This content has been archived, and while it was correct at time of publication, it may no longer be accurate or reflect the current situation at Microsoft.
It’s now much easier for Microsoft to extract important SAP telemetry insights from its business operations data.
Microsoft relies on SAP to handle a multitude of business processes, including global invoicing and tax. Historically, limited visibility in the Enterprise Resource Planning (ERP) platform has made it difficult to identify and resolve issues.
Not anymore, thanks to work by Microsoft Digital, the organization that powers, transforms, and protects Microsoft.
“Service health is our number one priority, but SAP has been a black box for us,” says Michelle Concannon, a group principal manager within Microsoft Digital’s Cloud + AI group.
Put bluntly, what happened in SAP stayed in SAP. The proprietary nature of an ERP system can create challenges with surfacing insights that can pinpoint issues in a processing flow.
“We couldn’t scale with the number of requests, so how could we get smarter?” Concannon says. “We have lots of teams working on tax, and most of it runs on SAP. If taxes aren’t calculating correctly or there’s an issue, we need to know about that.”
Since the Microsoft Digital team rolled out new end-to-end SAP telemetry with Microsoft Azure, all the teams at Microsoft that use SAP to manage their work processes are benefitting from key insights being delivered to them. With newfound visibility into their workstreams, engineers can pinpoint and predict problems before they happen, transforming how they respond to SAP service health. As a result, troubleshooting has become proactive, tax and invoices are being calculated correctly more often, and important business intelligence is being captured.
[Find out more about Microsoft’s monitoring platform for SAP in Microsoft Azure. Learn how Microsoft Digital has created a unified telemetry platform for monitoring end-to-end enterprise health.]
Working with limited visibility
A massive amount of data flows into Microsoft’s SAP systems.
Business teams from across Microsoft, whether they’re in North America or Brazil, India or Singapore, use SAP for back-office operations and processes.
One such business is Microsoft Azure Volume Licensing. Microsoft Azure, Microsoft’s cloud product, uses a volume licensing approach to offer customers a flexible method of deploying services on the cloud. In turn, that allows for Enterprise Agreement subscription models and pay-as-needed models.
“Not only is there an increasing number of customers using Azure, but each invoice also has a larger number of line items,” says Maeve Tait, a software engineering manager with Microsoft Digital. “We didn’t have a framework to see if the data inflow, both the volume and the nature of the invoices, coincided with lags in performance.”
During the same time window every month, annual billing for subscriptions and impromptu overage billing creates a high volume of data to be processed by SAP.
This increase in Microsoft Azure revenue is good for business, but it brings with it an equally large spike in volume of anomalies signaled between Vertex, Microsoft’s external tax system, and SAP.
“When you’re running a live system and a customer goes to place an order, the order confirmation may not appear immediately, or the tax charged in the pricing may be incorrect due to a systems integration communication issue,” says Louise Kennedy, a senior program manager on the SAP Tax team for volume licenses. “That’ll impact the customer.”
This breakdown goes beyond a poor experience for Microsoft’s customers. It also results in wasted resources, including costly delays and lost revenue.
“Triaging is time-consuming,” Kennedy says. “Overcharges and undercharges have an operational cost for the people who have to credit and re-issue invoices with correct taxation amounts, particularly when there are multiple invoices caused by the same issue.”
Leveraging Azure fundamentally opens up the black box. It allows us to be more proactive and agile.
-Michelle Concannon, group principal manager, Microsoft Digital
Without clear visibility, errors occurring within SAP only land with an engineer after someone else has been impacted and has manually escalated it in the form of a ticket. Finding the issue requires a significant effort, and at the end of the day, that’s costly.
“The end customer would say their invoice is incorrect,” Tait says. “That would come into our first-line support, who wouldn’t necessarily know what was wrong. They would escalate to the second line, then eventually it may go to a generic engineering team without tax knowledge.”
Microsoft Digital saw an opportunity to transform the entire practice into an automated process with telemetry, but to do that, they’d need to take advantage of some of Microsoft’s modern services.
“Leveraging Azure fundamentally opens up the black box. It allows us to be more proactive and agile,” Concannon says. “We can use the output of the work effectively to be automatically notified of whatever issues would be emerging. Ultimately, this gives us time back.”
Taking a peek into SAP
By moving SAP data onto the cloud and introducing a framework for end-to-end telemetry, Microsoft Digital could extract SAP telemetry with Microsoft Azure.
“The system we have for tax determination is Vertex,” Tait says, referring to a third-party product that Microsoft uses to calculates tax. “If we’re seeing a stress on the system because volume licensing is going up, and we know Azure billings are going up a certain percentage point every month, we can use that to forecast the stress on the service. We have the data now.”
As transactions flow from Microsoft’s front-end customer portals to SAP, the tax telemetry framework unlocks insights into the end-to-end business process flows in SAP to Vertex performance.
