Custom Next.js builds
Server-rendered React on Vercel. Pages arrive as finished HTML, so they load fast on a phone and every crawler can read them.
Monolith Web · development, tracking + AI search
Websites built for people, search engines, and AI agents
We design and build fast custom websites, run the domains and hosting underneath them, set up GA4, Google Tag Manager, and ad-platform tracking you can trust, and make your pages easy for Google, ChatGPT, Perplexity, and browsing agents to read.
Bring your URL and whatever logins you have. We'll look at the site with you.
Platforms + tools we work in
01 · Design + development
Custom websites that load fast and report what they do
Most of our builds run on Next.js and Vercel. Pages render on the server, load quickly on a phone, and every change ships to a preview link you can click through before it goes live. If your team edits content without a developer, we connect a headless CMS or work inside the platform you already use.
The team builds with AI coding agents and reviews every change by hand. The agents make the work faster, and a person still signs off before anything reaches production.
Platform builds
WordPress, Shopify, Webflow, Wix, Squarespace, and Drupal, for teams that need to keep editing in the tool they know.
Migrations that keep rankings
A full URL inventory, a 301 redirect map, and a crawl before and after launch, so search traffic follows you to the new site.
Core Web Vitals
We measure LCP, INP, and CLS from real visits and fix what causes them, which is usually oversized images and scripts that block rendering.
Accessibility built in
Semantic HTML, keyboard paths, readable contrast, and reduced-motion support, checked against WCAG 2.2 AA.
AI on the site
Site search that answers questions, and assistants that answer only from your own pages.
Every build ships with
The same launch checklist on every project, whatever the platform.
- A private preview link for every change
- Core Web Vitals report at launch
- XML sitemap, robots.txt, and llms.txt
- JSON-LD structured data in the templates
- GA4 and Tag Manager with consent mode
- A 301 redirect map on every rebuild
- Search Console and Bing Webmaster Tools verified
- Repository, hosting, and domain in your name
02 · Hosting + domains
Hosting and domain management, with every account in your name
Plenty of outages start underneath the website: a renewal notice sent to someone who left the company, a DNS change nobody wrote down, a certificate that expired over a weekend. We inventory every domain, registrar, DNS zone, and login, move the scattered pieces into accounts your business owns, and watch renewals and certificates from there.
Email delivery lives here too. SPF, DKIM, and DMARC are DNS records, and one careless change can send your invoices to spam.
Domains
Registration, transfers, renewals, and registrar locks, consolidated into one account your business owns.
DNS
Zone records kept in one place, with a written log of every change and why it was made.
Hosting
Vercel, Cloudflare, or the host you already use, with a private staging site for each build.
Certificates
SSL/TLS that renews on its own, HTTPS everywhere, and one canonical version of every URL.
Email authentication
SPF, DKIM, and DMARC set up and rechecked after DNS changes, so invoices stop landing in spam.
Monitoring
Uptime, certificate, and renewal alerts that go to a team, not to one person's inbox.
yourdomain.com
- Registrar lockOnPass
- Auto-renewOn, 291 days leftPass
- DNSSECSignedPass
- TLS certificateValid, auto-renewsPass
- SPF2 includes, -allPass
- DKIM2048-bit keyPass
- DMARCp=noneFix
- www to apex308 redirectPass
- HSTSEnabledPass
8 of 9 passingDMARC: move from p=none to quarantine, then reject
03 · Analytics + tracking
GA4, Tag Manager, and ad pixels that agree on the numbers
When GA4 reports 40 leads, Google Ads claims 62, and Meta claims 85, nobody trusts any of them. We rebuild measurement from one written plan: each conversion named once, sent through Google Tag Manager, gated by consent, and delivered to every platform with the IDs it needs to avoid double counting.
Reporting comes last, as a Looker Studio dashboard that puts GA4, Google Ads, and Search Console on one page.
What we set up on each platform
| Platform | What we set up | Why it matters |
|---|---|---|
| Google Analytics 4 | Events and key events, custom dimensions, audiences, cross-domain tracking, BigQuery export | One record of what visitors do on the site |
| Google Tag Manager | Web and server-side containers, naming conventions, consent mode v2, versioned change notes | Tags change without a code deploy, and every change is logged |
| Google Ads | Conversion actions, enhanced conversions, offline conversion import, GA4 audience links | Bidding learns from real leads instead of page views |
| Meta | Pixel plus Conversions API, with event IDs so the two never double count | Recovers conversions that browsers and ad blockers drop |
| TikTok | Pixel plus Events API, matched to the same event names as every other platform | The same server-side backup and event names as Meta |
| Insight Tag, conversion tracking, and the Conversions API for lead forms | B2B leads tied back to the campaigns that found them | |
| Microsoft, Pinterest, Reddit, X | UET tag and platform pixels, added to the same measurement plan | Every channel reports against one definition of a lead |
04 · SEO + agentic search
Search has three kinds of readers now
For most businesses, Google still sends the most search traffic. More questions now go to ChatGPT, Perplexity, Gemini, Claude, and Copilot first, though, and some of those assistants open your page, read it, and fill in the form themselves. We run technical SEO the way it has always worked and add what the new readers need.
Search engines
Crawls and indexes your pages, then ranks them in results and AI Overviews.
- Who reads your site
GooglebotBingbot- What it needs
- Crawlable pages, fast loads, clean internal links, structured data, and links from sites people trust.
- How we measure it
- Clicks and impressions in Search Console, plus visibility tracking in Semrush and Ahrefs.
AI answer engines
Pulls a handful of pages and writes a single answer, citing the sources it used.
