- Market
- Saudi Arabia
- Written for
- CTO
- Decision
- Build a system on PostgreSQL and make the platform decisions around it well
Information security for software delivery

ISO/IEC 27001
ISO/IEC 27001 certified
Why it wins
How a database nobody owns became the one serious teams reach for
We do not have an office in Saudi Arabia
We serve organisations there from our Riyadh office and our engineering hubs in Lahore and Karachi, travelling for the parts of an engagement that need to happen in person.
PostgreSQL earned its position the slow way: decades of unglamorous correctness, a licence that permits anything, and a habit of absorbing new workloads — geospatial, time series, full-text search, and now vector search for AI retrieval — as extensions rather than as reasons to buy another product. There is no company behind it to negotiate with, and none to be a partner of; it is developed in the open by a community, which is precisely why every major cloud offers it and no invoice ever arrives for the software itself. When we recommend it, that is an engineering judgement, not a reseller margin.
The consequence is that choosing PostgreSQL is the cheapest decision you will make about it. Everything that determines whether it serves you well is downstream: a schema that models the business rather than the first user story, extensions chosen for their maintenance reality rather than their demo, a replication topology matched to how much data loss and downtime the business can actually tolerate, and a hosting decision made on facts about your team and your data. That downstream work is what this page is about.
The working surface
The core is one decision; the four rings around it are the rest
A production PostgreSQL system is the engine plus everything fitted around it. These are the parts we design and build, grouped by the decision each one answers.
The data model itself
- Schema design
- Constraints
- Migrations
- Partitioning
Capabilities added as extensions
- PostGIS
- pgvector
- Full-text search
- TimescaleDB
Staying up and losing nothing
- Streaming replication
- Logical replication
- Patroni
- PgBouncer
Where it runs
- Amazon RDS
- Azure Database
- Cloud SQL
- Self-managed
The hosting decision
Managed PostgreSQL or your own — the trade behind the monthly bill
Because the software is free, the real cost of PostgreSQL is operational, and the hosting question decides who pays it. Neither answer is right in general; one is usually right for a specific workload, team and residency constraint.
| Option | Who carries the operations | Extension and version freedom | Cost shape over time | Residency and control | Best for |
|---|---|---|---|---|---|
| A managed cloud service | The provider patches, backs up and fails over; your team consumes an endpoint. | A curated subset of extensions and versions, on the provider’s schedule, not yours. | A predictable monthly bill that climbs with instance size, storage and replicas. | Data sits in the provider’s region under configurable, contractual controls. | Teams that want the database to be someone else’s pager. |
| Self-run PostgreSQL | Your team — or a partner engaged for it — owns patching, backups, failover and the on-call rota. | Any extension, any version, upgraded when the workload allows rather than when a provider decides. | Cheaper compute on paper, paid for in engineering time and operational cover. | Full authority over where data lives and who reaches it — and full responsibility for proving both. | Workloads with residency constraints or extension needs a managed tier will not host. |
Our role
What we engineer into PostgreSQL and what your team owns on top of it
- 01
The schema, treated as the asset
Application code is rewritten every few years; the data model outlives all of it. We design the schema, constraints and migration discipline for the system you will have in year five, not just the feature shipping this quarter.
- 02
Replication chosen, not defaulted
How many replicas, synchronous or not, and what fails over automatically are answers to how much data loss and downtime the business tolerates. We put that question to you plainly and build the topology the answer justifies.
- 03
A hosting decision argued on facts
Managed or self-run is decided from your workload, your residency obligations and the team you actually have — and because no vendor pays us, the recommendation has nothing behind it except the reasoning we show you.
Which cloud you sign with, and where your data is allowed to live, are your decisions; we build the schema, the replication and the tooling so that whichever way you decide, the database is not the thing that locks you in.
Common questions
What buyers ask
PostgreSQL is free — so what does a PostgreSQL project actually cost?+
The licence costs nothing and always will; the project’s permissive licence allows unrestricted commercial use. What you pay for is everything the licence does not include: the schema and migration design, the extension and replication decisions, the hosting bill if you choose a managed service, and the operational cover if you do not. That trade is usually favourable — you spend on engineering that fits your system instead of on a licence that fits a vendor’s price list — but it is a real spend, and pretending the database is free end-to-end is how it ends up unmanaged.
When would you advise against PostgreSQL?+
Honestly, rarely — but the cases exist. If your organisation already runs deep on a commercial database and the skills, tooling and contracts around it, migrating for ideology is a poor trade. If a workload genuinely needs a horizontally distributed system across regions from day one, a database built for that shape may fit better than stretching PostgreSQL to it. And a cache or a queue should be a cache or a queue. What we push back on is the opposite error: adopting a niche database for a workload PostgreSQL handles well, and inheriting a small community and a thin hiring market with it.
Are extensions like pgvector safe to build a product on, or a lock-in risk?+
Judge each extension by three things: whether it is actively maintained, whether the managed services you might ever run on support it, and how hard your usage of it would be to unwind. PostGIS and pgvector pass all three comfortably — mature, hosted by every major managed tier, and expressed through SQL you could migrate. The riskier pattern is an obscure extension that only self-run PostgreSQL can load, because it silently makes the hosting decision for you. We flag that dependency at design time, when changing course costs a conversation rather than a migration.
In this cluster
Next step
Scope it with us
A design session that ends with a schema direction, an extension shortlist and a hosting recommendation your team can defend.
