Structure first
Content modelling is settled before anything is built, because retro-fitting fields onto live content is the expensive way round.
Serious content architecture — and an honest view of whether you need it.
Drupal is the right answer for genuinely complex content structures and strict editorial workflows. It is an expensive answer for a brochure site. Most of my work here is upgrades, custom module work, and migrations in both directions.
Drupal was built around structured content, and that remains the real reason to choose it. Content types with typed fields, taxonomy that can model genuinely complicated relationships, and Views – a query builder that turns that structure into listings without anyone writing SQL. If your content has a shape more complicated than "posts and pages", this is where that shape lives comfortably.
The second reason is editorial control: revision histories, moderation states, workflows where content moves between roles before publication, and permissions granular enough for organisations where who may publish what is a real question rather than an afterthought.
The cost is complexity, and it is not small. Developers are scarcer and more expensive, major version upgrades have historically been substantial projects rather than routine updates, and a straightforward site carries all that overhead for no benefit. A great many Drupal sites exist because someone chose it for a brochure site in 2014 and it has been costing more than it should ever since.
So the work splits three ways. Keeping a complex site that genuinely needs the platform healthy and current. Building custom modules and Views where the structure justifies it. And migrating sites that never needed it onto something cheaper to run, with the content structure mapped properly rather than flattened.
Moving to current major versions, with contributed module availability checked before anything is promised.
Security support without discovering mid-project that a key module has no successor.
Modules written to the platform’s own APIs and coding standards, not bolted on beside them.
Functionality that survives upgrades and that another developer can read.
Content types, fields, taxonomy and relationships modelled around how your organisation actually works.
Structure that still fits in three years, instead of fields added ad hoc until nobody understands it.
Complex listings, filters and feeds built with Views rather than hand-written queries.
Editors can adjust what a listing shows without a developer in the loop.
Moderation states, revision handling and role permissions set up for real approval chains.
Control over what gets published, enforced by the system rather than by email.
Structured content moved to WordPress or a newer version, fields mapped and URLs redirected.
A cheaper platform to run, without losing structure or search visibility.
Mostly about respecting a system other people will have to maintain.
Content modelling is settled before anything is built, because retro-fitting fields onto live content is the expensive way round.
Custom code written to the platform’s own conventions, so the next developer can read it without archaeology.
If a well-maintained module already does it, that gets used. Custom code is a maintenance liability, not an achievement.
Configuration and code tracked and deployed, not clicked into production and hoped for.
If your site does not need this platform, you will hear that, with the cost of moving to something simpler.
Content model, custom modules and configuration written down, because complexity without documentation is a trap.
Organisations with real content complexity, and organisations who were sold it and do not have any.
Many departments, many editors, and content that must move through approval before it appears.
Accessibility obligations, multilingual requirements and audit trails on published content.
Large archives where relationships between content are the point rather than a nice extra.
Running a release past end of life, where the upgrade path needs planning rather than attempting.
A brochure site on an enterprise platform, costing enterprise money for no benefit.
A platform your team does not work in, on a site that still needs supporting.
Bangalore has more of this expertise than most Indian cities, largely because enterprise and public-sector projects here adopted the platform early. It is still specialised work, and the gap tends to be at the small end – organisations with one site, no in-house developer, and no appetite for agency retainers.
That is the work I take: a single site that needs looking after properly, or an honest assessment of whether it should stay on the platform at all. I work with organisations across Bangalore and India, and with clients in the UK, USA, Canada, the UAE and Singapore.
Working with organisations across the city and India.
Available without an agency retainer.
Told plainly if the complexity is not earning its keep.
India, the UK, USA, Canada, the UAE and Singapore.
Serving Chennai, Bangalore, India, Canada, UK, USA, Dubai, Singapore.
The sectors where the complexity is doing real work.
Typically needs: Dozens of departments publishing independently under one brand and one set of rules.
How I help: Content types per department, role-based permissions, and workflows that enforce review before publication.
Typically needs: Accessibility standards, several languages and a record of who changed what.
How I help: Accessible templates, multilingual content, and revision history that satisfies an audit.
Typically needs: Deep archives where taxonomy relationships drive discovery.
How I help: Taxonomy modelled properly, with Views generating listings editors can adjust themselves.
Typically needs: Multiple facilities, practitioners and services with shared and local content.
How I help: Entity relationships that let one practitioner appear correctly across several facilities.
Typically needs: Programmes, regions and reporting, often across countries.
How I help: Structured programme content with regional filtering, and a hard look at whether the platform is warranted.
Typically needs: Tiered access, member directories and gated resources.
How I help: Permission structures matched to membership tiers, with a directory built on real entity data.
Six stages. The content model is settled before anyone builds anything.
I audit what exists: core version, every contributed and custom module, the content model, the configuration management setup and what has been patched.
You get: A written audit, with an honest view of whether the platform suits the site.
Why it matters: These sites accumulate custom code, and nobody should plan around a codebase they have not read.
The content model comes first – content types, fields, taxonomy and the relationships between them – then the module strategy and the upgrade or migration path.
You get: A content model and technical plan you have signed off, with a fixed quote.
Why it matters: The content model is the one decision that is expensive to change afterwards. Everything else is negotiable.
Templates designed against the real content structure, at phone width first, with the editorial experience designed too rather than inherited.
You get: Designs for the key content types and the editing experience.
Why it matters: On a site with many editors, the admin experience matters as much as the public one.
The build: theme, custom modules, Views and configuration, all in version control and deployed rather than clicked into place.
You get: A working site on a staging environment, deployable repeatedly.
Why it matters: Configuration changed by hand in production is why so many of these sites cannot be upgraded.
Testing across devices, every permission role exercised, workflow states verified, accessibility checked, and for migrations the full redirect map tested.
You get: A tested site, an accessibility report and a verified redirect map.
Why it matters: Permission and workflow errors are invisible until the wrong person publishes something.
Launch, editor training against the real workflow, and written documentation of the content model and custom code.
You get: A live site, trained editors and documentation that outlives me.
Why it matters: A complex site with no documentation is the reason the previous developer was irreplaceable.
The platform, what it runs on, and where sites usually move.
Scoped from the audit, not offered as a package.
Well-modelled content is easier to mark up accurately, because the page already knows what each piece of information is. The common problems are elsewhere: path aliases left inconsistent, taxonomy pages generating thin listings, and upgrades changing URL patterns without anyone mapping the old ones.
Migrations carry the real risk. Moving off the platform without a complete URL map is how sites lose visibility they spent years earning.
A site where every fact lives in a named field rather than inside a blob of HTML is inherently easier for anything automated to read. That is the one genuine advantage this platform has here, and it survives a migration if the fields are mapped rather than flattened.
The practical reasons, rather than the adjectives.
A large share of these sites are brochure sites on an enterprise platform. If that is yours, you will hear it, with the cost of moving to something cheaper to run.
The content model is the decision that is expensive to change later. It gets settled before any building starts, with you in the conversation.
Custom code is a maintenance liability someone inherits. If a maintained module already does the job, that is what gets used.
Most of this expertise sits inside agencies on annual contracts. A single site needing occasional proper attention is work I am happy to take.