Web design and landing pages
A site you can't measure is a site you can't improve.
Almost every site gets delivered finished and blind: nobody knows which section gets read, where the form gets abandoned or which campaign brought the sale. We build it and deliver it measuring, because that's the part that decides whether the rest was worth anything.
- Measuring from day one
- Fixed scope in writing
- The site and the domain are yours
Control KPIs · not promises
The targets we use to decide what to touch. They come from how your site behaves today, not from an industry average that doesn't know it.
| Month 2 | Month 5 | |
|---|---|---|
| Site conversion rate | +15% | +35% |
| Landing page bounce rate | −10% | −25% |
| Cost per lead | −10% | −20 to −30% |
Working targets, not contract clauses. They require enough traffic to measure: without volume, any variation is noise.
What we find almost every time
Five things repeat across the sites that come in for review. Most don't require rebuilding anything.
- 01
Nobody can tell what you do
The headline talks about values, track record or innovation, and the visitor still doesn't know what you sell. It's the most expensive failure and the easiest to fix.
- 02
Nobody knows where the leads come from
Analytics is installed, but no conversion events are defined. You can say how many visits there were; not which campaign turned them into customers.
- 03
The campaign page is the homepage
The ad promises something specific and lands on a homepage that speaks to everyone. Expensive traffic gets paid for only to get lost in the menu.
- 04
The form asks for too much
Eight fields for a first conversation. Every unnecessary one costs leads, and almost none get used later in sales follow-up.
- 05
Only the vendor can change anything
Changing a phone number or uploading a promotion means requesting it, waiting and sometimes paying for it. The site stopped being yours even though you paid for it in full.
Sometimes the answer isn't a new site.
Some projects are missing a page, and others are missing a fix to the one that already exists. We start by deciding that.
-
Campaign landing pages
One page, one goal, one message that continues the ad's. It's what holds up paid campaigns and the first thing worth having if you're going to invest in advertising.
-
Corporate sites
What your prospect checks before replying to you. Clear structure, content that answers objections and contact paths that lead to where you actually respond.
-
Online stores
Catalog, payments and shipping on Shopify or on the platform you already run. With purchase tracking set up properly, which is where it breaks most often.
-
Redesign & migration
When the site exists but it's heavy, can't be edited or is locked to a vendor. We migrate while preserving the URLs that already bring you traffic.
-
Conversion optimization
Audit, heatmaps and session recordings with Microsoft Clarity, and a list of improvements ranked by effort versus impact.
-
Measurement & tagging
GA4, Tag Manager, pixel and conversions API, contact events and consent that actually work. Included, not an add-on.
The clock starts when the project has what it needs to move.
Projects rarely fall behind because of the build. They fall behind waiting for copy, photos and access, so that gets resolved before we start.
- Week 1
Discovery
How your customer arrives today, where they drop off and what decides the sale. This produces the page map and the hierarchy of each one.
- Week 1–2
Structure & content
What each screen says and in what order, before a design exists. It's the part that moves conversion the most and the one the industry skips the most.
- Week 2–3
Design
You see the screens and comment before any code is written. Changing a design costs hours; changing a built site costs weeks.
- Week 3–4
Build
Assembled on the agreed platform, designed mobile-first and with the technical search-engine foundation resolved from the start.
- Before launch
Measurement & testing
Each event is tested separately. Speed is checked, real forms are submitted and, for stores, a purchase is completed. If measurement doesn't pass, it doesn't go live.
- Handoff
Training
We teach you how to edit it and hand over access, files and documentation. From there you don't depend on us for day-to-day changes.
Four rules you can verify.
Four concrete conditions, not a list of values. Each one can be checked from your side.
- 01
The site doesn't launch without measurement
Analytics, contact events and consent are tested before going live. If you can't answer where a lead came from, the site isn't finished.
- 02
Fixed scope, with milestones and a date
A flat fee against a written scope. If something outside it comes up, it's quoted separately and you approve it before anyone works on it. No charges appearing mid-build.
- 03
Domain, hosting and files in your name
Everything is registered in your name or your company's. You can request your backups and access whenever you want, whether you work with us or not.
- 04
You manage it
Delivered on a platform your team can operate, with training included. Whether we keep doing maintenance has to be your decision, not a consequence.
When we're not the right fit
- You're looking for the lowest price on the market: we're not going to be that quote.
- You need the site in two weeks and the content doesn't exist yet.
- You want a site that only looks good; here conversion wins over aesthetics.
- There's no one on your side who can make decisions. With four unconsolidated opinions, the project stalls.
What people ask before signing
-
I already have a site. Do I need a new one or can it be fixed?
Often it can be fixed, and saying so takes some effort because a full redesign bills better. But rebuilding a site that works is expensive and risky: you lose the rankings you've already earned and start from zero with something that hasn't been tested yet.
When fixing makes sense
- The site gets traffic and something converts. There's a baseline to measure against. Adjusting headlines, hierarchy and forms usually moves the needle more than a new design, and the effect shows in weeks.
- The problem is the message, not the structure. If nobody understands what you sell in the first few seconds, that gets fixed by writing, not by coding.
- Measurement is missing. It's the cheapest intervention with the biggest effect. Without it you can't justify any later investment.
When rebuilding does make sense
- You can't edit it without depending on someone else, or the platform no longer receives security updates.
- It loads slowly and can't be fixed because of how it's built. Some limits don't get solved by optimizing images.
- The structure doesn't fit what you sell today. The business changed and the site still tells the old story.
- It doesn't work well on mobile. That's where most of your visitors decide, and patching a desktop design rarely does the job.
