B2B manufacturer website, not a template: what it really costs

The problem

A manufacturing company decides to go public-facing and discovers it either has no site at all or a five-year-old one-page business card. The brief is almost always worded the same way: "we don't want to look like everyone else", "we need it fast", "we want to add products ourselves later". What actually gets bought is a ready-made theme with demo content — and six months later it turns out three competitors run the same template, mobile PageSpeed sits at 35, and changing a price in a spec table means emailing the contractor and paying by the hour.

In B2B the cost of this isn't aesthetic. An engineer or a procurement officer lands on the site with a specific question: do you have an item with these parameters, do you make to order, what quality documentation comes with it. If there's no dedicated product page with a spec table and a request-a-quote form, the enquiry goes to the competitor who has one. The second loss is quieter: the enquiry drops into a mailbox, the sales manager opens it three hours later, and by then the buyer already has an answer from someone else.

What makes it hard

The main constraint isn't technical, it's the budget expectation. A benchmark of roughly $250–300 combined with a requirement for "a unique look + a 15–20 item catalogue with individual product pages + a blog + copy written from scratch" does not exist in one project. At that budget you get a single-page site on a website builder with a template look, no catalogue as a data structure, and no written content — which is precisely what the client asked to avoid. This has to be said at the kick-off, not discovered at handover.

The second constraint is source material. Manufacturers usually have no ready copy, no photography of their own production floor, and no structured list of items with specifications. So the contractor writes the copy, licensed stock imagery fills the gaps, and the product list becomes a separate dependency: until it arrives, product pages stay empty and the deadline slips. The third is timing. "Two or three weeks" tied to a fixed date — a trade show or an event — often means the full scope has to be squeezed into 10 working days. It's more honest to split it: by the deadline the site is live with design, catalogue and forms; SEO polishing and articles follow the week after.

How we solve it

We build the design straight in code, working from the client's references. The client supplies 3–5 sites and explains what exactly they like about each; the homepage is deployed to a working URL and reviewed in a browser and on a phone. This removes the "first a Figma mockup, then the same thing again in code" stage — duplicated work that gets paid for twice. Stack: Next.js with a built-in CMS (Payload) and PostgreSQL, static generation, automatic WebP/AVIF images, target mobile PageSpeed 90+ with the score recorded in the acceptance document. The alternatives lose on specific points: website builders deliver 40–65 on speed plus a $16–49/month subscription forever; WordPress with plugins means an update zoo and a recognisable theme.

The catalogue is built as a content type, not as a pile of pages: product = name + spec table + photos + applications + files/drawings + quote request form. Growing from 15 items to 60, adding categories, filters and a second language then becomes configuration rather than rebuilding the site. Multilingual support is switched on at CMS level from day one even if we launch in a single language: the only separate job left is translating the content, and that's logically done when the export direction actually materialises. Submissions from every form go to email and Telegram simultaneously, so the sales manager sees them within seconds; spam protection works without a CAPTCHA that scares buyers off.

A separate block exists so the site never becomes a hostage. Domain, hosting, mail and analytics are registered in the client's own name from day one, and the source code sits in the client's repository from the start of work, not after the final payment. The admin panel is in the client's language plus a 15-minute video walkthrough: copy, products, specifications, gallery, FAQ, articles and contacts are all editable in-house. If the site is meant to generate enquiries from search rather than serve as a link in sales emails, an SEO layer is added: collecting the queries B2B buyers actually type, structure and copy built around them, Schema.org Organization/Product on every product, Google Business Profile, and several engineering-grade articles — those are what capture the long technical queries an engineer searches with.

What drives the price and timeline

Market range for 2025–2026 (custom code, not a builder): a one-page site on a custom codebase — $400–550 and about a week; a corporate site with a 15–20 item catalogue, individual product pages, a blog and turnkey copy — $900–1,200 and 12–15 working days; the same site plus keyword research, product microdata, articles and CRM-connected enquiries — $1,300–1,600 and 16–20 days; a version with a second language, a catalogue of up to 60 items with filters, a selection configurator and a dealer portal — $1,900–2,400 and around 25 days. Small running costs are paid directly by the client: domain $10–25/year, hosting with CDN from $5/month, SSL free, licensed imagery $0–70 one-off.

The main price driver is not the number of pages but the volume of content creation and the depth of the catalogue. Writing 12 pages of copy and 20 product descriptions with specifications is more work than building those pages. The second heaviest factor is whether a second language and catalogue filtering are needed: if the structure is laid down at the start, adding them later is a top-up payment; if it isn't, the rework costs about half the price of the site. Post-launch support typically runs $90–130/month for updates, backups, monitoring and a few hours of edits, or $25–40/hour with no retainer.