Databases that handle the load. Built on SQL Server.
MSSQL development services for organisations running SQL Server on-prem, in Azure, or both. Query tuning that cuts report times from minutes to seconds. Always On configurations that actually fail over cleanly. Azure SQL migrations that don't surprise you with worse performance on a bigger bill. We work inside SQL Server — not just around it.
Four things every SQL Server environment actually needs.
The SQL Server environments we audit always have the same problems — missing indexes on columns that get filtered millions of times daily, `SELECT *` in production queries, maintenance plans that haven't run in months, and an Always On group that's never been tested for actual failover.
Performance01
Query tuning fixes 90% of SQL Server "slowness."
Missing indexes, outdated statistics, implicit conversions, parameter sniffing — the usual suspects. We profile with Query Store and wait stats before touching anything. The database is rarely slow — the queries hitting it usually are.
Architecture02
Proper indexing strategy, not indexes on everything.
Covering indexes for critical queries. Filtered indexes for common WHERE clauses. Index maintenance that balances fragmentation against I/O cost. Not 80 indexes on one table because someone kept clicking "Create Missing Index."
High availability03
Always On that's tested, not just configured.
AG groups with synchronous commit for zero data loss. Async for DR across regions. Failover tested quarterly — not assumed to work because it was set up two years ago. Read-only replicas for reporting that take load off production.
Migration04
Azure SQL with a performance baseline, not a blind lift.
Performance baselined before migration. Service tier matched to actual workload, not default. Hybrid Benefit applied to reduce licensing cost. The migration that performs the same or better — because we measured before and after.
When SQL Server, when not
SQL Server is a tool. Not the only database.
SQL Server is the most widely deployed enterprise relational database on Windows. It's also expensive to license and overkill for many workloads. We'll tell you on the first call if MSSQL is the right fit.
STAY ON SQL SERVER WHEN
You're a Microsoft shop — .NET, Azure, Windows Server — where SQL Server integration is native
Enterprise workloads — SSRS, SSIS, SSAS, cross-database queries, linked servers — that PostgreSQL can't match without rewrites
Always On Availability Groups for HA/DR — proven, battle-tested, well-understood by your ops team
Regulatory or vendor requirements mandate SQL Server specifically
CONSIDER ALTERNATIVES WHEN
Greenfield with no Microsoft dependency — PostgreSQL is free, open-source, and handles most OLTP workloads identically
Cost is the primary constraint — SQL Server licensing adds up fast, especially without Azure Hybrid Benefit
Your workload is read-heavy analytics — a data warehouse (BigQuery, Redshift, Snowflake) may serve it better
WE SAY NO WHEN
"SQL Server because we always have." That's habit, not architecture.
"Enterprise Edition for a 10GB database." Standard Edition handles that. Save the licensing cost.
"Migrate to Azure SQL in two weeks." Baseline first, migrate second.
What we build on SQL Server
Six product surfaces. One quality bar.
The shapes of SQL Server development work we deliver most. Each tuned, maintained, and production-ready.
S01
Query performance tuning
Query performance tuning
Query Store analysis, wait stats profiling, index recommendations, execution plan review. The queries that take minutes reduced to seconds — with measured before and after.
QUERY STOREWAIT STATSEXECUTION PLANSINDEXES
S02
Always On high availability
Always On high availability
AG groups configured for your RPO/RTO. Synchronous for zero data loss, async for cross-region DR. Failover tested. Read-only replicas for reporting offload.
ALWAYS ONAG GROUPSFAILOVERREAD REPLICAS
S03
Azure SQL migration
Azure SQL migration
On-prem to Azure SQL Database, SQL Managed Instance, or SQL Server on Azure VM. Performance baselined, service tier matched, Hybrid Benefit applied. Zero-downtime cutover via DMS.
AZURE SQLSQL MIDMSHYBRID BENEFIT
S04
Database schema design
Database schema design
Normalised schemas for transactional workloads. Proper data types — no `nvarchar(max)` for a 10-character code. Foreign keys, constraints, and check conditions enforced at the database level — not just in app code.