- Who reads your site
OAI-SearchBotPerplexityBotClaude-SearchBot- What it needs
- A plain answer near the top of the page, the same business facts everywhere, and content that renders without JavaScript.
- How we measure it
- Citation checks across ChatGPT, Perplexity, Gemini, and Copilot, and AI referral traffic in GA4.
AI agents
Opens your site on a person's behalf to compare options, fill in a form, or book a time.
- Who reads your site
ChatGPT-UserClaude-UserPerplexity-User- What it needs
- Labeled forms, stable URLs, no hover-only steps, and facts stated in text rather than baked into images.
- How we measure it
- Agent traffic in server logs, and whether those visits end in a completed form or booking.
What an agent-ready site looks like
An agent behaves like a patient customer who can't see hover menus or read text inside images. These are the fixes we make most often:
- Page content in the HTML response, not only after JavaScript runs
- Forms with real labels, clear error messages, and no hover-only steps
- A stable URL for every service, product, and policy
- robots.txt rules set per crawler, and an llms.txt that points to your key pages
- The same name, address, phone, and hours everywhere your business is listed
- An MCP server or structured feed when assistants need live data like inventory or availability
Who is reading your site
Each AI company runs separate crawlers for search, for user requests, and for model training, and you can allow one while blocking another. By default we let the search and user bots in and decide on the training bots with you.
| Crawler | Operator | What it does | Our default |
|---|---|---|---|
Googlebot | Indexes pages for Google Search, including AI Overviews | Allow | |
Bingbot | Microsoft | Indexes pages for Bing, which also grounds Copilot answers | Allow |
OAI-SearchBot | OpenAI | Finds pages to show and cite in ChatGPT search | Allow |
ChatGPT-User | OpenAI | Fetches a page when a person asks ChatGPT to | Allow |
Claude-SearchBot | Anthropic | Indexes pages for Claude's search results | Allow |
Claude-User | Anthropic | Fetches a page when a person asks Claude to | Allow |
PerplexityBot | Perplexity | Indexes pages for Perplexity answers | Allow |
GPTBot | OpenAI | Collects content for model training | Your call |
ClaudeBot | Anthropic | Collects content for model training | Your call |
Google-Extended | A robots.txt token for Gemini training and grounding; it does not affect Search | Your call |
User agents can be faked, so we verify the major crawlers by published IP range or reverse DNS before trusting what the server logs say.
The SEO toolkit
Semrush, Ahrefs, and Moz estimate your traffic; Search Console reports it. We use the estimates for competitor research and check anything about your own site against Search Console.
| Tool | What we use it for |
|---|---|
| Semrush | Keyword research, position tracking, competitor gap analysis, and scheduled site audits |
| Ahrefs | Backlink and referring-domain audits, content gap research, and rank tracking |
| Moz Pro | Domain Authority benchmarks, link research, and keyword difficulty |
| Google Search Console | First-party clicks, impressions, indexing status, and Core Web Vitals |
| Bing Webmaster Tools | Bing indexing and IndexNow submission. Bing's index also feeds Copilot |
| PageSpeed Insights + CrUX | Field data for LCP, INP, and CLS from real Chrome users |
05 · How a project runs
Audit first, then build on a preview link
Every engagement starts with an audit, even when you already know you want a new site. The audit tells us what is worth keeping, which is usually more than people expect.
- 01
Audit
We crawl the site, read the tag setup, check DNS and email records, and ask the AI assistants what they say about you. You get a written list of problems in priority order.
- 02
Plan
Scope, platform, and measurement plan agreed in writing before work starts: what gets built, which conversions get tracked, and who owns each account.
- 03
Build
Work ships to a private preview link. You click through every page and every form before anything touches your live domain.
- 04
Launch + measure
DNS moves at a planned time, redirects and tags are verified on the live site, and the first month's report is checked against each platform's own numbers.
Questions
Before you book
Something else on your mind? Email hello@monolithdesign.agency or call (509) 352-8993.
How is Monolith Web related to Monolith?
Monolith Web is Monolith's web development and AI search practice. Brand, video, campaigns, and AI training live at monolithdesign.agency. The same team runs both, with the same booking link.
Do you only build on Next.js?
No. Next.js on Vercel is the default for new custom builds because it is fast and easy to maintain. We also build on and repair WordPress, Shopify, Webflow, Wix, Squarespace, and Drupal when your team needs to keep editing there.
Can you fix tracking on a site you didn't build?
Yes. We audit the GA4 property, the Tag Manager container, and every ad pixel, write down what fires and what double counts, then rebuild from one measurement plan.
Who owns the domain, hosting, and analytics accounts?
You do. We work from access you grant, under accounts registered to your business. If we stop working together, there is nothing to transfer back.
What is agentic search?
Search where an AI assistant does more than answer. It opens websites for a person to compare options, fill in forms, or book appointments. Sites that serve content without scripts, label their forms, and state facts in plain text are easier for those agents to use.
Can you guarantee rankings or AI citations?
No one can promise that. Search engines and AI assistants decide what they show. We work on the causes: crawl access, page speed, clear answers, accurate business facts, and structured data, and we report what changes.
Do you work with agencies?
Yes. We take on development, tracking, and technical SEO for agencies, under your brand or ours.
Where is the team based?
Monolith is based in Spokane, Washington, and works with businesses and agencies across the United States. Businesses in the Inland Northwest can start with Monolith's Spokane web design page.
Next step
Bring your URL. We'll show you what the stack is doing.
A 30-minute call to look at your site, your tracking, and how search engines and AI assistants see you. You leave with a short list of what we would fix first, whether or not we work together.