigns, run smaller budgets, test, learn, reiterate and scale properly, following the advice in the course.

There are no cut and paste formulas for success, because there are so many different industries out there. The way we are going to teach you is a system, and we want you to run it properly. Do everything you can to stay within these rules, and make notes as you go along.

What is the biggest mistake once things are working?

Getting lazy and thinking you have won. There are plenty of people out there who get a massive ROI and it is brilliant. Do not stop there. Always want the best, always want the lowest cost of acquisition, keep finding those better placements, keep learning more about your customers. This is the game. This is what we are there to do: learn and learn and learn more and test. You will hear me say test and learn all the way through the course.

It is all in the data. The more data you feed to Google, the more you are going to get back and the better the result. So do not stop thinking, do not stop adding new conversion events, do not stop running new campaigns, do not stop scaling where you can.

The bottom line is that everything will come out in the data you glean from these campaigns. You could become a full time data analyst on the back of the information you are going to get out of Google. There is so much data and so much opportunity to learn about your customers, where they live, what they do and what they like. Google can serve you as many of the right people as you want. You just have to feed Google the right data so the AI can work for you and not against you.

So when you are scaling, do it gradually, work the system, run multiple campaigns and do not have all your eggs in one basket. Otherwise you become lazy, think you have won, sit back on your laurels, and two months later you find you are not getting anywhere near the conversions you used to get.

Just on that note, you will often see that the first two, three or four weeks of a campaign can be fantastic and then it drops right off, and you wonder what has happened. It is because you are not looking at the patterns, or you have not got enough campaigns running across an audience to make a success of it.

Hopefully that has been useful, and I will see you in the next little piece.