Custom WhatsApp and web AI agents for enquiry qualification, booking, quoting, dispatch, and customer support. Built on N8N, OpenAI, and Claude, wired into Google Sheets, Airtable, HubSpot, or Pipedrive. From £4,500 one-off, optional monthly maintenance from £400. Three to five week delivery. Full ownership of the workflow at handover.
Not a chatbot that answers FAQs. An agent that moves work through a defined pipeline with real integrations and real handover rules.
A well-built AI agent is a pipeline, not a chat window. A customer or prospect makes contact through a channel (WhatsApp, web, email). The agent classifies the intent, asks the few questions needed to qualify, pulls data from your systems to answer or quote, books or dispatches where that is the next step, logs the interaction in your CRM, and passes to a human when the conversation needs judgement the model should not make. The whole thing takes seconds and runs 24 hours a day, but the quality depends almost entirely on the prompt engineering, the integrations, and the handover rules, not on which model sits behind it.
The model (OpenAI GPT or Anthropic Claude) is the reasoning layer. It decides what the user is asking and produces the response. The orchestration layer (N8N) is the pipe that routes the conversation through your systems. The data layer (Sheets, Airtable, HubSpot, Pipedrive) is where the business context lives. Treat the model as a component, not the product. The product is the pipeline.
Agents pay back fastest when you have repetitive qualification, booking, or dispatch work happening at volume. A recovery business taking 40 WhatsApp enquiries a day, half of which are outside service area, benefits enormously. A bespoke law firm taking five complex enquiries a week probably does not, because the qualification is where the firm's judgement lives. We will tell you honestly on the first call which category you fall into, and if the answer is the second, we will refund any retainer paid and suggest other places to spend the money.
Sectors and workflows we have delivered in the last 18 months. If your use case is not here, ask, we are happy to say honestly whether it fits.
WhatsApp agent takes the enquiry, checks location, qualifies vehicle and fault, quotes distance-based pricing, dispatches to the nearest available driver, logs to Sheets or HubSpot. Replaces overnight dispatcher on most enquiries.
First-touch enquiry handler across web and WhatsApp. Qualifies appointment type (consultation, treatment, follow-up), checks practitioner availability, offers slots, books, logs to PMS. Front-desk time on inbound enquiries typically drops 30 to 50%.
Web widget on Shopify or WooCommerce. Handles order status, returns workflow, basic product questions, with handover on anything complex. Support ticket volume to humans typically drops 40 to 60%.
Intercepts demo requests on the marketing site, qualifies ICP fit (company size, role, use case), books qualified demos directly into the sales calendar, redirects unqualified to self-serve or nurture.
Plumbing, electrical, HVAC. Agent qualifies the job, pulls the pricing matrix, quotes for standard jobs, books site visits for bespoke. Pricing stays consistent, quotes go out in seconds.
Takes new enquiries across web and WhatsApp, triages by practice area, qualifies for conflict checks, books consultations for valid enquiries, passes to solicitors with full context.
Five weeks from discovery to live agent, for standard-scope builds. Each week has specific deliverables rather than vague milestones.
Kick-off call, use-case mapping, integration inventory, handover rule drafting, edge case cataloguing. Written scope document by end of week one.
System prompt drafted, tested against 30 to 50 representative enquiries (real or simulated), iterated until response quality is stable. First demo mid-week two.
N8N workflow built, OpenAI or Claude connected, CRM and data-source integrations wired, handover triggers configured, WhatsApp or web channel integrated.
Agent runs in shadow mode against real enquiry volume. Outputs reviewed daily. Prompts tuned, integrations hardened, handover thresholds calibrated.
Go-live, full documentation, credential handover, operating manual. Two weeks of post-launch monitoring included from here.
If you opt in to the maintenance retainer: weekly review dashboard, monthly prompt tuning, model migrations when OpenAI or Anthropic ship updates, integration monitoring.
Single-channel single-workflow at £4,500. Multi-workflow at £9,000. Enterprise dispatch and custom integrations at £18,000. Fixed price after the discovery call, not time-and-materials.
Baseline deliverables at every tier. Scope and integration depth scale with tier.
Written scope document by end of week one. Covers use case, integrations, edge cases, handover rules.
Drafted, tested against representative enquiries, iterated to stable response quality.
Full orchestration workflow built, documented, and delivered. Self-hosted on your infrastructure or N8N Cloud, your choice.
OpenAI or Anthropic Claude wired in, with honest guidance on which model fits the task and cost envelope.
HubSpot, Pipedrive, Salesforce, Zoho, or bespoke via webhook. Contact creation, conversation logging, next-action flagging.
