Conversion Rate Optimization

Coliving sites lose prospects at three predictable points. First, the booking path: the tour request or application flow that takes six clicks, demands a phone number and a life story upfront, or ends in a form that emails someone who replies two days later. Every added field and every extra step sheds real prospects — and the ones who push through then hit the response-time problem: research from the MIT-affiliated Lead Response Management study (Oldroyd, 2007) found the odds of qualifying a lead are 21x higher with a 5-minute response versus 30 minutes. A fast form feeding a slow inbox is still a leak.

Second, room pages. This is where the decision actually happens, and most operators treat them as inventory listings: one dark photo, a vague 'from £X' price, no availability, no sense of who lives there. Prospects don't email to ask for the missing information — they open the next operator's tab.

Third, mobile speed. The majority of rental discovery happens on a phone, often on a commute or a lunch break, and Google's research with SOASTA found the probability of a bounce increases 32% as page load goes from one second to three. A beautiful site that takes five seconds to load a room gallery is losing tour requests to physics. The audit finds all three leak types and puts numbers on each, so fixes land in order of leases recovered — not in order of what's fun to redesign.

A prospect who doesn't book a tour usually isn't unconvinced — they're uncertain, and uncertainty has a checklist. Zillow's 2024 Consumer Housing Trends Report found 79% of renters call at least one digital feature essential when deciding which home to rent: 56% say photos, 49% say floor plans, 25% say a 3D/virtual tour, 20% a recorded video walkthrough. That's the baseline for a room page that converts — real photos of the actual room (not the show flat), a floor plan, and a tour they can take at 11pm from another country. The same report found 19% of recent renters never toured in person at all, which for coliving's relocation-heavy audience makes virtual-viewing infrastructure a direct revenue channel, not a gimmick.

Beyond the visual layer, three doubts dominate. Price doubt: publish all-in pricing — one number covering rent, bills, and amenities — because 'from £X' plus hidden costs is the exact anxiety coliving exists to remove. Availability doubt: show live or regularly-updated availability by room and move-in date; 'enquire to check' converts like a closed door. And people doubt — the coliving-specific one: who will I actually live with? Resident photos with faces, named reviews, community stories, and honest house pages do more here than any amount of amenity photography. We audit your pages against this full checklist and build out whatever's missing, in impact order.

CRO has a reputation problem, earned by agencies that run 40 button-color tests and report a 4% lift in clicks on a page nobody converts from. We anchor everything to one north-star metric: tour-request rate (with application rate and enquiry-to-tour rate close behind), because tours are the step that predicts leases. If a test can't plausibly move tour requests, it doesn't make the roadmap.

That discipline changes what gets tested. High-leverage coliving tests look like: pricing presentation (all-in monthly figure vs. itemised breakdown), tour CTA framing ('Book a tour' vs. 'See the house' vs. offering a video call option), form length and staging (what's genuinely needed to book a viewing — usually far less than the form asks), room page structure (community-first vs. room-first), and proof placement (where reviews and resident stories sit in the path). Each test states its hypothesis, its minimum sample, and its decision rule before launch — and where your traffic can't support a statistically honest split test, we say so and use sequential before/after measurement instead of manufacturing fake significance. Fewer tests, bigger questions, real answers.

Speed is the least glamorous conversion lever and one of the most reliable, because it acts on every visitor before your copy or photography gets a chance. Google's research with SOASTA measured it directly: as mobile page load goes from one second to three, bounce probability rises 32% — and it keeps climbing steeply from there. For a coliving site, the pages that hurt most are exactly the heaviest ones: image-rich room galleries, map embeds, and booking widgets loading third-party scripts.

Our speed work is conversion-prioritised, not score-prioritised. We fix the pages in the tour-request path first — home, city/property pages, room pages, booking flow — and we target the metrics that map to felt experience (Core Web Vitals: how fast content appears, how soon the page responds, whether it shifts under the prospect's thumb) rather than chasing a perfect lab score on pages nobody visits. Typical fixes: properly sized and lazy-loaded images, deferred third-party scripts, and stripping the plugin weight that accumulates on operator sites over the years. The result isn't a vanity metric — it's more of your existing visitors surviving to see the room.

We run CRO as a 90-day sprint with three phases, so you always know what's happening and what it produced. Days 1–15, diagnose: full tracking setup and verification, conversion audit of every path to a tour request, session-recording review, speed baseline, and the ranked leak list — delivered as a plain-English document with the estimated lease impact of each fix. Days 16–45, fix: the no-test-needed repairs ship — pricing clarity, form reduction, room page rebuilds, speed work, trust-layer gaps. These are the changes that move numbers fastest because they remove outright blockers rather than optimising preferences.

Days 46–90, prove: the test roadmap runs on the real questions the audit surfaced, winners get held and losers rolled back, and the measurement system starts producing its monthly one-pager — tour-request rate, application rate, and what changed, against the day-15 baseline. At day 90 you have three things regardless of what happens next: a site that converts measurably better than it did, a tracking setup you own, and a documented backlog of the next highest-value improvements. Operators who continue past the sprint move into ongoing test-and-tune cycles; operators who pause keep everything and lose nothing. Either way, the leak list is shorter than when we started — and you can see exactly how much shorter.