Platform specialists
We know
your stack
Every platform makes something easy and something else almost impossible. Knowing which is which is most of the job.
Ten platforms and three Ai engines. We have shipped technical SEO on all of them, which means we start on the real problem instead of learning your CMS on your budget.
A recommendation that cannot be implemented on your stack is not a recommendation, it is homework. We have shipped on all thirteen of these, so the audit you get back is a list of things that can actually be done, in the order they should be done.
Where the platform genuinely blocks something, we say so and price the alternative - a workaround, a plugin, a partial replatform, or living with it. That conversation is cheaper in month one than in month nine.
What we check first
How it renders
Server-side, static or client-side, and what the crawler actually receives. On a JavaScript-heavy build this single question explains most of the missing traffic.
How it multiplies URLs
Filters, sorts, variants, pagination, tags and session parameters. Left alone, a 400-product store becomes a 40,000-URL crawl trap and the pages that sell never get indexed.
How much control you have
Templates, head markup, redirects, robots and schema. This determines whether a fix is a ten-minute change or a development project, and it is why two identical audits get different roadmaps.
13
Platforms shipped on
Ten content platforms and the three Ai engines that now read them.
90d
To first uplift
Technical work is the fastest-moving part of any search programme.
0
Rankings lost
The only acceptable target for a planned migration, and the one we work to.
312%
Average growth
Across the search programmes built on top of that foundation.
Replatforming
Change the stack, keep the rankings
The riskiest month in a site’s life
Most traffic collapses after a migration are avoidable and entirely predictable: lost redirects, changed URL patterns, a staging robots.txt shipped to production, or content quietly dropped because nobody mapped it. We run migrations as a checklist, not a hope.
- Full URL inventory and redirect map before launch
- Parity check on content, metadata, schema and internal links
- Crawl, index and traffic monitoring through the first month
- 0
- Acceptable ranking loss
- 100%
- Of URLs mapped before launch
- 30d
- Of post-launch monitoring
Built on it,
or found in it
Whether you are fixing what you have or moving to something new, the sequence is the same and the order is not negotiable
We crawl the site the way a search engine does, then compare what the crawler receives against what a browser renders. On most modern builds the gap between those two is the single biggest finding in the report.
Every template, every parameter and every URL the platform can generate, inventoried. This is what turns a vague "the site has duplicate content" into a specific list of rules to write.
We change the template, not the page. One fix applied to a product template corrects four thousand URLs and keeps correcting them for every product added afterwards.
Crawl stats, index coverage, Core Web Vitals and rankings watched through the change window. Platforms update themselves, plugins regress, and a fix that is not monitored is a fix with an expiry date.
The problems we find on nearly every stack
Different platforms, remarkably similar failures. These five account for most of the traffic left on the table.
JavaScript the crawler never runs
Content that appears instantly in a browser and never in the index. Client-side rendering is not automatically fatal, but it has to be a decision rather than a default nobody reviewed.
Filters generating infinite URLs
Faceted navigation is the most common crawl trap in e-commerce. The crawler spends its budget on colour-and-size permutations while your actual category pages go unseen for weeks.
Performance regressions nobody noticed
A site is fast on launch day and slow eighteen months later, one plugin and one tracking script at a time. Core Web Vitals decay is gradual, which is exactly why it goes unreported.
Redirect chains from old migrations
Three or four hops from the original URL, each one leaking a little authority. Every replatform since 2015 is usually still visible in the chain if you follow it.
Schema that validates but says nothing
Plenty of sites emit technically valid structured data that describes the template rather than the business. Language models read that, and it is a large part of why an Ai answer gets a brand wrong.
Built for the engines that read your stack
What clients say
Australian businesses whose platform was the thing holding them back.
Brain Stream
Find out what your stack is costing you
A free technical audit on your real site, on your real platform, with the findings ordered by what to fix first.
A real crawl of a real site. No sales call required.
Questions about platforms and builds
Running something not listed? Tell us what it is.
-
The one your team will actually maintain. WordPress and headless builds give the most control, Shopify and Webflow trade some of it for speed and stability, and the wrong answer is switching platforms to chase a ranking. We would rather fix what you have.
-
On most platforms, yes - template changes, schema, redirects, performance work and configuration. On custom or headless builds we usually specify the change precisely and work alongside your developers, because they know the codebase and we know the requirement.
-
No, and now is much better than after launch. The cheapest moment to get URL structure, rendering strategy and redirects right is while the build is still in staging. After launch it becomes a recovery project.
-
Regularly. We scope exactly who owns what, hand over specifications rather than vague recommendations, and check the implementation afterwards. The friction usually comes from ambiguity, so we remove it in writing upfront.
-
The technical foundation is largely shared - if a crawler cannot read it, a language model cannot quote it. What is added is entity clarity, structured data that describes the business rather than the template, an llms.txt, and presence on the third-party sources those models already trust.
-
Yes. Staging audits are some of the most valuable work we do, because everything found is still cheap to change. We will need access and a note on what is intentionally incomplete.
Areas we service
Brain Stream
Sydney NSW Australia
5.0 on Google