← The Digital Marketing Podcast
The Digital Marketing Podcast
The Rise and Growth of Direct Traffic
00:18:38
Have you found the direct to website traffic reported in your Analytics has grown and grown in recent months? Daniel and Ciaran discuss the growing trend in direct traffic growth being reported within Analytics and the reasons behind it. Daniel shares...
Loading summary
Transcript35 lines
- [00:01]
A
Welcome to the Digital marketing Podcast brought.
- [00:03]
B
To you by targetinternet.com hello and welcome back to the Digital marketing Podcast.
- [00:18]
A
My name is Kieran Rogers.
- [00:19]
C
And I'm Daniel Rawls.
- [00:20]
B
And today, Daniel, we are going to be talking about the rise and growth in direct traffic.
- [00:27]
C
Yes, we are. So if you're looking at your analytics at all, you may have noticed or you may not that your direct traffic has gone up. Now, just to. Just to clarify what direct traffic is, direct traffic is supposedly when somebody comes through to your website and they've typed in your website address directly into their browser and therefore have come straight through to your website. That's how most people consider it. What direct traffic really is is basically traffic where the analytics package has got no idea where you've come from. And that is a number of different places, but it's kind of growing and making things slightly confusing. So if I tell a bit of a story, it might help shape things. So we started doing some display campaigns. So we're doing some display advertising and we saw lots of increased traffic from display, as you would hope. But we also saw a massive increase in traffic coming through directly through to the website. Now, my instant thought on that was that we're not actually identifying this traffic as being direct traffic, so as being, sorry, display traffic. So we need to do something gremlin.
- [01:29]
B
In your tracking code.
- [01:31]
C
Yeah. So we thought that and we added additional tracking code. So we added tracking code onto the links that were coming to the website and we tagged them directly as display traffic. So if you're not sure how to do that, just search URL builder into Google and you'll find a way of doing that. But that didn't solve the entire problem. So we were seeing some more traffic, was actually being marked as being. As being this display traffic, but not all of it was. So we dug into this a little bit deeper.
- [01:56]
B
So you do this at this point I just go, just because I'm awesome is awesome.
- [02:01]
C
And it just. People are just coming to our website.
- [02:03]
B
Throwing themselves at the site.
- [02:05]
C
Yeah. Unfortunately, I wasn't that confident about what we were doing at this particular point. So what we found was a couple of things. First of all, when you're doing display advertising and that display advertising is going into app, when somebody clicks on a link in an app, more often than not, that will show up as direct traffic because you just aren't showing where that's coming from unless you add tracking code into that as well. Now, I should say, as well, the quality of traffic we were getting from Apps was I thought was going to be appalling because what I assumed is that you're playing some sort of game or you're doing something in an app and then this ad pops up and you click to get rid of it and you accidentally click on the ad. And there definitely was a certain percentage of the traffic that was happening to. But then what we were seeing and we separated all these different display ads out and we actually worked out that app traffic was actually working quite well. But what it was doing was driving search traffic. So you are playing a game. You see an ad, you either click on it, you don't click on it because it looks interesting or you've clicked on it by accident. But then a certain percentage of those people were then going, well, that looked quite interesting. I wasn't really in the mood for it then because I was in the middle of playing whatever I was playing or doing whatever I was doing. But actually I went into the search engine a day or two later and I did a search and I found the website and went through to it and then I still wasn't quite persuaded enough to buy their stuff. But the third time I went back to the website, I then started typing in the website address into my bar of my browser and the browser remembered the website and therefore I came to the website directly.
- [03:33]
B
So I want you all to take note here because this is fascinating. I feel a little bit like, like a wildlife kind of reporter here. But here we see the digital marketer in the natural habitat and he's clearly overly thinking what's actually going through his customers minds. But that's what you have to do in this world and that's how you understand what's going on. Thank you Daniel.
- [03:56]
C
You're very welcome for observing me in this way.
- [03:58]
B
Insight into the inner workings of your search traffic's mind is fascinating.
