Web architecture
Headless E-commerce & B2B Quoting
We build storefronts and B2B platforms two ways - a custom headless architecture, or advanced engineering on the WooCommerce you already run. Either route: per-account pricing, quoting engines with your own rules, and instant search across thousands of SKUs.
2017 — todayOne relationship, two architectures · Industrias Cruz Centro
Sound familiar?
- Your store is slow, and every extension you add makes it a little slower.
- You sell to businesses at different prices per account, and the platform has no concept of it.
- Catalogue search fails to find products that do exist, and you see it in the cart.
- Every quote comes out of a spreadsheet only one person knows how to use.
Is the problem WooCommerce, or how it was built?
Almost always the second. WooCommerce moves an enormous share of the world commerce and handles large catalogues without breaking a sweat. What it does not handle is the habit of solving with extensions what calls for code. Each one adds its own JavaScript, its own tables and its own minimum version, and the day arrives when updating one breaks two others. The store is not slow because of hosting or the CMS. It is slow because of the patch layer.
So the first question is not which platform, but what your store is doing on every visit that it should not be doing. Often the answer fits inside WooCommerce, with the storefront rewritten and the engine untouched.
When does separating storefront and engine pay off?
When the business breaks the average case and the mould starts costing more than it saves. Negotiated prices per account, volume tiers, products only a distributor can see, quotes following rules only your sales team knows, or one catalogue feeding store, app and marketplace at once.
What that looks like in practice
The storefront is served as HTML generated at the edge, so a product page appears immediately even with a hundred thousand SKUs behind it. Search does not hit the database - it hits an index built for the job, typo-tolerant and filterable by attribute without a page reload.
And the catalogue stops living inside the theme. It lives in a panel your team edits, and that your ERP, your own app or whichever marketplace you sell on next can read from just as easily.
What if you sell quotes rather than units?
Same engine, different output. Instead of a cart, a configurator that applies your rules and produces a quote with its PDF, its validity window and its status. The difference from the spreadsheet you use today is not cosmetic: it leaves a trail, it does not depend on one person, and the customer can check it without sending an email.
What you get
- Custom Next.js storefront, rendered at the edge with the catalogue prefetched.
- Or that same storefront over your WooCommerce, consumed by API, leaving your orders and customers untouched.
- Catalogue and content editable by your team from the panel, without touching code.
- Instant search with Meilisearch, typo-tolerant and filterable by attribute.
- Pricing rule engine - per account, per volume, per price list and per validity window.
- Quoting flow with PDF generation and status tracking for every quote.
- Integration with your ERP or billing system for stock, prices and orders.
How we work
- 01
Catalogue modelling
1 to 2 weeksAttributes, variants, pricing rules and who edits what. This is the phase that decides whether the platform holds up in year three.
- 02
Architecture and integration
2 weeksERP or billing connection, search index and caching strategy. Your technical team and whoever owns the sales process both take part.
- 03
Build
5 to 8 weeksStorefront, checkout or quoting flow and editing panel, with performance QA running alongside.
- 04
Launch
1 to 2 weeksCatalogue migration, redirects, live payment gateway and monitoring through the first 30 days.
Investment
Scoped quote
A costed proposal within 3–5 business days
Scope drives price. Send us the project: within one business day we tell you whether it fits, and the proposal — what is in, what is out and what it costs — arrives within three to five business days.
The range depends on catalogue size, per-account pricing rules and how deep the ERP integration goes. The number comes out of the scoping call, not a price list.
Proof of work
Frequently asked questions
Headless or WooCommerce?
We build both, and the choice is not ideological. WooCommerce is a serious commerce engine: it handles catalogue, orders, customers and tax, and your team already knows it. What usually fails is not WooCommerce but WooCommerce without engineering - twenty stacked extensions solving what code should solve. If that base is healthy, the fastest route is to keep it and put a custom storefront in front. If the business model breaks the mould -negotiated per-account pricing, rule-driven quoting, several channels over one catalogue- then it pays to separate storefront and engine entirely.
Do I have to rebuild my store from scratch?
Rarely. If your WooCommerce has orders, customers and catalogue in good order, the usual move is to keep it as the engine and change only what is seen and what is searched. History is preserved, your team relearns nothing, and the project runs in weeks rather than months.
Can I keep my payment gateway?
Yes. We integrate whichever one you already use, including local Latin American providers. What changes is the storefront, not the payment.
How large a catalogue can the search handle?
Meilisearch answers in under 50 ms across hundreds of thousands of SKUs. The practical limit is not volume but how the attributes are modelled, and that is decided in the first phase.
Can my sales team change prices without us?
Yes. Pricing rules live in the panel, not in the code. We define with your team what can be changed safely and what needs review.
What happens to my current store rankings?
They are preserved. Before touching anything we export the URLs that carry traffic and build the 301 redirect matrix. No indexed product URL can end in a 404.