After this framework was introduced, things changed dramatically. But generating all this data wasn’t immediately useful.
“It’s a blessing and a curse,” Tait says. “You can pull so much data that you get dizzy trying to figure out what you’re solving. To show how your service is performing, you need to narrow down what you’re looking at.”
Amber Singh, a software engineer who designed the framework for capturing SAP telemetry with Microsoft Azure, agrees.
“How are we going to implement this solution?” Singh says. “Telemetry isn’t new, but the way information between SAP and Vertex is captured at runtime is.”
Microsoft Azure may have opened the black box, but without a plan in place, the intelligence would have remained hidden in a sea of data.
“Having clarity in the problem statement helped focus things,” Tait says. “We knew there was some sort of communication issue.”
Once trends were visible, Microsoft Digital was able to introduce service health improvements.
“When we see spikes in failure rates between Vertex and SAP, we can assign people to go look at that specifically,” Tait says. “Cleaning up the code within SAP has already improved our performance by 30 to 40 percent in some areas.”
It used to take two to three days to be notified of an issue with an invoice. Now, teams are alerted immediately, and the correct support team is working on the problem through an automated ticket within half an hour.
Putting all this visibility to work
With access to end-to-end telemetry, Singh could introduce other functions and tools to empower the SAP Tax team.
“Telemetry empowers digital transformation for Microsoft’s tax team,” Singh says. “It improves productivity, decreases the cost of operation, and should make life better through monitoring, built-in alerts, and automation, all on a reusable framework.”
The telemetry framework implemented in Microsoft Azure is always looking for any tax bound failure. This constant monitoring enables a new intelligent production support model of automated tickets for failures. The tickets are assigned to respective user groups, utilizing Microsoft Azure navigation or embedded XMLs consisting of first-hand data inputs to help resolve the issue.
“The built-in alert framework emails respective groups when a transaction has failed, telling them the impact,” Singh says. “Someone can now go in, monitor, and create mitigation steps.”
Similarly, having visibility into SAP allows for better business learnings, including comprehensive tax data that can be exported to Microsoft Power BI. This type of intelligence is a benefit to everyone.
“It helps us be a better partner for Vertex,” Tait says. “These insights make their system sturdier and more reliable. They’ll be able to cope with future trends as our businesses continue to grow.”
Dashboards for specific personas, including business leaders and engineers, significantly enhance the way users interact with the business intelligence.
For users like Kennedy, pulling SAP telemetry with Microsoft Azure has been all upsides.
“I can see us applying this data to different aspects around service health,” Kennedy says. “In preparation for and during fiscal quarter or year ends, we are asked to check service health. This is how we do it—we use telemetry to accurately report and have metrics to back everything up.”
The data captured in Microsoft Azure can also be applied to other modern practices, like machine learning.
Having actionable telemetry, actions on the back of seeing a weakness occurring, and having that automated ticket, gets you way ahead of the curve.
– Louise Kennedy, senior program manager, Microsoft Digital
“Today we have notifications,” Concannon says. “Tomorrow, with all of that data, as we solve these issues and build a historic dataset, we can do auto-healing and have full automation from end to end.”
No longer waiting for a problem
By painting a clear picture of SAP telemetry with Microsoft Azure, teams no longer have to wait and search out problems. They can derive intelligence from the data and strategically forecast. Finding process issues before they impact service health has been a big cost saver.
“Having actionable telemetry, actions on the back of seeing a weakness occurring, and having that automated ticket, gets you way ahead of the curve,” Kennedy says. “You’re not the one with the red face at the end of an escalation. You’ve caught the problem before it happened.”
These improvements mean 90 percent faster service line health reporting and 80 percent faster hyper-care reporting for the SAP Tax teams.
Designed to be reusable and scalable, the framework is now being introduced to other lines of business across Microsoft. SAP Tax has been simple, with little to no customization required. These additional SAP telemetry insights with Microsoft Azure will only strengthen Microsoft.
“We’ve onboarded different lines of business,” Tait says. “We can see all of their transactions going through. I expect from that we’ll see more trends in the data so that we can start looking at other service health improvements.”
Singh sees this telemetry framework as being a solution for other companies who want to run SAP in the cloud.
“We can show how we transformed the way we do taxes and how it can change things for clients,” Singh says. “SAP is a huge product and Azure has a lot to offer. We just scratched the surface.”
The SAP Tax team is only one part of a multifaceted and complex team, but by combining two industry leaders, SAP and Microsoft Azure can improve the process for everyone.
“We want Azure to be the best cloud for SAP to run on,” Concannon says. “As part of that, the SAP insights should be visible both downstream and upstream, at a level where tax subject matter experts, partners, and stakeholders can be aware of alerts through a business lens.”
Find out more about Microsoft’s monitoring platform for SAP in Microsoft Azure.