- [04:03]
C
So we basically identified that our display ads were working. But actually click through rate becomes a bit of a nonsense in a lot of cases because what happens is you do see a click through rate and you're judging on that. And I'm paying on cost per click, so I do care about that. But also what was happening is people are seeing our ads and it's driving an increase in search traffic because they're kind of thinking about our brand and they're thinking about our product, they become aware of it and then they're finding us. And also those people that are clicking are then potentially revisiting on different devices. And then what's happening is that we're not Maybe connecting those journeys they come through. The first time they remember the website address they go through and they search for us on a different device. So attribution wise, it doesn't make things much easier. But just accept that if you see an increase in direct traffic, it can often mean that things like display advertising and anything else that's driving people to come back to your website or search for your website or just typing your website domain afterwards can go up from that point of view as well.
- [05:04]
B
There's another angle on it as well. E consultancy did a really good paper on this, I think back in June of this year. But you know, very often you notice the growth because you're doing year on year, month on month comparisons, or comparing this month, this year with the same month last year, which we all love to do. But what's happened is kind of really big change in consumer and kind of customer behavior really. And the difference is that more and more people are using mobile devices. And you know, a lot of sites that I've looked at recently, it's not unusual to see, you know, 50, 55, sometimes 60% of the traffic coming through can be from mobile stroke tablet, tablet devices. And when they're using those, typically what they tend to do is just type words rather than using a search box, you just type it into the browser address bar and the browser will instantly link to previous websites that you've been to and take you there rather than going via the search engine. So whereas you would have previous seen good organic search traffic. So I think that the direct to website or none traffic as it's labeled in Google Analytics is actually reflecting this. It's just reflecting that people are using their devices in different ways. And it makes perfect sense because when you're on a browser you don't want to have to have extra clicks if you're on a small mobile browser going through to search engines and the results pages and whatever it wants to intuitively get you where you want to go. So the devices are doing a better job, but they're kind of cutting out the middleman really.
- [06:33]
C
They are. I mean it's interesting, the iOS particularly. So Apple have been very, very clever at that. You do something on one device and the browser on the other device will remember what you've searched for, what you've been looking at previously. But also to the extent with iOS where you have something open on a particular page in one browser, it will be open in the other browser on that page and so on as well. So it does cut out the search engine. That journey, which again means your direct traffic is starting to show up in a much higher way. And don't always rely on the fact that you're seeing low click through rates for a campaign. It's not having an impact. So I'd be very careful of when you run a campaign, have a look at that and then look concurrently, are you getting more direct traffic and are you getting more search traffic at the same time? Which means you need to do some isolated testing because if you're running four or five things at the same time it's very, very hard to work out the impact these things are having. And you might not have that luxury if you're a big brand, a big organization. You can't just say, well everyone stop everything, we are just going to do this at the moment and then see what happens. Because that could obviously be a bit of a disaster. So I think the point that Kiera made about observing is a really important and driving some inferences because actually attribution modeling analytics are fantastic to try and get you scientific detail but there'll be some stuff you just won't have the detail on because you don't entirely understand what your audience is doing.
- [07:52]
B
And therein lies the problem with analytics because I'm finding this increasingly all analytics does is it tells you what it very rarely answers with conviction the why. And actually a what story is quite dull. You need to start to understand the motivations of your key characters in order for a story to be rich and fulfilling. And every marketing story should be rich and fulfilling. I think so yeah, it was fascinating for me because I, you know, I do exactly the same thing and it wasn't until I saw you doing it there, it's kind of like yeah, actually you are, you're walking in your customer's shoes, you're getting under their skin, you're trying to understand their motivations and actually that can lead to some really great hypotheses which you can then go out and test out and you do have to test them out. So it is a danger with this stuff because you can get slightly carried away with a good story as we all do and assume that's it. You can't, you have to test.
