Ammarah Marjan, Charles Graham, Margaret Bruce & Andrew Mitchell
Over a third of all website traffic originates from free (Direct) rather than paid (social, organic, referrals) channels. There are misconceptions about Direct channel due to which marketers often ignore or sidestep it from the analysis. The aim of this whitepaper is to evaluate the potential of Direct as a valuable traffic source and address the issues associated with it. We used a dataset constructed from traffic flows totalling over 3 million hits observed over two years on five websites. Our results are potentially important for marketers who strive to generate more traffic to the website. We find that Direct channel can contribute to high traffic frequency. We also estimated its quality and propensity to convert which leads us to conclude that Direct traffic is a return of previous efforts and marketers should pay attention to it.
When was the last time you looked at the “Direct” traffic on your website? Just because you sociawebsite is the most important job a digital marketer must do, for this very reason firms spend a fortune on digital campaigns orientation such as SEO, PPC, paid social. Such campaigns have become increasingly important because marketers spend money on it. But less attention gets paid to the Direct traffic that is already present on your site with significant volume and consistent frequency. You might think it’s a “dustbin” for the things that don’t matter such as missing tracking codes, non-web documents or improper re-direction but this traffic could be valuable to the business - but no one has investigated it yet. In this paper, we look at the volume of Direct traffic, the quality of that traffic and the way it shapes up over time – another view that is not often taken.
Why ignore Direct?
In theory, Direct traffic contains the users who enter a homepage address Directly into the browser navigation bar, uses the bookmarks from the browser or uses the links from the history of web browsing (Artur et al. 2016) to enter the site. Generally, there are six sources which are commonly managed by marketers to acquire traffic. These are identified as channels by Google Analytics and consists of organic and paid search, email, referrals, social and Direct. While the first five channels are on the fingertips of most digital marketers, Direct as a source of driving traffic volume remains neglected. A common reason is that Direct traffic contains no trackable information of the source on Google Analytics as it is not prompted to the site as a result of any digital marketing campaign. Therefore, any resulting inbound traffic generated through Direct becomes invisible to Google Analytics and remains immune to further analysis.
The prevailing attitude amongst digital marketers is that Direct traffic is an unavoidable inconvenience; as a result, discussion of Direct is typically limited to ways of attributing it to other channels or side-stepping the issues associated with it. However, Direct traffic is likely to be an important traffic source. We have spent significant time on Google Analytics and examined data across five website types. After analysing our data in timeseries, we have got some interesting findings to report. In this paper, we answer three important questions about Direct traffic and why it is worth looking at it.
How much Direct traffic are you getting?
In Table 1 we looked into the traffic volume across five website types and the percentage of traffic each channel contains. The Direct channel contains a significant amount of users (37%) and accounts for more than one third of all website traffic. This works positively in our favour because 1) there’s lots of it and 2) all this traffic is free as this channel is not managed or invested in.
Table 1. Distribution of channel traffic in an average quarter
As much as anything, the users in Direct channel reach a website by typing URL of the website, such users at the very least are already aware of the website brand, and in the case of a bookmark are highly likely to be purposeful repeat visitors (Pakkala et al, 2012; Plaza, 2009) already engaged with the brand. Because of such substantial volume, Direct traffic is likely to contain high-value users who may relatively have higher propensity to convert.
Never mind the quality...
Many have reported that each traffic channel differs systematically as to both quantity and quality of traffic delivered. Since we have already established that Direct channel holds a fairly larger volume of traffic, it is also worth looking at the quality of traffic it provides. In Table 2 we analysed Direct compared to the total site with the help of Google Analytics behavioural metrics on three performance measures, for example, its bounce rates, page views/session and average time spent on the site.
Table 2. Behavioural metrics in an average quarter – Direct channel vs. total website
We found that Direct users had 20% higher bounce rate, 12% fewer page views and 22% less time spent on the site. Although on three of these metrics Direct did not appear to be better than the rest of the website there could be possible reasons for it. As a matter of fact, Direct users have previously acquired sufficient awareness and interest in the linked brand to visit its site. They are likely to be on the desire phase of the brand - prompted by typing the link into the address bar or using a bookmark. For one thing, such users perhaps know their way around on the site for the reason that they spend less time on the web. In fact on the e-commerce sites in our analysis, Direct appeared to be among the top three channels with the highest conversation rate. Hence at very least, Direct is a free source and contributing significantly to the website performance, therefore, we should pay attention to it by using meaningful engagement metrics and find ways to manage it.
