The starting point: a product flying blind.
RecruitingHUB was meant to give SMEs without recruiting experience everything in one place: design modern job ads, publish them on the right boards, manage applications – one invoice instead of five contracts. The product was already in development but not yet on the market. The external UX agency was gone, the concept patchy, and the stakeholders no longer exchanged constructively. Depending on whom you asked, a different target group was meant – each with a different feature set.
My official brief: work out detail screens, advance the concept work. My actual first job: make sure people talked to each other again – and gather evidence that we had to survey the customers before building further.
Research against resistance
With rudimentary personas from the existing data, I made visible to management how far the internal ideas about target groups diverged – and thereby won time for real research. Together with marketing we contacted 150 potential customers: questionnaires to known contacts, cold calls by phone. Which target group really needs this? Which can we exclude? Is our prioritisation right?
The rebuild: separate design and publish
The Hotjar recordings showed where the order flow lost customers. My hypothesis: the concept mixed two fundamentally different tasks – the design of the ad and its publishing on the job boards. I rebuilt the entire flow as lo-fi mockups, tested in quick iterations and cleanly split it into these two halves. In parallel, an atomic design system grew in Sketch, so design and development used the same reusable building blocks.
A month after the release of the new structure, we evaluated the funnels with marketing: around 30 percent more conversion in the order flow. And an unexpected bonus for sales: because people who dropped off had already entered a lot of data, the sales team could address their needs specifically. RecruitingHUB is, by the way, still on the market today.
What stuck
UX maturity can't be decreed, it has to be earned – with evidence, not appeals. The real world deviates from the textbook; whoever steers methodically straight at the goal gets stuck. And: a design system accelerates enormously – but only as long as someone takes ownership of it.
Figures: funnel evaluation Google Analytics/Hotjar after release (approx. +30% conversion) · reference letter Prospective Media Services PMS AG, December 2018 · own case documentation 2017–2018.