Module 3 - Finding the best keywords to use

How do you find keywords in Google Ads?

You find them in keyword planner, which sits under tools in the Google Ads interface. Here we are in the Google Ads interface again, in the s.media account, so let us talk about how we build lists of keywords that are going to work for us.

Within keyword planner you have the option to discover how many searches are being made for particular keywords on a monthly basis, for any search term that is going through Google. It is always a great idea to use the discover new keywords tool. Bear in mind that you can see Google flashing away. Sometimes it does not always do as it is told.

I have used in previous examples the idea that we do some work for a client who offers services to landlords. They do not want to cross over between the end user, people who are renting houses and might be typing in things to do with landlords, and the people they actually want, who are landlords or who work for a business that controls properties.

So let us try the things I talked about earlier. Evicting a tenant. Google will let you choose ten different keywords to put in here, so let us try landlord insurance, and let us just put landlord, because then it should give us a lot of other ideas within that. You can put a domain name in here. It is not essential, and I do not find it produces much better results than if you just search.

Why does geography change the numbers?

Because the volumes you are looking at are tied to the country you have selected. This list is based on the UK, so if I was to change that to the USA, where there are 299 million people, the numbers change. Probably not the greatest example, because I do not think they are called landlords in the US, but make sure you get the targeting in terms of geographic location correct for you.

At the top you can also change the date. If I wanted just the last month, it will change the numbers, and you can do the last twelve months. You can see a trend here too, although we have got a bit of a mixed bag at the moment, so it is not necessarily the greatest way to look at a trend, because we have got a mixture of keywords. Use this tool for trends when you are going a lot more targeted.

Over on the right hand side you have the opportunity to choose different parts within your search. At the moment there are 1,200 keywords displayed here, all of which, or a proportion of which, are not going to be relevant and will most definitely produce people we do not want.

What is the 51/49 rule?

The 51/49 rule says that if a keyword is 50/50, it is a no, because you are unsure. We need keywords we are sure are right. We need them to be in the 51, definitely sure that they are going to win over the other half.

A good example of that is something I said to one of my clients recently. He is an airline business and he wanted to target people 25 to 45. There is a massive search volume within those two age ranges, 25 to 35 and 35 to 45. I said to him, if you were going to fill your plane up with people 25 to 35 or 35 to 45, what would you prefer, who would spend more? If you got a hundred phone calls, fifty from 25 to 35 and fifty from 35 to 45, would you get more bookings from one group? He said, absolutely, I would get loads more bookings from the 35 to 45 group.

So what we then want to do is make sure we spread the ad budget across that group of people, not the 25 to 35. Yes, there are orders in that section, but in a 51/49 scenario the swing has gone straight to the older age group.

I know we are talking about the difference between age groups and keywords here, but in terms of keywords it is the same scenario. If you had a hundred keywords and the top twenty produced nearly all of your business, why run traffic to the others? You are just wasting budget, when you could have taken that extra bit of budget and stuck it into the keywords you know are more productive.

Sometimes you will have to go with your gut. But there will be choices over particular keywords where there is not really a choice, because you know that x keyword is better than y keyword, and in 51/49 you keep x and you get rid of y. You can always build separate audiences to test, but work from the top down. Keep the best and bin the rest.

How do you spot the keywords that are obviously wrong?

You read down the list and ask who would actually type that. If you are not used to Google Ads, you can always expand all of these columns. Looking at the most obvious mistakes here, Google has brought up 30-day notice to vacate. Absolutely, that is going to be tenants searching. 30-day notice to landlord, again, that is tenants.

So what you want to do is look down this list and decide which keywords only landlords are going to be typing in, exactly as you are going to do in your own business once you have made lots of different searches. You will get to a feel, and it is a feel. Once you start reading what everybody is typing in, you will understand what the best keywords are and what is more likely to produce a result for you.

What does a high top of page bid tell you?

A high bid usually means the keyword has both volume and a lot of intent. Another little tip here: go to top of page bid and sort by price, and the keywords with the highest bids come to the top.

Straight in there we have landlord insurance policy quote. No tenant is typing that in, well, we would hope not. Cheap landlord insurance quote. Landlord property insurance quote. When I ran this for the first time, and I went into this client blind, I was not expecting to end up with insurance within my search, because the client does not sell insurance. The client sells eviction services. They only work for landlords and they help those landlords evict their tenants. So we do not want the tenants. We only want to show the ads to the landlords.

Once you start looking at these lists you can absolutely start to understand why certain terms have high top of the page bids. When somebody is prepared to bid a lot and pay a lot, it means it is pretty highly focused. They know that if they get someone on the page they have got a good chance of conversion.

Why are YouTube ads cheaper than bidding on these terms?

