AI for marketing teams
We build AI for marketing teams that produce more than they can review: drafts written against your brand guidelines, reporting that brings every channel into one view and first-party data your tools can use. Editors and marketers approve everything before it is published or sent.
Where the work gets stuck.
More content than the team can review
Every channel wants more posts, pages, emails and ad variants. Drafts from general AI tools miss the brand voice and add claims nobody approved, and editors spend the time saved on corrections.
Reporting that starts with exports
Results sit in Google Ads, Meta, LinkedIn, GA4 and HubSpot, each with its own attribution. Someone copies them into a spreadsheet every Monday before the team can discuss what worked.
First-party data spread across tools
Customer, consent and behaviour data live separately in the CRM, the email platform and the website. Segments are rebuilt by hand for each campaign.
How we improve marketing teams.
Each one is scoped around your systems and rules, and each one keeps a person in charge of the decisions that matter.
Content drafts written to your guidelines
From an approved brief, an agent gathers input from product documentation, interview notes and past pieces and writes a blog post, landing page or email in your brand voice. The draft lands unpublished in WordPress, Webflow or HubSpot, and an editor approves it.
Brand and claims check
Drafts from your team, freelancers or AI are checked against your brand guidelines, tone rules and approved claims library, with each flag linked to the rule it concerns. The editor or legal reviewer decides what to change.
Campaign reporting across channels
Spend and results from Google Ads, Meta Ads, LinkedIn Ads, GA4 and HubSpot land in BigQuery under one campaign naming scheme, with a Looker Studio report on top. A weekly summary is drafted from the figures, and a marketer edits it before it is shared.
Search across your site and resources
Visitors find the right article, guide or product page when their wording differs from yours, through embeddings combined with keyword matching. Searches that return nothing are logged, so the content team can see what to write next.
First-party data in your marketing tools
Customer, product usage and consent data from the CRM and your product sync to Klaviyo, Mailchimp, HubSpot and ad audiences, with consent status on every contact. Marketers build a segment once, and people who opted out stay excluded in every tool.
Event and partner list imports
Lead lists from events, webinars and partners are cleaned, matched to existing contacts, checked for consent and loaded into the CRM with the right campaign attached. Marketing ops review the records the automation cannot match.
Built around the rules.
What we design for from the first week. Your legal and compliance people keep the final word.
EU AI Act transparency for content
Since 2 August 2026, chat assistants must tell people they are AI, and synthetic images, audio and video that could pass as real must be labelled. AI-generated text published to inform the public on matters of public interest needs a label unless a person reviewed it and carries editorial responsibility, so we log that review.
Advertising claims and the Reclame Code
Dutch advertising must meet the Nederlandse Reclame Code, and the ACM acts against misleading commercial practices, including environmental claims without proof. Generated copy is limited to claims in your approved library, and any new claim goes to a person.
Copyright in generated content
EU copyright protects work that reflects human creative choices, so purely machine-generated text or images may have little protection. We record the sources, prompts and edits behind each piece and check that fonts, stock and reference material are licensed for the use.
GDPR and ePrivacy for personalisation
Tracking, email marketing and ad audiences need a valid legal basis, which for cookies and marketing email is usually consent. Data flows we build carry consent status with every contact, and GA4 and ad tags follow your consent banner.
Works with what you run.
If a system has an API, a database, an export or an inbox, we can build on it. These are the ones we meet most.
- HubSpot
- Salesforce
- Google Ads
- Meta Ads Manager
- LinkedIn Campaign Manager
- GA4
- Google Search Console
- BigQuery
- Looker Studio
- Klaviyo
- Mailchimp
- WordPress
- Webflow
- Figma
Where to start.
Drafts for one content type
We pick one content type with a clear format, such as customer stories or product update emails, and build the eval set from pieces your editors already approved. Editors rate each draft and we track how much they change, so you decide on the next content type with those results in hand.
Talk it throughWhat it includes
- Brand guidelines and claims library set up as checks
- Brief template and approved sources
- Drafts delivered to your CMS as unpublished
- Editor ratings and edit tracking per draft
Guides.
AI readiness checklist: what to answer before your first agent goes live
The questions to answer before your first AI agent handles real work, from process and data to approvals, monitoring and ownership.
10 min read
Build vs buy software: a decision guide for operations and product teams
A practical guide to choosing between off-the-shelf and custom software, with a cost comparison, a decision matrix and a checklist for your team.
10 min read
Further reading.
LLM evals: how to test AI features before every release
A practical approach to LLM evals: build a test set from real cases, combine code checks with model grading, and block releases that regress.
5 min read
Embeddings and semantic search: a practical guide for product teams
How embeddings power semantic search, how to choose a model and a vector store, and how to measure whether your results actually improved.
5 min read
Human-in-the-loop AI agents: design patterns for production
The patterns we use to run AI agents safely in production: approval checkpoints, confidence thresholds, tool permissions, audit trails and fallbacks.
4 min read
Common questions.
Well enough for a first draft your editors can work with, when it starts from your guidelines and a set of pieces you already approved. We measure this on an eval set your editors score, and an editor approves every piece before it goes live.
Google does not penalise content for being AI-generated. Its guidance looks at whether content is helpful to readers, and its spam policies target pages mass-produced mainly to rank. We design content workflows around your own expertise, sources and editorial review.
Sometimes. Under the EU AI Act, chat assistants must say they are AI and realistic synthetic images, audio and video must be labelled, while text on matters of public interest needs a label unless a person reviewed it and takes editorial responsibility. Ad platforms such as Google and Meta add their own disclosure rules for some ad types.
For individual writing tasks, yes. A build pays off when every draft has to follow your guidelines and claims library, use your own data and arrive in your CMS or reporting with a review step your team can see.