
Get found in search. Get recommended by AI. Make the next action easy.
Customers now find businesses three ways: a search results page, an AI-generated answer, and an assistant that compares options and takes the next step for them. Growth Nuts works on all three, starting with whichever is costing you the most inquiries, bookings or sales.
01Three moments
Three moments where a customer is won or lost
Every sale that starts online passes through the same checkpoints. Lose the customer at any one and the steps after it never happen.
Found
Your pages are crawlable, indexed and match what buyers search for. This is SEO.
Recommended
AI search tools understand what you offer and have accurate, citable reasons to suggest you. This is GEO, or generative engine optimization.
Acted on
When an assistant tries to book, request or buy, your site and systems let it finish, with the customer’s approval. This is agentic AI optimization, and for some businesses agent and API readiness.
02Which problem is yours
Which problem sounds like yours?
Customers cannot find us
You do not show up when people search for what you sell, or you show up for the wrong searches. SEO
AI answers overlook or misrepresent us
AI search tools name competitors, skip you, or get your services, prices or service area wrong. GEO and AI search
Assistants cannot complete tasks on our site
An AI assistant reaches your site but cannot confirm what you offer or finish the booking, request or purchase. Agentic AI optimization
Our systems need to work with customer agents
You want software acting for customers to get accurate answers from, or take permitted actions in, your systems. Agent and API readiness
03How the services differ
How the services differ, and how they work together
SEO has not been replaced. AI search tools draw on pages that are crawlable, indexed and clearly organized, the same pages that rank in Google. An assistant cannot complete a booking on a page it cannot find or understand.
So we sequence the work in that order unless the evidence points elsewhere: fix discoverability, then representation in AI answers, then task completion, then system integration when a real use case justifies the engineering.
SEO and GEO are ongoing services with defined methods. Agentic AI optimization is sold as scoped audits and implementation projects. Agent-to-agent and API readiness are sold as pilots, because they involve custom engineering in your systems and the standards behind them are still settling. Every proposal says which category the work falls into.
You may see labels such as AAIO, A2AO or APIO. They are descriptive labels Growth Nuts uses for this work, not established industry standards.
Compare the five services side by side
| Service | Customer goal | Work performed | How it is measured |
|---|---|---|---|
| SEO | Get found when people search for what you sell. | Crawl and indexing fixes, site architecture, intent-mapped pages, content, local signals, legitimate authority, conversion tracking. | Qualified organic traffic, leads, and attributable revenue. |
| GEO and AI search | Be described accurately, and recommended, in AI-generated answers. | Buyer-question testing, entity consistency, clear service and policy information, evidence-backed content, matching structured data, third-party coverage. | Sampled mention, recommendation and citation rates, accuracy of what answers say, observable AI referrals. |
| Agentic AI optimization | Let an assistant evaluate your offer and complete a task for the customer. | Task inventory, authorized agent testing, semantic pages, labeled forms, clear errors, accurate availability and policy data, workflow fixes. | Completion rate on a defined task set, failure points, fixes confirmed by regression tests. |
| Agent-to-agent readiness Pilot | Let a customer’s agent communicate with an agent or system you operate. | Use-case definition, capability description, discovery and communication design, identity and permissions, human approval points, audit trails. | Successful authorized exchanges in a sandbox, errors, and handoffs to people. |
| API readiness for AI agents Pilot | Make accurate business information and permitted actions usable by software. | API inventory, specifications, data models, scoped access, rate limits, useful errors, idempotency, versioning, sandbox, monitoring. | Successful authorized requests, data accuracy, latency, failures, duplicate actions prevented. |
04Three ways to engage
Three ways to engage
All pricing is set from a written scope after a short review of your site and goals.
Search & AI opportunity audit
A defined assessment of where customers are being lost across search, AI answers and assistant tasks.
Who it suits, what is included, access and scope
- Who it suits
- Businesses that want evidence before committing to ongoing work, or that suspect AI tools describe them badly.
- What is included
- Technical and content review, a documented buyer-question test across applicable AI search tools, a small assistant task test where relevant, and a competitor comparison.
- What you receive
- Findings with evidence, the business implication of each in plain English, and a prioritized list of next steps.
- How success is evaluated
- Every finding is evidenced, and every recommendation names an owner, the effort involved and the expected effect.
- Access needed
- Read access to Search Console and analytics, your core services and markets, and details of any booking or checkout tool.
- What sets the scope
- Number of sites, locations and service lines, and how many buyer questions and tasks are tested.
Implementation sprint
A scoped project that fixes the most important discoverability or task completion problems.
Who it suits, what is included, access and scope
- Who it suits
- Businesses with a known, prioritized problem, usually identified in an audit.
- What is included
- Crawl and indexing fixes, page content and structure, entity and listing consistency, structured data, and form or booking flow changes.
- What you receive
- Completed changes, a change log, before and after evidence, and retest results against the baseline.
