Choose a refresh when the foundation still works
A refresh suits a site with accurate positioning, a usable page structure and a stable platform. The work may update typography, spacing, imagery, components or selected copy while keeping URLs and core journeys largely intact.
This option is attractive when speed matters, but it should not be used to avoid necessary content decisions. If visitors cannot tell services apart or proof is missing, new styling will preserve the confusion.
- Services and audiences are still accurate.
- Navigation matches how buyers explore the offer.
- Editors can update the current platform safely.
- Forms and measurement already capture useful outcomes.
Choose a redesign when several systems must change together
A redesign is justified when positioning, page hierarchy, copy, conversion routes or technology are materially wrong. It creates room to rebuild the information model and then express it through design. It also carries more migration and approval work.
Common triggers are a merger of services, entry into a new market, a shift from general enquiries to qualified consultations, inaccessible mobile experiences, or a platform that cannot support required integrations and governance.
- The homepage explains an outdated offer.
- Important services compete on one undifferentiated page.
- The platform blocks routine content or measurement work.
- Language versions and redirects need a new architecture.
Use a six-part decision scorecard
Score positioning, navigation, content, proof, technology and measurement as sound, repairable or obstructive. A single obstructive area may be handled as a focused project. Several connected obstructions usually make a redesign more coherent because changing one will affect the others.
For every weak area, write the buyer or operating consequence. This prevents preference from becoming the decision rule. ‘The design feels old’ is weak evidence; ‘mobile users cannot compare the two main services or complete the form’ is actionable.
- What changed in the business?
- Which buyer task currently fails?
- Can the present platform support the correction?
- Which valuable pages or URLs must be protected?
Turn the diagnosis into a bounded brief
Write the chosen option, excluded work, priority pages, dependencies and acceptance criteria. A refresh might cover a component library and six templates without changing URLs. A redesign might cover a new architecture, representative content, build, redirects and measurement.
If evidence is incomplete, begin with a short discovery phase rather than estimating an undefined rebuild. The output should be a decision and scope that another qualified provider could understand.
- Chosen approach and reason
- Pages, languages and integrations
- Materials and approvals required
- Launch risks and measures of success
Frequently asked questions
Frequently asked questions
Is an old-looking website automatically a redesign project?
No. Age alone does not show that the structure is broken. If the offer, journeys and platform remain sound, a visual refresh may solve the problem with less disruption.
Can a redesign be delivered in phases?
Yes. Discovery, priority templates and content migration can be phased, provided the final architecture and URL plan are agreed early enough to avoid rework.
Should cost decide between a refresh and redesign?
Cost is a constraint, but the decision should start with the problem. A cheaper refresh is poor value if it cannot correct the structural issue; a full rebuild is wasteful when targeted work is sufficient.