Because so few people are using YouTube ads properly, you can reach the same searcher for pennies instead of pounds. If somebody is paying fifteen, twenty, thirty pounds or more for a click, this is where YouTube ads are your friend. I would not want to be paying ten, fifteen, twenty or thirty a click when I can show your YouTube ad to a potential client for one or two pence, or one or two cents, maybe a little bit more than that depending on the channels you are going to bid into.

If your keywords are really relevant and your content is great, so your video is good quality, you know that the intent is there. Somebody searched something really specific. I am not saying tighten your campaigns right up, but it is obvious that some of these keywords are spot on, absolutely perfect for this client. We would obviously want to show those people those videos.

If we show that video to the person who searched those very expensive, very targeted keywords, they get to look at us for thirty seconds before we pay anything. If they are still engaged and they then click through, the chances are we have got a lot more chance than a standard PPC ad. All the guys bidding on these terms are paying eight, ten, fifteen, twenty pounds or dollars a click. We could have got a thousand views for that. If we only get one per cent click through, we have still got ten times more people onto the landing page using a YouTube ad than we would have done if we had just bid against three or four other people at the top, especially in keywords that are particularly competitive like this.

The chances are people will click on two or three of those ads. Are they in the mood to buy at that point? Maybe they are just doing some research. If they are doing research, let us show them a video instead. Let them learn about our business, let them go away, let them go and search these terms and go and visit competitors' websites. There are ways to target people who have been on your competitors' websites, which we will learn about later in the course.

In this scenario, if our competitors are bidding ten or fifteen a click on normal PPC, they are letting people click on their ads, they are spending all the money, and they are warming the traffic up for us. All your competitors are doing the warm-up, the people are doing the research, and we show them the video for one, two, three or four cents for a view. Then people come through already minded to buy.

The only other consideration is that the video matches the landing page experience, so take that into account as well. Hopefully you can see that doing it this way is a lot more targeted and less costly to work with. That is why the conversion percentages and ROI from YouTube ads beat traditional PPC hands down.

How do you refine the keyword list?

You use the menu on the right hand side to strip out the phrases and brands you do not want. These are particular insurers. We might not want farmers. We might keep insurance companies. We might start taking out particular brands. I know that British Gas is not what I want. That would be a definite end user search, as would home service. Let us get rid of the ones that are clearly not us. Co-op is a UK insurer. We could go through this whole list: insurance, renters insurance, business insurance. Let us get rid of renters. Property, I do not know on this one just yet.

Once you start unticking, it is updating your list. So now that I have done that, there are still 960 in here.

What I would then highly suggest you do is go and download it, because you cannot just sit and look. You become snow blind. You can see so many keywords here that you get to the point where you are just going yes, no, maybe, yes, and you can start making bad decisions. In the top right there is a download button. I always download it to a Google Sheet. In this case it is going to give me 900 and something, and I will probably try to refine that down to 50 or 100 keywords.

While I am looking through that list I am going to find other keywords, little threads and little signals of other potential keywords that I might like. Then I will bring those keywords back into this tool and repeat the exercise again. It is going to take you a couple of hours or three hours to get your head around it, but it is probably one of the better two or three hours you will spend, because you will understand what your potential audience are searching for.

How many keywords do you actually need?

You probably want 50 or 100 keywords, but they have to carry enough search volume for your budget. Do not just have a bunch of keywords where you have not got enough volume. You do need to have volume.

That might mean you have lots of keywords that are highly focused, and you might find some with decent volume. Depending on your budget, you need to tailor the amount of searches to the budget, because if you are trying to spend thousands of dollars a day and you have only got a couple of thousand searches a month, clearly it is not going to work. You have to look at other avenues such as topics, and other types of in-market or affinity audiences, which we will talk about later in the course.

The lesson and the takeaway here: use that keyword tool, and open your mind up to other keywords that people might be searching. It does not necessarily have to be exactly about your product.

When should you split keywords into separate campaigns?

Split them when you spot two or three separate threads that you do not want crossing over. Go backwards and forwards between the keyword tool, refine it, take it out into Google Sheets, pull out the best ones, bring some more in, and eventually you will end up with a decent list.

An example of that is a client we have got with a medical device. That device works for circulation, but it also works for disease or conditions. So their keywords relate to pain and cramps and all this sort of thing, but they also relate to diabetes and poor circulation. Because there is a separation between those two bunches of keywords, I have created two different campaigns, because I do not want crossover.

Within the example we are given here, we might have landlord insurance, we might have landlord services, we might have eviction services, and we might want to split those out depending on how many keywords we have got in each of the groups.

Hopefully that has been helpful. Test, reiterate, bring those keywords out, find the best ones, and use that 51/49 rule. If there is a keyword you are not sure about, you can always ask the client or ask someone else. Would you expect the person searching this to fall into the could be a customer category, or is it 51 the other way, and they are more likely to not be the right type of fit? Hopefully that is useful, and see you in the next lesson.