- How success is evaluated
- The agreed problems are fixed and confirmed by retesting the same checks, questions or tasks.
- Access needed
- CMS or code repository, a staging environment where one exists, and admin access to form or booking tools.
- What sets the scope
- The number and complexity of fixes, and whether changes require developer work in your systems.
Ongoing growth partnership
Continued SEO, content improvement, sampled AI visibility monitoring, conversion analysis and maintenance.
Who it suits, what is included, access and scope
- Who it suits
- Businesses where search already drives, or should drive, a meaningful share of revenue.
- What is included
- Ongoing SEO and content work, AI visibility retests on an agreed schedule, conversion analysis, and technical maintenance.
- What you receive
- Monthly reporting in plain English tied to leads and revenue, plus a rolling priority list.
- How success is evaluated
- Qualified leads and attributable revenue against the baseline agreed at the start.
- Access needed
- Analytics, Search Console, CMS, and CRM or call tracking where available.
- What sets the scope
- Market competition, number of locations and services, and content volume.
Agent-to-agent and API readiness pilots are scoped separately, each with its own sandbox and success criteria.
05How we measure
How we measure, and the limits
Every report separates three kinds of result: directly observed (logged events, captured answers, completed test tasks), estimated (modeled values, labeled as estimates), and influenced (outcomes the work plausibly contributed to but cannot isolate).
A conversion is counted once, in the channel where it was recorded, so SEO and AI search are never both credited with the same lead.
Every metric, its source and its limits
| Work | Metric | Denominator | Source | Limits |
|---|---|---|---|---|
| SEO | Qualified organic leads and attributable revenue | Organic sessions; total leads for share of pipeline | Search Console, GA4, CRM or call tracking | Revenue is reported only where your CRM records it. Attribution depends on tracking quality. |
| GEO | Mention, recommendation and linked citation rates | Answers captured for the agreed question set, per platform | Dated test captures with query, locale and conditions | A sample. Answers vary by user, run and day, so this is not total exposure. |
| GEO | Representation accuracy | Factual statements about you in captured answers | Captures checked against a fact sheet you approve | Only covers the questions tested. |
| GEO | Observable AI referral conversions | Sessions with an identifiable AI referrer | GA4 referrer data | Many AI-influenced visits arrive with no referrer. Google AI Overviews and AI Mode traffic is reported inside normal Search Console web data, not separately. |
| Agentic AI optimization | Task completion rate | Task attempts in the defined set, per assistant and run | Recorded sessions in an authorized test environment | Assistants change often. Results describe the test date and conditions. |
| Agent and API | Successful authorized requests | All authorized requests in the period | Server logs and monitoring | Sandbox results can differ from production load. |
| Agent and API | Data accuracy | Sampled responses | Automated checks against the system of record | Sample based. |
| Agent and API | Latency and failures | All requests | Monitoring | Reported as median and 95th percentile latency, and error rate by type. |
| Agent and API | Duplicate actions prevented | Retried action requests | Idempotency logs | Covers only actions built with idempotency keys. |
06Request an assessment
Request a search & AI assessment
What happens after you submit
- You land on a confirmation page only once our system has accepted the request. If it fails, the form tells you and keeps what you typed.
- We look at the site and the problem you chose.
- We reply by email with what we saw at a first look and which next step fits: an audit, a sprint, or nothing at all.
- If you want to go ahead, you receive a written scope and price before any paid work begins.
Prefer to start on your own? The free site scan runs a live technical, content and AI readability check with no account.
07Questions
Frequently asked questions
Has AI search replaced SEO?
No. AI search tools and assistants rely on content they can find and trust, and most of that work is still SEO: crawlable pages, clear information and real authority. Google states that its AI features have no additional requirements beyond being indexed and eligible to appear in Search.
Where should we start?
Start with the checkpoint where you are losing the most customers. For most local businesses that is still search visibility or conversion. If you already rank well, the next question is usually how AI answers describe you.
Can you guarantee ChatGPT or Google’s AI recommends us?
No, and nobody can. AI answers vary by user, wording and day. We improve the information those tools rely on, test a defined set of questions on a schedule, and report what changed.
Do small local businesses need agent or API work?
Usually not yet. A local business with a working booking platform typically needs clearer pages and better use of that platform’s existing integrations. Custom agent or API work makes sense when a specific, valuable task requires it.
What happens after we request an assessment?
We review the site and the problem you chose, then reply by email with a first view and a recommended next step. Paid work starts only after you approve a written scope and price.
How we work
Growth Nuts is a boutique practice, so whoever scopes your plan also does the work and answers for it. Our method was proven in cannabis, where paid ads are mostly off the table: our founder built Green Rush Daily into a $7 million sale to High Times almost entirely on organic search.
Not sure which of these you need?
Tell us the problem in your own words. We will tell you where we would start, and whether this work is worth it for your business.