Is it consistent?
Variation in traffic frequency is common due to short-term marketing investments but it is also possible that our analysis in time series captured a trend between Direct and other channels. In order to find out, we compared the different sources of traffic acquisition for the employee engagement website over seven quarters.
Figure 1. Traffic acquisition strategies on employee engagement website
Online campaigns bring immediate growth in traffic acquisition as found in Figure 1 (Display with 62% in Q1, Paid Search with 73% in Q3). This spiking pattern is seen widely across many website types. It is evident that marketing investments deliver a boost on par in acquiring high-frequency traffic. However, as soon as the investment is pulled out, the total traffic declines sharply. But what does not change throughout is the Direct curve. As seen in Figure 1, traffic volume in Direct channel is stable, and this bunch keeps coming back with the same consistency quarter after a quarter which indicates they are repeat visitors. This is a motivating factor, as repeat traffic is of particular concern to build customer brand engagement through repeat activity on a website and such pattern is a great interest for us.
We extended our analysis to all five websites to validate our findings further. In Table 3 we calculated a coefficient of variation for each channel over eight quarters on each website and found that while all other channels are wobbling around due to campaigns orientation, Direct channel varies only slightly (0.3).
Table 3. Coefficients of variation in channel effects
Direct traffic does not fluctuate much, it implies that such users consistently came back to the website quarter after quarter without any effort or management. Such behaviour not only tells that these users are highly engaged and aware of the brand but also that Direct traffic contains purposeful repeat visitors that are loyal customers to the website.
Consider it as a payback time
Direct is far from an inconvenience as perceived by many. In fact, it’s a return on the investments you have already put in to be on top of your customer’s mind. Look at all these users in Direct channel on your website who know your brand and potentially repeat visits. In the light of our findings, Direct contains quality traffic and if managed well, has a high propensity to convert. Therefore, marketers shouldn’t overlook Direct because it’s payback time for the efforts they previously made.
Minimise false Direct traffic
For many reasons, Direct is considered as a contingency option of analytics tools when their processing logic has failed to attribute a session to a source. As some report, Direct contains clutter for the things that don’t matter including missing and broken links, improper redirection, non-web documents and more. But these are quite unlikely to be present in a large proportion. In any case, there are fixes to minimise false Direct traffic allocation (Bennet, 2017). Therefore stepping aside these issues, Direct channel is an extremely valuable segment to analyse for all the right reasons. There’s not much clutter and marketers should make an effort in getting to know these users.
Know it contains dark social
Though there is one considerable issue with Direct for companies looking for the best ways to reach consumers as analytics tools cannot pick up on where this traffic is coming from –and that is Dark Social. However, at the minimum, it increases the importance of Direct as the traffic source because it contains traffic that is coming from social media private messaging platforms. We explore Dark Social in our next whitepaper. But in the meantime, look at Direct channel as there is good value in it.
Bennet, T (2017). The Complete Guide to Direct Traffic in Google Analytics. Mox.com, 29 November. Available from https://moz.com/blog/guide-to-Direct-traffic-google-analytics
Pakkala, H., Presser, K., & Christensen, T. (2012). Using Google Analytics to measure visitor statistics: The case of food composition websites. International Journal of Information Management, 32(6), 504-512.
Plaza, B. (2009). Monitoring web traffic source effectiveness with Google Analytics: An experiment with time series. In Aslib Proceedings (Vol. 61, No. 5, pp. 474-482). Emerald Group Publishing Limited.
Strzelecki, A., Furmankiewicz, M., & Ziuziański, P. (2016). The use of management dashboard in monitoring the efficiency of the internet advertising campaigns illustrated on the example of google analytics. Studia Ekonomiczne, 296, 136-150.