Use this as your checklist. Most bad hires happen because one or more of these steps got skipped.
Step 1 — Diagnose the actual problem
Before you talk to anyone, write down the answer to one question: "What is actually wrong, and how would I know if it were fixed?"
A bad version: "Our store needs improvement."
A good version: "Our mobile conversion rate is 1.2% (industry benchmark for our category is 2.5–3%). Customers add to cart but don't reach checkout. We've already enabled Shop Pay and Apple Pay. We have 45,000 monthly mobile sessions. Success means lifting mobile conversion to at least 2% within 90 days."
The good version filters out 80% of candidates immediately — generalists can't engage with it, specialists can.
If you're not sure what your problem actually is, work through the relevant diagnostic article first. Examples:
Step 2 — Write a one-page project brief
This is the most underrated step in hiring. A clear brief saves you weeks of misaligned conversations.
Include:
- The problem in one sentence. Specific, measurable.
- Context. What's your store, what's your scale (sessions, revenue, plan), what's your stack (theme, key apps, integrations).
- What you've tried. Honestly. Saves you from being pitched things you've already done.
- Constraints. Budget range, timeline, must-not-break, must-stay-on-Shopify, internal resources available.
- What success looks like. Specific metrics, deliverables, or acceptance criteria.
- What kind of engagement. Project, retainer, audit, fractional? See Shopify Freelancer vs Agency.
Send this to every candidate before any call. The way they respond to it is the strongest filter you have.
Step 3 — Source 3–5 candidates
Don't talk to just one. You need comparison points to know what's a fair quote and what's a good fit.
Sources, ranked roughly by signal quality:
| Source | Signal | Best for |
|---|
| Vetted directory (Shopexperts, etc.) | High — pre-screened, Shopify-specific | All categories, especially when you don't have a personal network |
| Direct referral from a peer founder | Very high — based on real outcomes | Anyone, if available |
| Past collaborators | Very high | When the work overlaps your previous projects |
| Shopify Partner directory (official) | Medium — broad, not filtered by specialty | General developers, less for specialists |
| Freelance marketplaces (Upwork, Fiverr) | Low to medium — flooded with low-quality | Small, well-scoped one-off work |
| LinkedIn / cold outreach | Variable | Senior specialists and consultants |
| Communities (Reddit, Slack groups) | Variable | Specific niches and emerging specialists |
| Agency websites direct | High for fit | When you know you want agency-level work |
For most merchants, a Shopify-specific vetted directory is the highest-leverage starting point — pre-filtered by specialty, focused on Shopify, and structured to compare candidates side-by-side. Browse Shopify experts by specialty.
Step 4 — Interview with diagnostic questions
The mistake most merchants make is interviewing on portfolio and price. The right questions are diagnostic — they test whether the candidate actually understands your problem.
Bad questions:
- "Show me your portfolio." (Tells you they can design, not whether they can fix your specific issue)
- "What's your rate?" (Tells you cost, not value)
- "Have you done this before?" (Yes/no answers are useless)
Better questions:
- "What do you think is most likely causing the problem I described?"
- "What would you check first if you took this on?"
- "What's a similar issue you've solved before, and how did you approach it?"
- "What would you need from us to be successful?"
- "What's the most common reason this kind of project fails?"
- "What would you NOT do?"
Good candidates engage with the specifics of your situation and ask questions back. Generalists give templated answers and don't dig in.
A useful trick: at the end of the call, summarize back what you think they understood about your situation. If their understanding is shallower than yours, that's a signal.
Step 5 — Vet portfolio, references, and case studies
Once you've narrowed to 1–2 finalists, dig deeper.
Portfolio: Look for relevant work in your category. A Shopify expert who's done 50 jewelry stores has pattern recognition you can't easily replace, even if a generalist is "technically capable." For technical work, ask to see live Shopify stores they built or fixed — not mockups.
References: Ask for two references in roles similar to yours. Don't accept "I have great references but they're all NDA'd." Call the references and ask:
- "What did they actually deliver?"
- "What was hard?"
- "What would you do differently if you hired them again?"
- "Would you hire them again? If not, why?"
The "if not, why" question gets the truth more often than "would you hire them."
Case studies: Read them critically. A case study showing "increased conversion 200%" with no baseline numbers and no methodology is meaningless. Look for case studies with: specific starting metrics, specific deliverables, specific outcomes, and honest discussion of what didn't work.
Step 6 — Start with a small paid trial project
This is the single most-skipped step and the one that prevents the most disasters.
Before any large engagement, scope a small paid project — a few hours to a few days of work — that gives both sides real signal on:
- How they actually work (process, communication, attention to detail)
- Whether their estimates were accurate
- Whether they hit deadlines
- Whether they understand your business
- Whether you actually want to work with them
A good trial is paid (you're testing fit, not getting free work) and scoped (not "go wild and see what you can do," but "audit this specific thing and deliver this specific output").
If the trial goes well, expand the engagement. If it doesn't, you've spent a fraction of what a bad full engagement would cost.