Why does your system keep going down, and why do customers notice first?
Monitoring & Observability
Know about a problem before your customers do. We watch your services the way your customers use them, from USSD and mobile apps down to the servers underneath, so outages are caught early and fixed fast.
What we monitor
Monitoring and observability services
Seven monitoring platforms we implement and support, from digital experience monitoring down to metrics and logs, followed by six capabilities we build on top of them. Select any one to read what it covers.
Apogee digital experience monitoringDo you only find out a service is failing when customers complain?Read more →
Datadog cloud monitoringIs your cloud estate too spread out to see in one place?Read more →
Prometheus metrics monitoringCan you tell a slow day from a system that is about to fail?Read more →
Grafana dashboardsDoes anyone actually look at your monitoring dashboards?Read more →
Zabbix enterprise monitoringToo many servers and network devices to keep an eye on by hand?Read more →
Nagios infrastructure monitoringWant a proven, dependable alert when something goes down?Read more →
Log management and analysisWhen an incident happens, how long does it take to find out why?Read more →Banking and fintech monitoringWould you know within minutes if transactions started failing?Read more →USSD, SMS and STK monitoringAre customers hitting failed USSD sessions or M-Pesa prompts you never see?Read more →Mobile app experience monitoringIs your app crashing or slow for customers you never hear from?Read more →Web and API monitoringIs your website or API down right now, and would you know?Read more →Synthetic transaction monitoringWant to catch a broken journey before a real customer does?Read more →Alerting, dashboards and reportingToo many alerts, or not enough of the right ones?Read more →Watched 24/7/365, because outages do not keep office hours
Monitoring is only worth having if somebody is watching it. Every engagement on this page can be covered by a 24/7/365 SLA support and maintenance agreement, so an alert reaches an engineer rather than an empty dashboard at two in the morning.
We set thresholds with you, tune them once real traffic is flowing, and agree in advance who gets called and when. That conversation is the difference between monitoring that gets acted on and monitoring that gets muted.
Watching the service, not just the servers
A server can be up, the database can be healthy, every dashboard can be green, and a customer can still be unable to send money. Infrastructure monitoring tells you the parts are running. It does not tell you the service works.
Digital experience monitoring closes that gap by continuously performing the transaction a real customer would perform, on the channel they would use, and checking whether it succeeded. TecFlax is a Breakpoint End-to-End Monitoring and Observability Solutions Partner, and Apogee, the platform behind that partnership, drives scripted transactions over real GSM hardware, so USSD, SMS and STK are tested the way a customer on a basic handset experiences them.
That matters more here than in most markets. Mobile money and mobile banking in Kenya run on exactly those channels, and they are the channels conventional monitoring tools reach least well.
Underneath the experience layer we run the ordinary discipline too. We implement five monitoring platforms: Datadog, Prometheus, Grafana, Zabbix and Nagios, plus centralised logging on the Elasticsearch, Logstash and Kibana stack, with thresholds tuned so that an alert still means something six months after go-live.
They are not five competing answers to one question, and we will not pretend otherwise. Prometheus and Grafana are one stack, not two products: Prometheus stores and alerts, Grafana draws. Zabbix and Nagios are alternatives to each other for classic infrastructure, network gear and appliances, and Zabbix is the modern one. Grafana sits on top of any of them, including monitoring you already own. Datadog replaces the lot as a hosted service, at a subscription cost and with your telemetry leaving the country. Which of those is right depends on your estate, your data governance and your budget, and we would rather have that conversation honestly than sell you the one with the best margin.
Find out what your customers are actually experiencing
Tell us which services matter most and which channels your customers reach them on, and our engineers will come back with a monitoring proposal.