Business case · UX/UI
Casavo property listings
Casavo’s listing page put a map in 70% of the viewport and the properties in the margin. I audited the page against Nielsen’s heuristics and answered each finding with a paired wireframe and interface, so every visual decision could be traced back to a specific problem.
- Type
- Self-directed business case, a brief taken on during a design selection process and run with my own method
- Scope
- Heuristic audit, four paired wireframe and UI solutions
- Surface
- Property showcase page, desktop
- Tools
- Sketch
The brief
Casavo is a leading instant buyer in southern Europe: it buys homes directly, then resells them, with the promise that selling and buying a house can be fast, simple and convenient.
Its website has a page showcasing the properties it sells. The task I set myself was to find the pain points a visitor hits on that page, and propose solutions that make browsing feel like the promise the brand is making.
The process, and what was real
Empathise
Partly delivered- Heuristic walkthrough of the live listing page
- Interviews with sellers and buyers (proposed)
- Analytics on filter and pagination use (proposed)
Define
Delivered- Six pain points, each tied to a heuristic
- What good looks like for a browsing session
Ideate
Delivered- Four solutions covering all six findings
Prototype
Delivered- Low-fidelity layout for each solution
- High-fidelity interface for each solution
Test
Method proposed- Task-based test on list vs. map default (proposed)
- A/B on the default view (proposed)
What already worked
Delivered
An audit that only lists faults is a bad audit: it gives the team no idea what to protect. Three things on this page were working and had to survive the redesign.
Keep
- Clean, minimal layout. The structure is complete and easy to manage, and the visitor feels a few clicks from what they came for. (8th heuristic: aesthetic and minimalist design.)
- Photography leads. The images do a good job of communicating the type of property Casavo deals in.
- Price is prominent. After the photo, price is the most emphasised element, which is correct for this category.
Fix
- The map dominates the viewport by default.
- There is no way back after opening a property.
- No sorting options at all.
- No call to action on a listing card.
- Pagination leaves you stranded mid-page.
- Filter dropdowns are raw and unexplained.
Findings and fixes
Delivered
Six findings, each tied to the heuristic it breaks and to the solution that answers it. The mockup references point to the four designs in the next section.
- The map takes 70% of the horizontal viewport
Displayed by default at that size, it pushes the property cards, the thing people came for, into a narrow column at the side.
8th heuristic: aesthetic and minimalist designFix
Default to the list, one property per row, with an explicit button to see the same result set on a map. Map becomes a mode, not the frame. (Solution 1) - No way back after opening a property
Opening a listing offers no “back”. The only route is the browser button, which throws away the filters the visitor had set, punishing an accidental click with the loss of their whole session.
3rd heuristic: user control and freedomFix
Add an intermediate step: the row expands in place to preview the property, vertically in list view and horizontally when the map is open, so browsing never destroys context. (Solutions 2 and 4) - No sorting
There is no way to order the result set, so a visitor with a budget or a size in mind has to scan everything and hold the comparison in their head.
7th heuristic: flexibility and efficiency of useFix
A sort dropdown above the list, persistent across pagination. (All four mockups) - Listing cards have no call to action
Nothing on the card invites a click. The one CTA in the header changes its label colour to match its own background on hover, so it becomes unreadable exactly when it is being used.
Fix
A visible action inside each card with a real hover state, and fix the header button’s contrast. (All four mockups) - Pagination strands the visitor
Clicking a page number loads a new set but leaves the scroll position where it was, so the visitor is looking at the middle of a list they have not seen the top of.
2nd heuristic: match between system and the real worldFix
Scroll to the top of the result set on page change. The cheapest fix on this page. - Filter dropdowns are unexplained
The filters are default, unstyled selects. Nothing communicates the range of what can be filtered or how the options interact, so people under-use them and then complain that the results are wrong.
5th heuristic: error preventionFix
Custom controls (sliders for ranges, icons for property attributes) that show their bounds and their effect before submission.
Four solutions, drawn twice
Delivered
Each solution exists as a low-fidelity layout and as an interface. Working in pairs keeps the conversation about structure honest: if the argument only works once the photography is in, it is not a structural argument.
Solution 1: list first, map on demand
Solution 2: expand in place instead of navigating away
Solution 3: sorting and card actions
Solution 4: the map view, redesigned
Looking back
Method proposed written now, looking back: what I would keep and what I would do differently with the experience I have today
The traceability is what I still like here: six findings, four designs, and a named heuristic behind every one. Anyone can check my reasoning, which is the whole point of writing an audit down.
What I’d do differently: the map decision deserved evidence, not judgement. “Default to the list” is a real bet. For a buyer who cares primarily about neighbourhood, the map-first layout may well be right, and I asserted the opposite from a single walkthrough. Today I would frame it as a hypothesis, put an A/B test behind it, and look at how many sessions actually touch the map before rearranging the page around the answer.