« Nous avons des SLO sur la plupart des éléments de notre système et ces workloads sont définis en fonction des flux des clients clés dans le produit. Ainsi, nous avons la capacité proactive de voir ce qui impacte ses flux client dans le système et de le gérer à l'avance, » explique Craig McGuff, Responsable principal Ingénierie chez Serko.
« Lorque l'on prend une posture plus proactive, on fait une recherche à l'avance. On voit les problèmes, on les résout avant qu'ils ne se déclenchent, et on libère ainsi le temps des ingénieurs, qui peuvent alors être plus productifs ailleurs, » continue Craig McGuff.
« Nous voyons tout en un coup d'œil : si quelque chose se produit et a un impact sur la recherche, par exemple, et nous pouvons l'explorer et voir s'il s'agit d'un problème d'infrastructure sous-jacent. Est-ce qui se passe autre chose ? Cela nous donne une visibilité totale sur ce qui se passe à chaque étape de notre entonnoir et, au bout du compte, cela stimule les conversions, » confie Craig McGuff.
"We measure downtime in minutes now, not hours. We've had to change the axis on our reporting charts because the improvement is so dramatic you can't see the current data on the year-on-year scale."
Key metrics include:
- 60% reduction in high impact incidents
- 30% improvement in automated issue detection
- 20% increase in customer satisfaction
- 2 hours saved per engineer, per week
- 6 year partnership
- 155 users across 17 departments
A custom metric for Serko is engineering hours spent on incidents per thousand bookings. What once consumed hours of engineering time now takes fractions of a minute. Those reclaimed hours are being reinvested into building better products. The shift is visible at every level of the organisation, including incident performance reporting to Serko’s board.
Customer experience has been transformed, with customer satisfaction increasing by 20%. The existential incidents of previous years, such as complete platform outages that left clients stranded, are gone.
Internally, the efficiency gains are equally significant. Engineers save at least two hours per week previously lost to reactive troubleshooting. That capacity is now able to be directed toward innovation, including Serko.ai, the new multi-agent AI solution designed to act on behalf of both travellers and companies. Dashboards are being surfaced more broadly across the organisation, making operational health a visible, shared concern for both engineering and product teams.
The future: Scaling with confidence
Serko is entering its most ambitious growth phase yet, with major initiatives like Serko.ai and an expanded Booking.com partnership underway. Both are being built on Google Cloud Platform (GCP), as Serko transitions its primary cloud infrastructure from Azure to GCP over the coming months. New Relic is central to that migration, ensuring that as systems move and new services come online, observability coverage is immediate and complete.
The priority remains to support Serko's strategic direction in consolidating technology, simplifying complexity and building to the scale and stability that a global corporate travel platform demands. With New Relic as a core pillar of that strategy, Serko has the visibility to stay ahead of issues and the confidence to grow without compromising the experience customers depend on.
Simon Young concluded, "The partnership with New Relic is a very important part of our strategy. New Relic gives us that single pane of glass across all of our systems, so we know what’s going on behind the scenes at every point of the user journey and what needs fixing or improving before the customer does."
