Understand why useful visitors stop short. Diagnose the journey, prioritise changes and measure whether they help people take the right next step.
Conversion optimisation improves the journey between a website visit and a meaningful action: a suitable enquiry, completed purchase, membership application or booking. I review the available evidence, identify friction and prioritise changes to content, layout, forms or handoffs. Implementation and evaluation are defined in the engagement, so the work can progress beyond an audit.
A low conversion rate does not prove the design is wrong. Traffic may be poorly matched, the offer may be unclear or tracking may be counting the wrong thing. We establish what is happening before choosing a fix. For sites without enough traffic for a useful A/B test, we use practical research, usability checks and carefully interpreted before-and-after evidence.
CRO works best when the business is willing to inspect the whole decision, including lead quality and what happens after submission.
You can see visits, but it is unclear where or why people stop. We check measurement and separate audience mismatch, missing information and interaction problems before recommending design changes.
Forms are being submitted, but many enquiries are unsuitable. We review the promise and qualification process, treating a good-fit response as more valuable than raw submission volume.
You want a prioritised set of changes rather than a speculative rebuild. If the site is broken, tracking is missing or the business offer is unsettled, those foundations become the first actions.
Each recommendation should explain the observed problem, proposed change and how we will judge it.
Define the primary action and quality signals, then check whether the data represents them. Review traffic sources, device differences and funnel steps where available. Separate form starts, completed submissions and real opportunities rather than pooling them as equivalent conversions.
Inspect key pages and interactions on mobile and desktop. Review the offer, information gaps, trust evidence, form behaviour and third-party handoffs. Use customer questions, approved feedback and agreed usability research to understand issues the numbers cannot explain.
Create a ranked queue using evidence strength, likely customer impact, effort and dependencies. Agree a focused batch of changes to copy, forms, structure or interface behaviour. Record the hypothesis and implementation date so later evaluation has context.
Use a controlled experiment when traffic, conversion volume and the decision justify it. Otherwise, test task completion, check defects and review directional changes over a comparable period. Report uncertainty, campaign changes and seasonality rather than presenting correlation as a guaranteed lift.
Illustrative diagnostic scenarios, not reported conversion gains from client projects.
Mobile visitors reach a service page, but staff report incomplete or unclear booking requests.
The approachTest the form on real screen sizes, inspect labels and validation, clarify whether the time is requested or confirmed, and agree what details staff actually need.
What to checkReview successful requests and their usability for the service team, with device and traffic context.
Visitors view products but contact the business to ask about dimensions, delivery or custom options before proceeding.
The approachCheck whether essential decision information is visible at the right point. Distinguish an information gap from a checkout fault before changing the buying flow.
What to checkReview relevant product-to-action progression and the recurring questions received by the business.
Prospective members stop before submitting because eligibility and category differences are unclear.
The approachMake category fit and required information explicit, then test the application task with representative users. Preserve necessary qualification rather than removing every field.
What to checkReview correct category selection, valid application completion and support requests.
A change earns its place by explaining the problem it is meant to solve.
| Evidence | Next action | How to judge it |
|---|---|---|
| The form fails validation on a common mobile path. | Repair the defect and test all relevant field and error states. | Successful task completion; an A/B test is unnecessary for a known fault. |
| Prospects repeatedly ask the same missing service question. | Add a clear answer before the enquiry action. | Task understanding and enquiry quality; compare later behaviour with context. |
| Conversions are scarce and visitor volume is low. | Use qualitative task checks and fix supported problems in a documented batch. | Directional evidence without claiming statistical proof. |
| Traffic and action volume can support a controlled test. | Set a primary outcome, guardrails and analysis plan before launching a meaningful variant. | Evaluate the planned result without ending the test just because an early number looks favourable. |
| Submissions rise but staff reject more enquiries. | Inspect traffic and qualification rather than celebrating form volume. | Suitable enquiries and progression alongside total actions. |
Agree the customer action, business-quality signal and baseline. Check data quality and record known campaign, seasonal or operational changes.