Sheets, Airtable, or bespoke. Agent reads live business context rather than hallucinating answers.
Configurable triggers: low confidence, out-of-scope intent, specific keywords. Full transcript passed to human.
Full operating manual, credential handover, optional two-hour team training session.
Weekly sample of live interactions for quality review. Essential for catching drift early.
You own the workflow, the prompts, the integrations, the code. No vendor lock-in.
Monthly prompt tuning, integration monitoring, model migration. From £400/mo.
One number after discovery, not time-and-materials billing.
Standard integrations below. Custom integrations quoted on the discovery call.
Fixed-price builds, quoted after the discovery call. Optional monthly maintenance billed separately.
The £4,500 Single tier covers a standard-scope build: one channel (WhatsApp or web), one workflow (qualification, booking, or support), one data source, and one CRM destination. The £9,000 Multi-workflow tier doubles scope: two channels together, two to three workflows, and conditional routing. The £18,000 Enterprise tier handles multi-channel dispatch, quoting logic, custom integrations, and multi-language prompts. Each tier is a fixed price, not a starting price.
The optional monthly maintenance retainer starts at £400 and covers weekly review sampling, monthly prompt tuning, integration monitoring, and model migrations when OpenAI or Anthropic ship new versions. It is not required, and a sensible in-house team can take the agent on maintenance themselves after handover. We document everything with that assumption in mind.
Payment is 50% on kick-off, 50% on go-live, consistent with how we invoice website projects. Maintenance is billed monthly in advance. Everything month-to-month after the first 90 days. No clawback on the build fee if you leave the maintenance retainer.
Receipts available on request, happy to show live Search Console on a call.
Built 40+ postcode-level landing pages, cleaned up a messy schema stack, deployed a WhatsApp AI dispatch agent, earned local press across east London recovery services.
Rebuilt an ageing site, added product & review schema, rewrote category pages in plain English.
180-page city-service template that reads human, plus a WhatsApp agent handling 60% of intake.
Four verified reviews from active engagements. Every review ships as schema.org Review markup alongside the visible quote, same claim on screen and in the structured data.
Three years in and still the best SEO money I have ever spent. Map Pack visibility across 40+ London postcodes, zero nonsense in the reporting, and I can text Syed directly when something breaks.
Organic revenue up 185% in 14 months. Product schema rebuild alone lifted rich-result capture by ~40%. No 12-month lock-in, month-to-month, which meant I could judge the work on results rather than on contract friction.
Moved from an NYC agency that billed $9k/month for junior-delivered work. Two years later, 23 practice-area terms on page one and qualified demos up 180%. Senior time, in USD, month-to-month, what US SaaS SEO should be.
Four-clinic group across Sydney. GBP work, postcode landing pages, review pipeline that actually complies with Google's rules. Patient bookings from organic up 3x in the first year. Remote but genuinely responsive.
SEO is the foundation. AI and custom web builds are how I ship outcomes in 2026, all connected, all from the same hand.
Crawl audits, schema that validates, internal linking, postcode-level landing pages, GBP, Map Pack, the foundation that makes everything compound.
Custom WhatsApp and web agents handling enquiries, quoting, booking, and dispatch. N8N, OpenAI, Gemini.
Custom sites on WordPress, Next.js, or hand-written HTML. Fast, SEO-ready, Core Web Vitals green from day one.
Topical maps that close ranking gaps. Editorial briefs your writers can follow. Digital PR that survives core updates.
Reporting, lead routing, content pipelines. If a task is repetitive and mechanical, I'll automate it with N8N.
Written SEO diagnostic with a ranked fix list. Two-week turnaround. Often the right starting point.
Replatforms, redesigns, rebrands. I protect rankings through the change, the riskiest work in SEO, done right.
Four tiers. Every tier is hand-coded, no Wix, no Elementor, no copy-paste from a template marketplace. Schema, sitemap, Search Console and Analytics configured on every project. 90+ Lighthouse speed target where technically possible. Express turnaround on sites up to 10 pages: 2 to 3 working days for an extra £500, or same-day launch for £1,000, subject to all content and brand assets supplied on day one. Lower than traditional UK agencies, because we don't carry London agency overhead.
Hand-coded 5-page site for founders validating a new business or single-service local operators.
Most common tier for growing SMEs. Full sitemap, services, about, blog shell, custom UI/UX in Figma.
Full UI/UX system plus hand-coded Next.js or WordPress build for businesses with multiple service lines.
Shopify / Saleor headless, multi-language hreflang, CRM / CMS / ERP API integrations.
The difference between a pitch deck and the people shipping your work is the difference between “scalable” and delivered.
