Skip to main content

For online retailers, intelligent traffic routing wins the sale

(Image credit: Image source: Shutterstock/Maxx-Studio)

Before the dawn of the internet, drivers typically relied on roadmaps, the kind that unfolded multiple times and you could never seem to get them folded back into one tidy rectangle. As internet adoption increased, drivers moved towards the digital cousin to these paper maps – printouts from websites such as MapQuest. Directions were spelled out, turn by turn, providing an easy system for finding where you needed to go. However, these printed-out directions certainly could not tell you if there was a traffic jam on the highway you were about to merge onto, if a sinkhole had developed overnight, or there was a road closure – helpful things to know to make your journey quicker and safer.

These days, you can download one of several apps to your smartphone and get up-to-the-minute information on traffic conditions and optimal routes to your destination. Now, imagine that you are a large retailer; would you rather your customers use a static directions provider or a real-time app to find the best route to the closest store? 

Clearly, you would rather they use an app; otherwise, they might find themselves stuck in traffic—or that sinkhole—and decide in frustration not to go to your store at all. When it comes to online businesses, the situation for your customers is quite similar. 

Online retailers, at the national or global scale, actually have websites that are comprised of multiple sites. To serve a large, geographically dispersed audience, your business needs many points of presence on the internet, just as the large brick-and-mortar retailer needs many store locations to thrive in today’s marketplace. This brings us to the question of how users actually find your site. 

Heavy demand

When a would-be customer types the name of your store into a browser, their PC, smartphone or other mobile device queries the domain name system (DNS) to find your site’s IP address. DNS delivers the IP address, and the device connects. This all happens automatically and typically within a couple of seconds. 

So, how does DNS decide which of your many internet store fronts the user should go to? Traditional DNS services operate more or less in the mode of MapQuest and other similar web mapping services. These services “know” where the user is located and send them to the internet store location that is the closest geographically. The DNS also may be able to check if the site is currently in operation. This works perfectly fine most of the time. 

For retail, though, “most of the time” is not sufficient; customers who cannot get to your site cannot purchase your goods. The internet route to the closest geographic location may be congested and slow. Conditions frequently arise in which the geographically closest point of presence does not deliver the best performing response. There can be a number of reasons for this: equipment outages within the internet infrastructure, server outages in the datacentre, or one of the most common and least predictable reasons – unplanned, local demand spikes.

User demand is highly variable. There may be heavy demand from users located in the eastern United States while relatively few users are coming from the west coast, for instance.

Rather than overload your east coast datacentres, wouldn’t it be better to send some users to the lightly loaded, west coast locations? And what if the DNS had more information, including:

  • The amount of capacity at each location to serve incoming customers?
  • Which of your locations is closest in “internet time” to each user trying to get to your ecommerce store?
  • The number of customers currently being served at each location?

By leveraging this information, you could send your customers to your site more quickly on average, and they would get faster service once they get there.

This is the power of intelligent DNS – and its business impact is significant. Historically, web performance has been the focus of developers and operations teams, but it is important for c-level executives to understand the business value of faster load times. Multiple studies have shown the positive effect that better performance has on ecommerce sales:

  • The Aberdeen Group found 25 per cent of users will abandon a web application after just three seconds of delay
  • Intuit increased conversions three per cent for every one-second improvement in page load time
  • Amazon found that every 100ms delay in page load time reduced sales by one per cent

Consumers have come to expect and now, in fact, demand a fast and satisfying customer experience. Your company spends heavily to drive user traffic to your website and to make the online shopping experience as appealing as possible. The fact however, is slow page load times and sluggish response will reduce both sales and repeat customer visits. Shoppers are far less likely to buy from your site again when they experience load time delays. There is an additional ripple effect here too – customers tell their friends and family about unsatisfactory experiences, and that can quickly escalate. In the digital world, word of mouth has far reaching consequences.

The good news is you can prevent many of these problems by using an intelligent DNS traffic management system. It is easy and straightforward to implement and works behind the scenes to take your customers where they want to go – to your “check out” page with a full shopping cart. With so many different outlets competing for consumer dollars, you do not want to give shoppers any reason to turn elsewhere. Intelligent DNS traffic management can be the key differentiator when it comes to your business’s success. 

Jonathan Lewis, vice president of product, NS1
Image source: Shutterstock/Maxx-Studio

Jonathan Lewis
Jonathan Lewis brings to NS1 over 25 years of experience in the IT industry in a career comprising product management, product marketing, customer service and systems engineering roles. Prior to NS1, Lewis led teams at Nortel that brought numerous network and security products to market including IPsec gateways, SSL VPNs and endpoint security. He played key product marketing roles, contributing to the success of mid-size and start-up companies, including Arbor Networks and SSH Communications Security. He holds B.S. and M.S. degrees from McGill University, an M.B.A. from Bentley College and a CISSP certification.