- [08:47]
C
One problem we've seen a lot is, is that people will try channel out. So say for example, I do some banner ads, do some display ads on the Google display network and I get a certain level of result and go do you know what? That wasn't as good as I expected. Or it wasn't as good as some Other channels. Right, let's stop doing it. And what actually you haven't done, you haven't explored all the variables. And I think this is a really important one to consider. I've seen this for years in social media where someone says, right, we're gonna do some Twitter. They go off and they run a few tweets. No real strategy behind it. They look at it and go, did we get much back? Not really. Okay, Twitter doesn't work for us, so let's not do it again. Actually, you didn't try out different styles of tweets, different tones of tweets, different tweets, different times a day, volume of tweets, building your audience. So there were so many other variables that could have been changed that actually you don't truly know if that worked or not. And it's too easy to throw things away. And I'll give you a really practical example. We did some testing really where we did a big push on pay per click in Google. Then we switched that off and we did a big push push on Google Display network to try and in isolation test these two things against each other. So we're in a fairly luxurious position to be able to just do this testing without too much concern. So we did those two things and we saw, we kind of benchmarked each of them to see what results level and what cost we got from each of them. Then the third experiment was to run them both at the same time. And when you run them at the same time, the sum of the parts is actually greater in what you end up with. So the sum of the whole. Because the reality is that the way that we're buying, there are multiple touch points and if you are touching people across different channels and actually in a simultaneous way, it has more of an impact. Just think about how branding has worked for years. It's fairly obvious that if I can keep engaging with you across different things, I'm going to have a cumulative impact. But actually we weren't really considering that. So one channel is impacting the other. So even if someone doesn't click on something, they maybe see a pay per click ad and they click on that because they've actively searched, then they leave, then they see a banner ad, then it reminds, and then they go back. When I think about how I buy things and I've tried to judge myself on this recently, it's a fairly chaotic, random process. It seems like because there's a few things on Facebook that I've purchased, products that's just you, maybe it is Maybe it is. And particularly the stuff I've been buying recently. So you might have seen those umbrellas that open upwards.
- [10:59]
B
You didn't?
- [11:00]
C
I bought two of them.
- [11:01]
B
Why?
- [11:02]
C
Because I just think it's brilliant. So. So these umbrellas that open up, the idea is you can get into a car or get through a door because the umbrella opens the other way around. If you've got no idea what I'm talking about, don't worry, it's all perfectly sensible. And someone had mentioned to me they really like this thing, they thought it was funny. So I thought well I'll buy them one for Christmas. So it's more of a Christmas gift. But I must have seen that thing five times, that video ad on Facebook before I ever clicked on it. Then I clicked through website and I went through three times before I purchased. It was quite expensive. And then eventually I saw another Facebook video and thought I am gonna buy it and I went through and did.
- [11:32]
B
It just tipped you over the edge.
- [11:33]
C
It did. And I might have had a glass of wine at the time. So that have also tipped me over there just making this crazy purchase. But the reality is that actually think about, look at how you purchase things and this sounds very consumery at the moment, like consumer products. Actually we were discussing this a little bit before this. You buy B2B services very often on emotive and emotional reasons. I like the person I spoke to, I like the consultant, I like the brand, I trust them, risk adversity, I don't want to look stupid. So I'm going to select a brand based on those things. All of these are emotional decisions and we are emotional beings. So actually even B2B purchases are being impacted in a similar kind of way.
- [12:13]
B
Just something to add into this because it's an area I've been taking a great deal of interest in recently and actually click through rate for me, coming from a PPC background was always the key. And actually what I learned very quickly is if you were trying to justify display campaigns on click through rate alone, you were lost. You were never going to be able to do that because actually in a lot of cases the click through rate is miniscule. Miniscule. Miniscule. But actually when you start getting to speak to some of the guys at the big display networks, they quickly point out, well, it's kind of a bit of an old fashioned metric. I'm like really? Yeah, well it's good for PPC but it doesn't really work in display. What I've learned is that actually if you've got a really good direct response offer, you've got some form of coupon or money off voucher that you're pushing out, then actually you can get some quite good click through rates if you're able to target people in the right places. Perhaps people sipping wine. Daniel, whole new area of advertising, pay per sip could that take?
- [13:20]
C
I think this would be my manipulative advertising, trying to advertise to drunk people and take advantage of their state. Or there is plenty of late night TV advertising that works like that. But anyway, we'll move on.