SCHEMA DESIGNCONSTRAINTSDATA TYPESNORMALISATION
S05
SSIS, SSRS & ETL pipelines
SSIS, SSRS & ETL pipelines
Data integration with SSIS packages, reporting with SSRS, scheduled ETL that runs reliably and fails loudly. Replacing manual Excel exports with automated pipelines.
SSISSSRSETLSQL AGENT
S06
SQL Server upgrades & maintenance
SQL Server upgrades & maintenance
SQL Server 2014/2016 to 2022. Compatibility level testing, deprecated feature remediation, maintenance plan configuration. The current database stays online throughout.
SQL 2022COMPATIBILITYMAINTENANCE PLANSBACKUP
The playbook
Patterns we ship on repeat.
SQL Server patterns from real production databases — not certification exam answers.
P01
Query Store for continuous tuning
Query Store enabled on every production database. Regressed queries flagged automatically. Plan forcing for critical queries. Performance regression caught before users notice.
P02
Index strategy, not index sprawl
Covering indexes for the top 20 queries by resource consumption. Unused indexes dropped. Index maintenance jobs that rebuild only when fragmentation justifies the I/O cost.
P03
Proper backup strategy
Full + differential + transaction log backups on a schedule matched to your RPO. Restores tested quarterly — not assumed to work because backups complete.
P04
Wait stats before tuning
`sys.dm_os_wait_stats` and `sys.dm_exec_query_stats` analysed before any change. We tune what the database is actually waiting on — not what someone guesses is slow.
P05
Parameterised queries everywhere
No string concatenation in SQL. Parameterised queries from every application layer. SQL injection eliminated at the data access layer — not at the WAF.
P06
Azure SQL right-sizing
Service tier matched to actual DTU/vCore consumption — not to the default. Elastic pools for multi-tenant databases. Serverless tier for variable workloads. Cost matched to usage.
Signature case
A reporting database, tuned from 14-minute reports to 8-second reports.
An enterprise reporting database on SQL Server 2019 — critical monthly report taking 14 minutes, 340 missing indexes identified by Query Store, statistics updated once a year, and an Always On replica that hadn't been tested for failover since setup. Tuned indexes, updated statistics, rewrote the top 10 worst-performing queries, and tested failover in 6 weeks. Monthly report runs in 8 seconds. Failover tested and verified.
A typical SQL Server engagement. We tune or migrate database by database — the current environment stays live while we work.
W01
Audit + RFC
Two senior SQL Server DBAs. Query Store analysis, wait stats profiling, index audit, HA configuration review. A ranked, dollarized RFC.
W02–03
Foundation + quick wins
Top 10 worst queries tuned, critical missing indexes created, statistics updated, maintenance plans configured. Measurable improvement in week two.
W04–08
Systematic tuning or migration
Database by database. Indexes optimised, queries rewritten, Always On tested, Azure SQL migrated with baseline comparison. Your applications keep running.
W09+
Handoff
Query Store monitoring configured. Maintenance plans automated. Runbook handed to your team — or we stay on retainer.
Stack
Tools we reach for first.
Our default SQL Server development stack — picked for production databases.
On-prem vs Azure SQL, performance tuning approach, Standard vs Enterprise — the questions we get on every SQL Server discovery call.
Azure SQL Managed Instance for most enterprise migrations — near-100% SQL Server compatibility, no OS management, built-in HA. Azure SQL Database for simpler workloads. Stay on-prem when compliance, latency, or vendor requirements mandate it. We baseline performance before migration so there are no surprises.
Wait stats first to identify what the server is actually waiting on. Then Query Store for the worst-performing queries. Then index strategy based on actual query patterns. We never tune blind — every change has a measured before and after.
Standard for most workloads. Enterprise when you need Always On with multiple read replicas, columnstore indexes, or advanced security features. We audit your actual feature usage and recommend honestly — not upsell.
Yes. The engineers who write the RFC do the tuning. No handoff mid-engagement. Direct access throughout.
Yes. We audit what's there, tune what's slow, fix what's broken, and migrate what needs moving. No rip-and-replace.
Founder-direct
Tell us whatyou're building.
Thirty minutes with the founder. We'll bring a senior SQL Server DBA, the relevant playbook, and a candid read on whether your database needs tuning, migration, or a different platform entirely.