Four products, all of them ours.
Three are carrying production traffic and one is still being built. The front page shows what the work looks like. This is the problem underneath each one, what was actually built, and where it stands today.
Flagship product · Operating since 2013
MyOwnConference
Browser-based webinar and video conferencing infrastructure, developed and operated by AKOVANA since 2013.
Problem
A webinar platform is judged in the first thirty seconds of a live session by an audience that will not come back if the video stalls. That standard has to hold on the presenter’s laptop, on an attendee’s phone, and across a connection neither of them controls.
What we did
Product, interface, real-time layer and infrastructure are one system rather than four suppliers. Broadcasting runs on our own machines in two European data centres with three CDN providers in front of them, which is what makes the latency ours to improve rather than someone else’s to explain.
Where it stands
Signal latency is under one second, down from the earlier sub-two-second stack. A next-generation dashboard is in development, and the platform has been in continuous operation since 2013.
Own product · Operating
RUA•Watcher
Email authentication and domain infrastructure monitoring, graded per domain and explained down to the record that has to change.
Problem
DMARC reports arrive as raw XML, separately from every mailbox provider that saw your mail. Reading them by hand is impractical, and most tools that parse them stop at a chart, which does not say which sending source broke or what to change in DNS.
What we did
Every domain gets a grade and then an explanation of that grade. Each check is its own signal with its own state, from MX and SPF through DMARC, DNSSEC and MTA-STS, so a failure points at one record rather than at the domain as a whole. Accounts with several domains see the portfolio first.
Where it stands
Running against our own domains and those of the platforms we operate, which is the only test set that argues back.
Own product · Operating
Statusle
Uptime monitoring, incident timelines and the public status page they feed, as several views over one set of records.
Problem
An incident interface is used under pressure by the person who is also fixing the thing. Every step between noticing a problem and telling customers about it is a step that gets skipped, and then the status page stays green while the product is down.
What we did
Checks, incidents, alerts and public pages are views over the same records rather than four separate features, so the page customers read cannot drift from what the monitors saw. The overview answers the only question worth asking at a glance.
Where it stands
Running. It watches the platforms we operate ourselves, which is where its own requirements keep coming from.
Own product · In active development
OpenMetrica
Privacy-first analytics for teams that need useful acquisition and behavioural data without cookies or fingerprinting.
Problem
Analytics that respects people usually stops at a page-view counter, and analytics that answers product questions usually pays for it with a third-party script and a cookie banner. Teams end up running both and trusting neither.
What we did
The product is built around what can be answered without identifying anyone, which makes the data model the first design decision rather than a later correction. We run it on our own sites, so the first demanding customer is us.
Where it stands
In active development. Product, interface and data architecture are ours, and the privacy position is a constraint on the build rather than a line in the marketing.
Let’s discuss the system behind your next product.
Tell us what the system has to do and where it has to hold. You get an engineering answer, not a sales deck.
contact@akovana.com