Rebuilding the Midsummer shop for clarity, trust, and modern e-commerce expectations

A full brand and shop site redesign for Midsummer's B2B renewable energy platform, tackling weak navigation, hidden stock visibility, and a logo that struggled for attention, all within the constraints of the existing backend.

This project reached final proposal stage, three brand and site directions presented to the board, before being placed on hold due to developer and marketing resourcing. Everything below covers the process through to that final proposal.

Role

Lead Designer

Company

Midsummer

Status

On hold

Tools

Figma, HTML, CSS, interviews, surveys

Midsummer shop redesign

Time to cart reduced

Stronger brand alignment

Clearer stock visibility

Improved navigation and discoverability

Context

Midsummer supplies solar panels, inverters, heat pumps, and other renewable goods to installers across the UK and Ireland. The site had grown into a powerful tool over time, but its interface hadn't kept pace with modern e-commerce expectations, users increasingly compared it unfavourably to slicker platforms they shopped on elsewhere.

The scope covered brand identity, product discoverability, information architecture, and visual interface, all within one fixed constraint: the existing backend had to stay largely intact.

Updated shop experience
Previous shop experience
Before After

Research

Internal workshops and a brand audit identified what was worth keeping, like the iconic sunpath in the logo, and where consistency and scalability were breaking down. Focus groups with internal sales staff, plus a customer survey, surfaced the clearest patterns:

  • Users expected stronger global search and filtering, this was one of the most consistently repeated requests from both customers and internal stakeholders
  • Stock visibility was often hidden or unclear
  • Navigation was inconsistent and caused confusion
  • Customers regularly missed products due to unclear categorisation

Shareholders separately flagged weak brand visibility, bolder logo directions tested as noticeably easier to see than the existing mark.

Affinity mapping

Goals

  1. Update the brand without losing recognition or trust
  2. Improve search and navigation to reduce user struggle
  3. Make stock availability visible and consistent across listings
  4. Deliver a solution that works within existing backend constraints
Brand audit

Brand audit

Reviewed the existing logo, type, and colour contrast. It had low visibility on light and dark backgrounds, and took up disproportionate space for the impact it delivered.

Brand direction

Iterated bolder logo directions around the sunpath, addressing the visibility issue while keeping recognition intact.

Brand direction

Board presentation

Presented three final directions, two exploring new brand territory, one developing the existing identity further, to the Midsummer board, who selected an approach balancing clarity, heritage, and modernity.

Site redesign visuals

Site redesign

Moved to top navigation, tested as easier to use and closer to typical online shopping conventions. Regrouped categories, added smarter persistent search and filtering (the most requested change), surfaced stock data in listing cards, and clarified product bundling.

Predicted impact

This wasn't measured, since the project didn't ship, but two predictions follow directly from the research above. Search and filtering was the top request from customers and staff alike, so time to cart is the headline metric the redesign targets. Stock visibility was the second most common complaint, addressed directly in the new listing cards.

Reflection

This project showed how much brand and UX intersect in a B2B context, where clarity and trust directly shape customer decisions. Working within backend constraints sharpened my prioritisation, since not every ideal solution was buildable within scope.

If revisiting, I'd push to test navigation variants with real customers earlier in the process, and make a stronger case for long-term investment in scalable component architecture, rather than only solving the immediate usability problems.

Copied to clipboard!