Review the journey and supporting evidence. Separate observed faults from hypotheses, rank the work and agree a focused set of improvements.
Make the agreed changes, test working paths and record what changed. If a controlled experiment is appropriate, define the test before exposing visitors to it.
Read outcomes in context, including lead suitability and uncertainty. Keep, revise or remove changes based on the evidence, then agree the next priority.
The right scope makes the work useful. Agree the deliverables, dependencies and responsibilities before starting.
The number of journeys, quality of existing data, research needs, implementation complexity and evaluation method drive scope. A focused enquiry-flow review differs from a multi-step store or member application programme.
Measurement and baseline checks precede evaluation. Known faults can be fixed promptly within the agreed work; learning from a change depends on traffic and action volume. We agree a review window without promising a result by an arbitrary date.
I diagnose, prioritise and deliver or specify the agreed changes. You supply account access, customer context and lead-quality feedback, approve changes and maintain follow-up. Any external developer, testing provider or research participant work has a named owner and scope.
The Te Ao Mārama website rebuild brings membership categories, benefits and signup routes into a clearer structure. It shows the information applicants need to make a decision. The enquiry guide explains practical reasons a website can receive visitors without useful responses.
Membership categories, benefits and signup routes brought into a clearer Wix website.
A practical starting point for inspecting the journey to contact.
Yes, when the method suits the evidence. A broken form, unclear offer or missing buying answer can be investigated without a large experiment. Low traffic often calls for task testing, customer feedback and documented improvements, with directional evaluation rather than claims of statistical certainty.
When traffic, conversion frequency, implementation options and the decision justify one. We estimate whether the test can answer a worthwhile question and define the outcome and guardrails first. Running two versions indefinitely on a quiet site is not a useful substitute for diagnosis.
The action that reflects the job of the journey: a completed order, valid application, confirmed booking or suitable enquiry. Supporting actions can explain behaviour, but a button click or form start should not be reported as the same achievement as a completed business outcome.
We can review audience expectations, service fit, qualification questions and the handoff to your team. Your feedback on suitability is essential. Sometimes clearer qualification means fewer submissions but a more relevant conversation; we agree how to judge that before changing the page.
The engagement states the implementation batch I will deliver and any work assigned to your developer. Findings are prioritised with evidence, effort and dependencies. Changes are then verified and recorded so there is a basis for evaluating them.
A repaired defect can be checked immediately. Business impact needs enough relevant observations and a comparison that considers traffic, campaign and seasonal changes. We agree a sensible review method and report what remains uncertain instead of guaranteeing a lift within a fixed period.
Choose the service that solves the problem in front of you.
Check the outcome data before using it to choose improvements.
Build a campaign-specific destination when the current route is the wrong fit.
Understand performance issues that may create friction in the journey.
Connect traffic quality, the destination and commercial outcomes.
Work directly with Dan Smith. Based in Christchurch, working with businesses across New Zealand.
Bring the journey, available reporting and a description of a useful customer action. We can identify what the evidence supports and choose the first improvement.
Good websites work harder. Cookies help us see what’s working.
Essential cookies keep things running. With your permission, analytics cookies help us understand how you use the site, and marketing cookies help us measure ads and show relevant ads. Optional tracking stays off until you say yes. You can change your mind in Cookie Preferences.
The essentials keep this site working. Analytics and marketing cookies are optional. Choose what suits you and change your mind here at any time.
Essential cookies enable basic functions and are necessary for the proper function of the website.
Optional analytics cookies help us understand visits and interactions using browser identifiers. They run only if you accept Statistics and the relevant service.
Google Analytics 4 measures page visits and meaningful interactions after you accept analytics cookies. We do not send contact form answers, names, email addresses or phone numbers in analytics events.
Service URL: policies.google.com (opens in a new window)
Marketing cookies are used to follow visitors to websites. The intention is to show ads that are relevant and engaging to the individual user.
Measures page visits and successful enquiries and helps Daniel James show relevant ads on Facebook and Instagram after you visit. This runs only if you accept Marketing and Facebook Pixel.
Service URL: www.facebook.com (opens in a new window)