How we decide
On the call we review what your site receives today and what it's measuring. That's usually enough to say whether the path is a targeted intervention or a full project. It's a thirty-minute conversation and it produces a concrete recommendation, even if the recommendation is that you don't need to hire us today.
-
How do I know if my site is measuring anything?
You can check it yourself in twenty minutes, without hiring anyone. It's four checks and they don't require technical knowledge. If all four fail, every dollar you invest in campaigns will be evaluated blind.
The four checks
- Open Google Analytics and look for the key events. If the list is empty, or only shows automatic ones like page views and scroll, your site counts people but doesn't count business.
- Submit a test form and watch it in real time. It's the simplest check and the one that fails most often. If the event doesn't show up, your leads aren't being measured.
- Click your WhatsApp button and look for the record. In Mexico and much of Latin America that's where a good share of demand arrives. It's almost never measured, so the channel that sells the most is the one that shows up least in reports.
- Accept and decline the cookie notice. If the difference changes nothing in practice, you're not compliant and your data is biased on top of it.
What each result means
If all four pass, your site is in better shape than most and the next job is conversion, not instrumentation. If one or two fail, it's usually a matter of configuring what's missing on top of what already exists. If all four fail, it doesn't mean you need a new site: it means you currently have nothing to decide with, and fixing that comes first.
Why we insist so much
Because it's the difference between investing and gambling. With measurement you can say which campaign brought which customer and at what cost, and from there scale what pays off. Without it you only know how much you spent. It's also why none of our sites go live without this part resolved, even if it means moving the launch date.
-
How much does it cost, and why don't you publish a price list?
Some agencies publish fixed packages and it's a respectable practice: it saves everyone time. We don't, because in site projects the same package can mean very different things, and a table that doesn't say how many sections or which integrations it includes ends up generating the discussion it meant to avoid.
What drives the price
- How many screens and of what kind. A campaign landing page and a twelve-section site have nothing in common, even if both are "a website".
- Who writes the content. With material ready, the project moves on its own. If it has to be produced, that's additional work quoted as such instead of showing up later as a surprise.
- What gets connected. A CRM, a payment gateway, a booking system or an automation add tests that someone has to do properly.
- Whether there's a migration. Preserving the rankings of a site with history is a job in itself, different from building from scratch.
How we quote
We close scope before price. Together we define the screens, the integrations and the timeline, and from there comes a fixed figure with payment milestones and an estimated date. Whatever falls outside the scope is quoted separately and you approve it in writing before it starts.
The fee is flat and we don't bill open hours. That means if the project takes us longer than planned because of us, it's our problem and not your invoice.
The most useful question
It isn't how much it costs but what has to happen for it to pay for itself. A landing page behind a campaign can recover its cost in weeks if the campaign already works. A corporate site that exists to build trust takes much longer, and that's fine, but it's worth knowing before signing. On the call we tell you which of the two cases you're in.
-
What do you need from me and how much of my time will it take?
Less than you fear, but concentrated in specific moments. Uncertainty about this stalls more projects than price does, so it's worth being concrete.
Your time, in total
- One discovery session of roughly ninety minutes at the start. It's the most important and the only long one.
- Two reviews — structure and design — of half an hour each, with written comments.
- One training session at handoff, one hour, so you can edit it yourself.
Beyond that, occasional messages. No weekly status meetings unless you want them.
The material that delays projects
- Logo in vector format, not a PNG pulled from an old presentation.
- Copy per section, even in draft. If you'd rather we write it, that's defined in the quote because it's scope.
- High-resolution photos, or the decision to use stock imagery. Product photos take up more of the calendar than people expect.
- Access: domain, current hosting and the accounts for whatever needs connecting.
What weighs the most: who decides
It's the most underestimated factor. When comments arrive from four people at different times and contradict each other, the project stalls while someone resolves the contradiction, and nobody gets that time back. Before starting we ask who approves, and if the answer is a committee, it's worth consolidating comments before sending them.
If you don't have something on the list, it's not an obstacle: it gets resolved, but it gets resolved before starting and with the time accounted for in the schedule. Starting halfway and discovering in week three that half the content is missing is the most common way a project doubles in length.
-
Who owns the site when we're done?
You do. The domain, the hosting, the files and the measurement accounts. It sounds obvious and it isn't: it's one of the things that causes the most problems later, and it's worth putting in writing before starting rather than discovering it when you want to move on.
How each piece is handled
- The domain is registered in your name or your company's. It's a brand asset, not a vendor input.
- Hosting you can contract yourself and we configure it, or we manage it if you'd rather not deal with it. Ownership doesn't change in either case.
- Files and backups are handed over at close and you can request them whenever you want.
- Measurement accounts live under a user of yours. We come in as administrators, not as owners.
Why we set it up this way
A relationship that holds because the client can't leave is a bad relationship, and it also produces bad work: when the vendor knows you're trapped, the urgency to respond to you drops. We'd rather you stay with us because it's in your interest.
If you're coming from another vendor
It's worth requesting in writing, before announcing you're leaving: administrator access to the domain registrar or the transfer code, a full backup of files and database, ownership of the analytics accounts and the tag container, access to the integrations, and a note on which platform and version was used.
If any of that comes with conditions, that's exactly why ownership is worth resolving at the start of any project. And if you're already in that situation, almost everything can be rebuilt: it takes time, it isn't impossible, and we've done it before.
arborix Almost no one knows where to start.
That's why we begin with a diagnostic: we review what you already have running, tell you where the most expensive leak is, and give you a 90-day plan. With that, you decide what to hire — or whether to hire anything at all.
Reply within 24 business hoursFree callNDA signed by Arborix