While many monitoring tools focus on a binary state where there are only two conditions; a white one where everything is OK, and a black state where the monitored service is down.
In reality there are multiple shades of grey in between which can have a large impact on how well your business users feel your service is behaving.
All providers of web services are looking to maximise completed transactions; this is the number of users of the service who complete an action which either results in revenue generation or produces data which is relevant to the company. Examples of such transactions are: site registration or sharing for a social website, check-out completion for an e-commerce application and viewing of rich media/advertising. On the other extreme there is transaction abandonment; when the user exits the web service before a transaction is completed.
Obviously for a web service to be successful the business will want to minimise or completely eliminate the user abandonment, as these users have been hard earned by the investment of marketing and public relations departments, and falling at the final hurdle should be eliminated from a technology standpoint whenever possible. Accordingly, it is no longer good enough to monitor the up/down status of a network or infrastructure element. You need to know where the customer-facing service is actually performing on the grey-scale between up and fully functional and down and inoperative.
If for a moment we just look at web page loading times. It seems obvious that slow web pages are a bad thing for any web service. Google, through some enlightening experiments have shown just what kind of impact a delay can have, even for such a ubiquitous service. Google have some interesting findings
“Our experiments demonstrate that slowing down the search results page by 100 to 400 milliseconds has a measurable impact on the number of searches per user of -0.2% to -0.6%”
While these percentages seem small the 0.6% represents more than 34 million searches per day for Google in 2014. Furthermore users experiencing slow loading were less likely to return with the number of searches steadily decreasing week on week.
“Users exposed to a 200 ms delay since the beginning of the experiment did 0.22% fewer searches during the first three weeks, but 0.36% fewer searches during the second three weeks.”
Let’s leave the world of web services for a second and discuss the performance impact for a more real-time service like the carrier world. Here performance issues can have disastrous effects.
With it being easier than ever to switch your communication provider (they call it “consumer churn”), merely providing a service that works isn’t good enough, they need to know how well it is working with as much meaningful granularity as possible. Their hard-earned customers who are subjected to poor service are swiftly moving on to other providers and what’s more concerning to communications providers is that 58% of Americans would never use a company again after a single negative customer experience New Voice Media. Continual service monitoring is essential, in order to be proactive, and to put remediation in place before any service impact is felt by the customer.
Let’s look at e-commerce: Akamai has carried out extensive studies of user behaviour and below are some of their observations. It is clear that slow loading web services and digital store-fronts are very sensitive to application and network performance. The key to improving these is to take a holistic approach to network efficiency and engineering the network around these key services. The only way to get a complete and accurate view of any complex network is to improve visibility by tapping, aggregating and then analysing the network and application data and its behaviour in motion and as-it-happens.
“Online shopper loyalty is contingent upon quick page loading, especially for high-spending shoppers. 52 percent of online shoppers stated that quick page loading is important to their site loyalty, up 12 percent from the 2006 study.
Shoppers often become distracted when made to wait for a page to load. 14 percent will begin shopping at another site, and 23 percent will stop shopping or walk away from their computer.
Retail and travel sites that under perform lead to lost sales. 79 percent of online shoppers who experience a dissatisfying visit are less likely to buy from that site again, up 17 percent from the 2006 study. 64 percent would simply purchase from another online store, up 16 percent from the 2006 study.”
The challenge when trying to monitor an end to end service is that there are multiple devices and applications both on the network layer and right through from the user to the applications which together provide the complete experience. Each link in the communication and processing chain between the user and the application they are interacting with needs to be identified, monitored and the elements of the transactions each of these items process, influence and affect stored and fully analysed.
Only with the complete and clear vision that complete monitoring and analysis provides will the many shades of grey coalesce into a complete picture that will highlight whats needs to be focused upon to keep user experience at its best.