System reliability and monitoring challenges threaten service quality

Askable is a leading user research platform helping enterprises scale their research efforts and uncover insights at speed. Powered by AI and backed by a global network of researchers and participants, Askable supports UX, design, product, and insights teams in making faster, more confident decisions. With tens of thousands of active users each month, it’s become a trusted partner for organizations focused on customer-centric innovation.

As Askable scaled rapidly, its engineering team encountered the kinds of technical challenges that come with fast growth. Occasional performance issues and limited visibility made it harder to proactively manage system health. Without centralized logging in place, creating automated alerts wasn’t yet possible, and the team relied on manual monitoring to stay ahead. “We sell a product, and if the product is down, we’re going to get complaints from our customers,” says Isaac Povey, software architect at Askable.

As Askable’s platform usage grew, its existing monitoring setup began to show limitations—requiring developer involvement for most issues and occasionally losing critical logs. With a rapidly expanding user base, improving system visibility, reliability, and performance became a key focus. The team set out to enhance platform stability and reduce resolution times, while also equipping support teams with better tools to assist users more efficiently.

Fast, integrated monitoring out of the box

Askable’s engineering team first considered developing their own monitoring solution. They also evaluated various other solutions but found most were slow and difficult to search effectively. Askable ultimately selected Datadog for several key reasons. First, the company needed a solution they could implement quickly, and Datadog had a low barrier to entry for non-technical users and comprehensive training resources for technical staff. “We are a small, lean team, so having the simplest solution possible and ease of use is critical. Time to value was the biggest thing with Datadog. It just seems to be better than any of the other solutions,” explains Povey. “The more we can do to enable developers with something that works out of the box, the better our monitoring becomes.”

The intuitive log querying experience was a deciding factor in choosing Datadog Log Management. Features like calculated fields, subqueries, and log joins by reference tables significantly enhance their investigation capabilities, enabling support staff to handle issues that previously required engineering intervention. 

“The ease with which the search works, and how non-technical people can actually search logs—and not just search logs but create metrics across logs and create quick graphs—is really powerful.”

The integration among Datadog’s products provides Askable with a comprehensive monitoring solution. “The biggest feature of Datadog is the great correlation between all of the different metrics, logs, and traces,” adds Povey. This seamless integration enables them to track requests across their entire infrastructure without extensive configuration.

The Askable team particularly values Datadog’s customizable dashboards. “It’s so easy to make dashboards in Datadog—so much easier than any other telemetry tool I’ve used before,” says Povey. The dashboards are displayed on large screens throughout the Askable office, providing at-a-glance visibility into system health.

Implementation was completed within a week, enabling Askable to immediately begin tracking down issues causing performance outages. “Having those metrics was of huge importance so we could know how bad the problem was,” says Povey. “We were able to trace the root cause behind the main issue and fix it.”

Today, Askable turns to Datadog for all platform issues. Backend engineers regularly analyze both logs and Application Performance Monitoring (APM) traces to diagnose bugs and track down performance issues. Frontend and mobile teams use Real User Monitoring (RUM) to track issues in web and mobile applications. Support teams use both logs and RUM to identify and triage reported user issues before reporting to engineering teams, which enables them to shift left easy-to-fix or common issues. And security team members use Datadog Cloud SIEM to centralize security logs, alert on suspicious activity, and conduct investigations.

Transforming performance and user experience

Centralizing and enhancing monitoring has enabled the Askable engineering team to improve application stability and performance.

“Most errors we can now fix within a day, whereas previously it would potentially take weeks.”

The impacts of these improvements have been substantial. Uptime has increased from 98 percent to 99.95 percent. Backend server unexpected restarts have dropped from over 20 times daily to zero. Bug reports from the support team to engineering have decreased by 50 percent, while time to resolution for critical issues have improved by 100 percent. “The decrease in time to solve critical incidents has been huge. Now we can get an alert from Datadog before it becomes a problem,” says Povey. “We also use Watchdog, which is really good for catching odd behavior that you can’t necessarily make an alert for.”

Askable has also used Datadog to enable better fraud detection. “Detecting fraud quickly within our participant panel is critical to keeping the quality of our customers’ research high,” Povey explains. “Rather than trying to build that into our internal admin portal and build out a database to store the metrics, we just started logging fraud events. We built a dashboard in Datadog and alerting based on that.”

Now, their customer service team receives a Slack notification anytime there is suspicious activity. They can then investigate it through Datadog dashboards and take appropriate action. “I was surprised at how easy it was to set up that integration,” adds Povey. 

Looking ahead, Askable plans to further centralize its operations around Datadog as it continues to help its customers transform user research through technology. “We’re starting to use a lot of AI and we’re seeing some of the features coming out from Datadog around AI observability,” says Povey. “I think that’s going to be critical for us.”