Where does selection begin?
Selection begins with a shortlist built from proven retail work, not from general design reputation. Retail brands need agencies that already know shopping behaviour, shopping habits online differ sharply from other digital products.
Retail carries special pressures that other sectors never face. Seasonal traffic spikes, basket abandonment, product search across thousands of items, and returns handling are all demands that generic portfolios cannot provide. Shortlists built from filtered searches across any directory of UX agencies move faster than open searching, because sector filters keep candidates outside the sector away before any call happens. Three to five names give enough range for honest comparison without stretching evaluation weeks longer than needed. Research at this stage covers each candidate’s client history, team size, and years of retail delivery, giving every later conversation a factual base to build on.
What proves retail skill?
- Live store evidence
Working shops carry more proof than any case study. Brands open each claimed store, place items in baskets, test search with misspelt words, and try checkout on a phone, because friction found in five minutes reveals what presentations alone cannot show. Phone testing carries extra weight here, since most retail visits now arrive through small screens.
- Measured results
Numbers behind each project separate completed work from display material. Conversion changes, basket size movement, and return visit rates tied to specific redesigns give brands something a reference call can confirm.
Which questions sort candidates?
Direct questions during first calls sort prepared candidates from confident presenters within one meeting. Answers naming people, methods, and past lessons carry weight that general promises never match.
- Who exactly works on this project, and what did each person build before?
- How does the team test with real shoppers rather than office colleagues?
- What proved difficult on a past retail project, and what changed afterwards?
- How does design handoff reach developers without detail loss?
How should trials run?
Small paid trials serve final decisions between two final candidates better than long contracts. One fixed piece of work shows collaboration quality faster than any proposal document.
- Pick one real problem, small enough for two or three weeks of work.
- Give both finalists identical briefs and identical access.
- Judge process quality alongside output, watching communication and question depth.
- Choose based on evidence from the trial, and offer the full engagement to one.
Trials structured this way protect both sides, since each party learns how the other works before any long commitment forms.
Reading working chemistry
Chemistry between teams shapes project quality as much as skill does. Meetings during evaluation show whether an agency listens before proposing, questions unclear briefs openly, and explains choices in words that merchants follow without translation.
- Listening balance – Candidates asking more than telling during early calls usually research the same way during projects.
- Honest dialogue – Agencies raising respectful questions about an untested idea during evaluation will bring the same care later.
Signals like these surface only in live conversation, which makes the trial stage doubly useful since chemistry and craft get tested together under real conditions.
Retail brands carefully follow evidence through every stage. A filtered shortlist, tested live stores, direct questions, a small trial, and attention to working chemistry together produce a choice the brand can defend with confidence. Partnerships formed this way tend to last across many seasons rather than one project, growing steadier as each side learns the other’s working rhythm.