How to find and hire a web developer

Five steps, in order: define the outcome (not the features), shortlist by proof of similar work, vet with references and live sites, start with a small paid test or tightly-scoped first milestone, then sign a contract with staged payments. Skipping the first step is why most hires disappoint — the full walkthrough is the hiring guide.

Five channels, each trading price, risk and effort differently: marketplaces (Upwork, Fiverr) for small defined tasks, vetted networks (Toptal, Codeable for WordPress) when you want screening outsourced, direct independents who publish rates and work, agencies for parallel multi-discipline builds, and referrals from businesses whose sites you can inspect. The honest comparison of channels is the alternatives section.

Search for the problem, not the profession: "WooCommerce checkout error", "Webflow CMS migration", "WordPress speed optimization". The developers who write clearly about the exact problem have usually solved it repeatedly. Then verify: their own site should be fast and clean, their client sites should still be live, and their proposal should ask you harder questions than you asked them.

Local wins when you want meetings, physical handover or regional context; remote wins on pool, rate and — usually — seniority per dollar. Web work has been remote-native for a decade, so "near me" is a preference, not a quality filter. This desk is based in Guwahati and works across four continents — both models work when the process is written down.

Same method as finding any developer — but first decide what you are finding: a person (freelancer or agency) or a platform (Wix, Squarespace, Framer). Platforms are fine for experiments; a person is right the moment the site must earn money, load fast, integrate anything, or be owned outright. The platform-versus-person decision is covered in the timelines & scope FAQ.

Proof over promises: live client sites (check them on your phone — speed is character), work similar to yours, clear writing in proposals, a stated position on code ownership, references you can actually call, and realistic pushback on your ideas. A developer who disagrees with you respectfully during the sales call will disagree respectfully when it matters. The checklist version is the vetting guide.

The six that expose the most: What would you cut to hit my budget? Who exactly does the work? What do you need from me, by when? What happens when something breaks after launch? Show me a project that went wrong. Who owns the code and when does that transfer? The expected answers — and the red-flag versions — are scripted in the interview template.

Verify three artefacts: their own website (fast, maintained, ranks for something), two or three client sites (still live, still fast, real businesses), and one reference conversation. Then test the working relationship with a small paid task — one week of real collaboration predicts a project better than any portfolio. Good developers welcome all three checks.

Portfolio depth (not breadth), a reference from a client who paid them, a repository or code sample you can have someone read, verifiable identity (domain email, registered business, consistent history), and a paid probation milestone before the full engagement. Red flags: full payment upfront, no written scope, "my last client owns the code so I can't show you", and pressure discounts. The detailed method is the vetting guide.

Designers shape how it looks and feels; developers build how it works and performs. Many independents — this desk included — do both, because for most business sites the disciplines are one craft. Split the roles when brand identity is the product (design-led, agency territory) or when the build is software-heavy (engineering-led). For a results-first business site, one accountable person doing both removes the classic designer-developer blame loop.

Budget-first: state the real band, get quotes against a written brief, and hire one accountable generalist on a proven CMS rather than the cheapest bidder or the biggest agency. Small businesses die from abandoned sites, not mediocre ones — so prioritise handover quality and maintenance terms over feature count. The sector-specific versions live in the industries section.