The Australian Government has made a clear decision about how it wants to work. The core work of the public service should be done by public servants, and outside expertise should be used where it is specialist, time-limited, and builds APS capability. That is the thinking behind the APS Strategic Commissioning Framework.
It is easy to read a policy like this as a restriction on suppliers. We read it differently. It is a description of what a healthy relationship between an agency and a specialist partner should look like, and it happens to describe the way we have been working for years.
Why keeping the work in-house is the right call
A web team that owns its platform makes better decisions, faster. It knows the content, the audiences, and the politics of its own department. In-house knowledge and knowhow helps maintain consistency and quality in a site as it develops over the years. Ultimately, it is up to the website owner to define the standards and principles that will take the site forward and that requires an in-house team with organisation and domain knowledge.
The framework puts that into policy. Agencies now ask of every engagement: is this our core work, and if we do go outside, how does the engagement make us more capable than we were? Those are good questions. Any supplier who finds them uncomfortable should think about why.
Measuring the right thing
In September, the Australian National Audit Office reviewed how the framework is tracking. Much of the public discussion since has been about dollar figures: how much work has moved in-house, and how much is still spent outside. Dollars are easy to count, but they are a proxy. What the government actually wants is capability. An agency can reduce its contractor spend and be no more capable than it was. It can also spend wisely on a specialist and come out far stronger.
So the more useful question for any engagement is what it leaves behind. Does the team understand its platform better than it did? Can it do more of the work itself next time? If that is the measure, the spending figures tend to deliver value.
Where specialists still belong
There is work that no agency should try to build a permanent team around:
- Platform and solution architecture.
- Complex migrations.
- Search engineering
- Privacy-first personalisation.
- Accessibility at the component level, where a mistake gets replicated across a thousand pages.
- Keeping a site current as Drupal moves forward.
This is deep, intermittent work, and it is exactly what the framework expects agencies to source from the market.
The trick is to buy that expertise in a form the agency can keep. Not as developers on a retainer who become the only people who understand the site, but as a well-built engine, documented and handed over, that the in-house team runs and extends.
That distinction is the whole game. Morpht builds the engine; the agency runs the site.
Crystallising specialist work into a product
Most of what we have learned about government websites over the last decade now lives in Convivial for GovCMS. It is our open-source starter site for the GovCMS platform: content types, components, layouts, search, and a theme that a non-developer can restyle from settings. It has passed an independent WCAG 2.2 AA audit. It ships with Convivial Personalisation, a personalisation engine that runs in the visitor's browser and sends no data anywhere.
The point of packaging our expertise this way is that an agency gets the specialist work without the specialist dependency. Editors assemble rich pages with components. Themers set colours and typography without touching code. The governance and accessibility rules sit underneath, so a content team can move quickly without being able to break anything that matters. Our engineering is in the product. Day-to-day delivery is in your hands.
A department that stopped needing us for new sites
For a federal government department, we built what the team calls a "Whitelabel" engine on Convivial for GovCMS: a governed set of components, templates, and configuration from which the department's own web team can stand up a new site without going back to a development partner. There is no code involved. A new site is a matter of configuration: pick the components, set the colours and typography, add the content, and publish.
The practical result is a team that is more productive and more skilled. New sites are a configuration exercise rather than a procurement exercise. Because every site starts from the same tested components, the quality is consistent, and the maintenance load is predictable. And because the team does the work itself, it has learned the platform properly, which is the kind of capability the framework is trying to build.
Where does that leave us? Doing the work the department cannot sensibly do in-house: extending the engine, adding new components to the shared library, and handling the occasional hard integration. Fewer engagements, higher value, and a client who is better off between them.
Handover is a phase
Knowledge transfer is the part of the framework most often promised and least often delivered. "Documentation will be provided" is not a handover.
Our handover starts on day one, because the architecture decides whether the agency can run the site. If we build bespoke code where configuration would do, we have built a dependency. So we favour reusable components and configuration the web team can see and change. If a new feature is needed, it is built in a reusable way. If a bug is found, it is fixed at the source.
Then we train, and we document. Training is hands-on and aimed at the people who will actually run the site. Documentation covers how the site is put together and how to operate it, written for the web team in a task-focused manner. At the end there is a plain statement of what the team now runs itself, what remains genuinely specialist, and what we would suggest next.
If that sounds like we are working ourselves out of a job, we are, for the routine work. That is the point. We would much prefer to build features and capabilities that pay off many times over in the future.
The relationship we want
The Strategic Commissioning Framework describes a public service that is confident in its own capability and deliberate about where it brings in outside help. That is a public service we want to work with.
Morpht builds the engine. You run the site. We would rather be the partner you call for the hard problems than the one you cannot stop calling.
If you are planning your next GovCMS site, or wondering how to make the one you have easier for your own team to run, we would be glad to talk it through. Get in touch with Morpht.
