1. Clarify the goal
Leads, brand, franchising, diligence, hiring — pick a primary goal. It decides the first screen and navigation. The rest is secondary. An unclear goal means rework. See Why every company needs a website.
Treat the site as “hire someone for a few pictures” and the project dies in color changes. Run it as engineering. Each stage has a sign-off: goal, navigation, content list, design, clickable pages, launch list, and who maintains it.
Leads, brand, franchising, diligence, hiring — pick a primary goal. It decides the first screen and navigation. The rest is secondary. An unclear goal means rework. See Why every company needs a website.
List visitor jobs, then draw navigation and key URLs. Manufacturing, export, and franchising each have a pattern. See manufacturer navigation, export homepage, franchise policy.
Product facts, certificates, real photos, case boundaries. No content, no detailed design. This is delay number one.
Build components from the VI. Do not paint an unbuildable homepage first. Consider responsive at the same time. See responsive.
CMS, inquiry notices, basic SEO, and language switch. Domain and ICP filing run in parallel. See domain and ICP filing.
Phone form fill, certificate download, links, analytics. Walk the inquiry path once.
Name an updater and a minimum cadence. Otherwise in three months you are back to nobody updates it.
Time and money depend on which phase you stop at. See cost and timeline. On an old site, decide redesign or rebuild first. See redesign or rebuild. Template versus custom belongs after step 2, not after buying a template and inventing the business. See template vs custom.
The usual stall is not technology. Nobody inside can sign “this sentence can go public.” Marketing wants it full. Legal fears full. Sales wants a hard lead time. Production disagrees. Without a content owner at kickoff, comps loop on “one more tweak.” In goal clarification, name who supplies facts, who reviews copy, who signs visuals — two rounds max. More than two usually means the goal is unset, not that design is lazy.
Acceptance is not “pretty or not.” Walk at least four paths: company-name search lands on this domain; a phone can submit the product form; certificates download with readable filenames; a WeChat forward is still the official page. On launch day test analytics and inquiry mail so “it looks live” is not a dead inbox. Same bar as how to diagnose no inquiries.
Phase-one launch is not the end. Products rename, phones change, certificates expire. Write upkeep into the contract or a job. Cheaper than another redesign talk. If you still hesitate between a digital card and an inquiry engine, read the two kinds of site, then decide how deep phase one goes.
At kickoff write what you will not do: no store, no membership, no ten languages in phase one. Narrow scope so quality can rise. Changes in writing, so verbal extra nav does not rewind design. A week after launch, check whether sales actually forwards pages. That tells you more than design satisfaction.
The content sheet should list required fields and examples per page, not “send us the materials.” Examples cut the loop. Photos: landscape, no watermark, commercially usable, so you do not discover a pile of phone portraits the night before launch.
On launch day check 404s and form test messages so a test note is not treated as a real customer. Easy to forget.
Lead generation, brand, franchising, or due diligence — the goal changes the navigation and the first screen. Send your current situation and goal; engineering will come back with a plan you can execute.