- [13:28]
B
But you can do that and with offers like that you can talk, particularly on Facebook. Facebook's brilliant for targeting the right audiences and right interests. You can still get good click through rates with that. But if you're going down more of a remember us and maybe targeting some of the products that people have looked at, retargeting campaigns and stuff, they work in a much more subtle way, as does email. I think. Now very often you see a correlation between when you send your email out and vast numbers of clicks on your pay per click campaigns or on your direct to website activity, or even on your organic activity. And it's because you have this layering of information and actually as a good business, a good brand, that's exactly what you want to do. You need to rise up above the noise that consumes all of us all of the time and just get that share of mind.
- [14:21]
C
I think that's absolutely right. Two things to add though, two limitations to that. One is that saying that click through rate is not a very good measure is also a really good way of selling really bad advertising space. So you run a campaign and then no one clicks on it and then you phone up and go, well this was terrible. No one clicked, ah well it's not about clicks, it's about branding. As I said before, and I'll continue to say branding is used as an excuse in digital marketing because you say to them, why did you do that campaign? So it was a branding campaign, which basically means they don't know why they did it and they don't know how to measure it. Now I'm not saying that branding isn't important, branding is massively important. Share a mind is really important. But just be cautious when you're buying ads, when someone tells you that click through rate isn't important because it's not that it's not important, but it's one of the factors. But actually you need to go beyond that. But if you're going to go beyond it, you need to know what you're going to measure. Secondly, retargeting we mentioned the idea is I go to your website, look at your product and then you start saying look, you didn't buy this thing or you've been to us before frequency cap, think about variations. And what I mean is that if I've looked at your product, I didn't buy it and then you just keep showing me ads for it, you will eventually drive me to distraction.
- [15:25]
B
It annoys us all. And so why would you do it it if it annoys you, why would you do it too? Gets me on my well, the reason.
- [15:32]
C
We do it and is you look at the statistics. If we just let's retarget everyone, we make more sales and okay, but at the expense of what how many people have you irritated and actually lost future custom from? So I think that's important. So frequency capping only shows a certain number of times. But also maybe show me the product again. But then what if I still haven't bought it? Maybe show me some other products, tell me some things about the product that I didn't know. Use it for loyalty. If I've bought something, say well you could now use this for this or you could now purchase it. So thinking about retargeting in a smarter way and just thinking about what works for us individually, but actually not being misled by the stats that say just blanket bomb everyone with this because then you're going to get more sales. Yes you will, but at quite a high expense I'd say as well. So just to sum up on this direct traffic piece, you will probably get more direct traffic. That's not necessarily a bad thing at all. To minimize it, make sure you have tracking code for your analytics on everything. So you label everything up. So you're going to minimize the amount of traffic coming in that you could identify directly where it's come from. And then have a look at the connection between each of the things you're doing and how that's impacting your search traffic, how that's impacting people coming in directly to your website. But also if you display do you get more clicks on your pay per click and so on. Look at the connections between those things, which is not necessarily easy to do when you're doing lots of things at the same time. But don't just look at direct traffic and go oh we're doing great, loads of people just know us and come straight through to the website because that's part of it. But there's also a lot more going on and you have to investigate.
- [17:02]
B
And the final thing I'd add in there, particularly with display, is leave yourself a little bit of lag time. Because in my experience there's always a minimum of kind of two to three days between showing the display ads and the effect that you get. So you need to need to factor that in, look for the correlation. You begin to see how that pans out or works for your particular market. And it does vary from market to market and from business to business. But you need to get a good sense of that.
- [17:28]
C
Yeah, I think the numbers I've looked at recently for a few brands, three days is when it should be statistically significant. So after the three days you should have had the impact from one channel or another and that should be enough of a time lag from a statistical point of view. So good luck with all your analytics analysis efforts. Do let us know any examples what you've been doing and we'll speak to you again on the Digital Marketing Podcast.
- [17:58]
A
Thanks for listening to another episode of the Digital Marketing Podcast, brought to you by Target Internet. If you'd like to get more information on the show, get hold of back issues of this podcast, or get details on any of the links we mentioned, please visit our website at www.targetinternet.com. if you've enjoyed the show, we would love to read your feedback. Please rate us in itunes or even better, write us a review. Or if you have any questions, please get in touch. We'd love to help.