A short introduction, your site URL, and what you’re trying to achieve. If it’s a fit, we’ll book a 30-minute call.
Every placement is negotiated and published by hand through a six-year network of editors and journalists. We never use AI bots or PBNs, they get detected, they get demoted, and your domain pays the price.
Ten contextual do-follow links from real UK and international sites with Domain Rating 50 and above. Topically relevant. Placed inside genuine editorial content, not link-farm footers. Index report delivered within 4 weeks.
Ten earned placements on national UK and US media with Domain Rating 70 and above, the kind of coverage that shifts rankings in competitive verticals and doesn't disappear in the next core update. Written, pitched, and placed by our PR team.
Google's last five core updates have all sharpened link-spam detection. Bulk-placed links from AI-generated host sites and public blog networks are being flagged faster than they can be bought. Our model is slower and costs more per link, but the placements survive every update and compound in value the longer they stay live.
Most agency SEO deliverables end at a recommendations document the client's developer never gets around to implementing. We write the schema, ship the SSR refactor, and merge the internal-link rebuild ourselves. The SEO work that needs code ships in the same sprint the audit flagged it.
Every client gets the same senior operator from first call to monthly review. Continuity is the product.
Two weeks. Crawl, keyword gap, backlink profile, on-page health. Written report, ranked fix list.
Schema, technical debt, site build or repair, internal linking. The work that makes everything compound.
Close topical gaps. Earn links honestly. Deploy AI agents where they save real hours, not just look clever.
Monthly call. Plain-English report. What moved, what didn't, what's next. Leave any time.
Syed leads the strategy and writes the monthly notes. Behind him is a tight network of expert developers and manual link-earning partners built over six years. Everything ships fast, nothing is outsourced to an AI bot that will earn your domain a penalty in the next core update.
For the right use case, yes. For the wrong use case, it will cost you more than a human answer would. Agents pay back fastest when you have repetitive qualification, booking, or dispatch work happening at high volume. If your enquiries are mostly nuanced first-time consultations, a human handles those better and the agent should stay on the first-touch layer only. We tell you honestly on the first call whether your use case justifies the build.
N8N as the orchestration layer (self-hosted on the client stack or cloud, your choice), OpenAI or Anthropic Claude as the reasoning layer depending on the task, and Google Sheets, Airtable, HubSpot, or Pipedrive as the data layer depending on your existing systems. WhatsApp integration uses the official Cloud API. Web chat uses a hand-coded widget that we host or embed. No low-code lock-in, no vendor dependency that forces you into a monthly platform fee forever.
Three to five weeks for most business agents. Week one is discovery and prompt engineering, week two is integrations and wiring, week three is testing and QA against real enquiries, weeks four and five are handover and monitoring. Complex dispatch or quoting agents with heavy external data dependencies can push to six or seven weeks, and we flag that on the first call rather than promising a timeline we cannot hit.
Every agent we build ships with human handover triggers. When the model is not confident, when the enquiry falls outside its scope, or when a specific keyword appears (refund, complaint, urgent), the conversation routes to a human with full transcript context. We also build a weekly review dashboard so your team can audit a sample of interactions and flag prompts that need tuning. Agents do not drift in isolation, they drift if nobody is watching.
Yes. Standard integrations with HubSpot, Pipedrive, Salesforce, Zoho, and most modern CRMs via webhook or native node. We also write custom integrations where the CRM is older or bespoke. The agent typically creates the contact, logs the conversation, and flags the next action, so your sales team picks up context rather than restarting the conversation.
It is the starting price for a standard-scope agent: one channel (WhatsApp or web), one data source, one workflow (qualification, booking, or support). Multi-channel, multi-workflow, or custom-integration builds scale from there. Most builds land between £4,500 and £9,000. Enterprise dispatch agents with live routing logic can sit higher. We quote a fixed price after the discovery call, not a time and materials invoice that balloons.
Prompts drift, APIs change, models update, and business rules evolve. The optional maintenance retainer from £400 per month covers monthly prompt tuning, integration monitoring, model migrations when OpenAI or Anthropic ship a new version, and a weekly review dashboard of agent performance. Not required, and plenty of clients maintain agents in-house after handover. We hand over full documentation either way.
You do. Full ownership of the N8N workflow, all prompts, all integrations, all code. We hand over the repo or workflow JSON, documentation, credentials, and a written operating manual. No vendor lock-in, no ongoing licence fee to us, no dependency on our hosting. If you want to take the whole thing to another developer tomorrow, you can.
Send the channel (WhatsApp or web), the workflow (qualification, booking, dispatch, support), and rough monthly volume. We reply inside a working day with an honest view on fit and a scoped build.