Can a website actually solve that?
Before building what was requested, we work together to clarify what you actually need to solve.
The request isn't always the real issue. The Hidden Issue
"We want more hires." "We want stronger social media." "We want a CMS."
The brief varies, and the real cause is often somewhere else.
What you asked for
The actual issue
Building a website is one means.
We start by clarifying what needs to be solved.
Once the issue is sorted out, the project moves more smoothly.
Once the issue is visible, the right means become visible too.
Before
- It's unclear what to build
- People aren't aligned
- Rework piles up
After
- The purpose is sorted out
- Decisions get easier
- Work moves more smoothly
My role is to bridge the gap between clarifying issues and execution.
Once things are sorted out, the next job is to carry them through. I stay with the work from clarification through operations.
This is the role commonly called Technical PM.
Business
- Leadership
- Business owners
- Sales / marketing
Bridge
Technical PM
Engineering
- Development team
- Designers
- Infrastructure
Support throughout the project.
Staying involved from planning through operations helps prevent misalignment and rework, and helps the project succeed.
Planning
- Requirements
- Identifying issues
- Goal setting
Design
- Tech choices
- Estimates
- Team setup
Development
- Progress tracking
- Problem-solving
- Quality
Release
- Validation support
- Operations design
- Retrospective
These may be signs it's worth taking another look. Worth a second look?
If any of this sounds familiar, it may be a sign to pause before a renewal or improvement.
We want to renew the website, but we don't know how to proceed
- There's intent, but it isn't structured enough to brief a production company
- It's unclear which company to ask, and how
We can't tell if a consulting or SEO proposal holds up
- Priorities and intent in the proposal aren't structured
- Reports arrive, but the next action is unclear
Technical decisions are hard, and the next step is unclear
- CMS choice and structure are hard to judge
- There's no one standing between production and the business
We want to sort out the project once
- Handover and documentation are thin, so the current state is unclear
- Specs and shared understanding aren't structured internally
It's fine if the requirements aren't specific yet.
We start by listening, then structure how to proceed.
Selected work Works
Information design and PoC for a Backlog status-sharing UI
I planned, designed, and built a support UI for the director team, using issues, comments, and update status already stored in Backlog. The aim was to make missing shared context, unclear next actions, and uneven review load visible as an observation layer on top of Backlog. After a PoC in real work, the team valued being able to see member load and case status that the default Backlog screens made hard to grasp, and a formal rollout across the business unit was decided.
To reduce missing shared context and uneven review load in production, I planned, designed, and implemented a Backlog API support UI. After the PoC, the business unit adopted it as a standard tool, together with Backlog operating improvements, and decided to roll it out across all cases.
Conversion-path design and measurement for B2B case-study pages
On a B2B corporate site, I improved how case-study pages lead people from reading a case to the next consideration step, such as a download or an inquiry. The existing pages mainly presented the case itself and did not connect well to related materials or contact. I designed and implemented a CTA that switches related materials by case category, plus a persistent FAB, then compared user behavior 56 days before and after with GA4 and Microsoft Clarity — including how the CTA itself was used and where interaction friction appeared.
Measurement from CTA click through to completed download was incomplete. Follow-up requirements include impression tracking, per-CTA events, carrying case and placement data through to conversion, offering materials by topic, and checking quick-back. The work was not limited to UI implementation; behavior analysis made both the effect of the path and the remaining issues visible for continued improvement.
Task design, go/no-go, and approval flow for multiple AI coding agents
While using several AI coding agents (Cursor, Codex, Orca), I designed a shared preflight and execution flow so pre-run judgment and approval would not scatter by agent. agent-budget-router evaluates task size, referenced issues, and budget as a shared core; agent-specific differences stay in thin adapters. Linear is the task source of truth, and implementation proceeds only after human approval — verified on real machines.
Rather than depending on agent performance alone, the system designs what is sent, in what unit, and to which model. Judging load and cost before a run turns AI use from ad-hoc trial into reusable operating rules.
What clients have said Voices
They've been a steady support on the project, and we rely on them as an essential partner.Web production company / Director
Their updates and task handling are careful, and they're someone we can trust with the work.
They pick things up quickly and keep stacking the necessary work calmly.In-house company / Web owner
CMS operations and daily reports are accurate, and even short tasks get finished reliably. There is a practical steadiness to the work.
They also take in specialized intent and shape it carefully, which we value highly.
They grasp the situation quickly, and even a vague request turns into structure and a proposal.IT company / Project lead
They bring in examples and perspective we wouldn't get inside the team, which helps.
Day-to-day handling is careful, and communication is smooth. The calm way of working makes it easy to leave things with them.Web production company / PM
They're a partner we'd like to keep working with.
Swipe to see more
What I do Services
I work across roles depending on the project, rather than staying inside a job title.
Clarifying purpose and requirements
- Deciding where to start
- Making the request specific
- Setting priorities
Structure and CMS design
- Information architecture
- CMS structure
- What's feasible to implement
Improvement and operational design
- User paths
- Measurement
- Ongoing improvement
Swipe to see more