Demand
- Technical SEO
- Service and comparison content
- Local and entity consistency
- Structured data
- AI answer-engine foundations
The category, explained
Growth engineering: building the machine that produces customers
A demand problem and a growth-system problem need different treatments. The first needs more qualified attention; the second needs a better path from attention to revenue.
In this model, a demand problem means qualified attention is missing. The response is to create or capture more of it.
A growth-system problem means demand already exists and the business loses some of it between the first search and the signed job. Adding more demand before fixing that path can increase the volume reaching the same failure points and make acquisition less efficient.
Growth engineering is the work of finding where that loss happens and building a system designed to reduce it.
The commercial model
Seven layers. One path to revenue.
Demand, conversion, capture, sales, retention, measurement, and automation are the public model. For the audit, sales and retention are decomposed into routing, follow-up, repeatable sales, and compounding. That produces nine testable checkpoints without inventing a second system.
Demand → conversion → capture → sales → retention → measurement → automation Every handoff visible
- 01
Demand
Earn attention through search, local visibility, useful content, and answer-engine eligibility.
- 02
Conversion
Turn relevant attention into a clear decision with positioning, UX, proof, speed, and a useful next step.
- 03
Capture
Record every enquiry, its source, and enough context to qualify and route it without inbox archaeology.
- 04
Sales
Give each opportunity an owner, a next action, a response standard, and a pipeline another person can run.
- 05
Retention
Keep useful follow-up, recall, reactivation, and referral prompts running after the first conversation.
- 06
Measurement
Carry source and campaign context through to qualified opportunities and closed revenue.
- 07
Automation
Remove repeatable manual work while keeping monitoring, failure alerts, and a human fallback.
Measurement and automation feed the next demand and retention decisions. The audit decomposes sales and retention into routing, follow-up, repeatable sales, and compounding, creating nine checkpoints that can be tested separately.
Definition
What growth engineering is.
Growth engineering is the practice of designing, building, and instrumenting the connected systems a business uses to attract, convert, follow up with, serve, and measure customers, so that revenue capacity depends less on the founder’s personal attention.
It borrows its method from engineering rather than from marketing. Map the system. Find the constraint. Change one thing. Measure what happened. Repeat.
The word “engineering” is doing real work in that sentence. Marketing optimises a channel. Consulting produces a recommendation. Engineering produces a working system with its behaviour instrumented, its failure modes handled, and its documentation written. That is the deliverable here: a machine, not a campaign and not a slide deck.
The problem
Why businesses stall even when the marketing works.
An established service business that grew through referrals, reputation, and the founder’s effort may still rely on one central operating system: the founder. Important decisions route through one person’s memory, judgement, and calendar.
That can work for a time. Common symptoms of the resulting constraint include:
- Leads arrive through several channels and live in several places.
- Response time depends on who happened to check the inbox.
- The CRM was set up once and now holds a partial, out-of-date picture.
- Follow-up happens for the leads someone remembers.
- Source-to-revenue attribution is incomplete, so budget decisions depend heavily on instinct.
- Hiring adds capacity without resolving the missing process the new person needs to follow.
Those are primarily system problems rather than channel problems. Adding marketing can increase the volume flowing into a path that already leaks, so spend may rise without a proportional change in revenue.
The model
Nine checkpoints inside the seven layers.
The finer audit view decomposes sales and retention into routing, follow-up, repeatable sales, and compounding. A checkpoint may be explicit or informal; the useful question is whether its behaviour is deliberate and measurable.
Audit view / nine diagnostic checkpoints Every handoff visible
- 01DemandSearch + AI
- 02ConversionClear decision
- 03CaptureOne record
- 04RoutingRight person
- 05Follow-upNo forgetting
- 06SalesRepeatable
- 07MeasurementSource to sale
- 08AutomationHands off
- 09CompoundingEvidence
Checkpoint 9 feeds evidence back into checkpoint 1. That loop is the point.
| Checkpoint | What it does | How it fails | How you’d know |
|---|---|---|---|
| 01 · Demand | Do the right people find us? | Invisible in search, wrong keywords, wrong audience. | Low branded traffic and no organic inquiries. |
| 02 · Conversion | Do they believe us enough to act? | Unclear offer, slow pages, no obvious next step, no trust signals. | Traffic without inquiries. |
| 03 · Capture | Does every inquiry get recorded? | Scattered channels, missed calls, silent form failures. | You find leads by searching your inbox. |
| 04 · Routing | Does it reach the right person fast? | Manual triage, unassigned leads, no response standard. | First-response time is unmeasured or regularly misses the business’s stated standard. |
| 05 · Follow-up | Does it get chased without someone remembering? | Depends on memory, stops after one attempt. | Lost or stalled deals have no recorded decision, reason, or next action. |
| 06 · Sales | Is the process repeatable by someone other than the founder? | The close is undocumented and unrepeatable. | Nobody else can run a deal end to end. |
| 07 · Measurement | Can we tell what produced revenue? | No attribution, vanity metrics, disconnected tools. | "I think most of it comes from referrals." |
| 08 · Automation | Does the machine run without hands on it? | Manual work, or automation nobody monitors. | The same tasks eat the same hours every week. |
| 09 · Compounding | Do we improve from evidence? | No experiments, no baseline, no review. | Busy, flat, and unsure why. |
Working heuristic: improve the active constraint before adding volume elsewhere. Ariadne therefore starts with a diagnosis, because spending on discovery while the loss sits later in the path can amplify waste.
Scope
What a growth engineering engagement includes.
Conversion
- Positioning
- Page architecture
- Copy
- Design
- Build
- Performance
- Conversion events
Capture
- Unified intake across forms, calls, and chat
- Qualification logic
- Validation and spam handling
Routing
- Assignment rules
- Response standards
- Notifications
- Escalation paths
Follow-up
- Consent-aware sequences under CASL
- Reminders
- Re-engagement
- Failure alerts
Sales
- CRM architecture
- Pipeline stages that reflect how you sell
- Documented process
Measurement
- Source-to-customer attribution
- Conversion tracking
- Reporting a founder can read
Automation
- Workflow automation
- Integrations
- Document and quote generation
- Applied AI where it helps
Compounding
- Experiment design
- Baselines
- Review cadence
A first engagement normally touches two or three of these, not nine. The diagnosis determines that scope before a build is proposed.
How it fits together
How the systems interact.
The layers are not independent, which is the entire argument for building them together.
SEO sends traffic to conversion. If conversion is weak, SEO spend may produce visits without a corresponding increase in qualified enquiries.
Conversion produces inquiries for capture. If capture is scattered, conversion improvements are invisible, because there is no consistent record to measure against.
Capture feeds routing and follow-up. A well-designed intake form has limited value if the resulting email sits unread, which is a CRM and sales systems problem.
All of it feeds measurement, which can reduce the cost of learning and give the next decision stronger evidence. Without it, improvements carry more uncertainty and budget conversations rely more heavily on opinion.
Automation sits across the whole loop rather than at the end of it. Automate a broken process and you get the wrong outcome faster, with fewer people watching. Design the process first. Automate second. Keep a manual path and an alert because any automation can fail.
Outcomes
What changes afterwards.
- In-scope inquiries enter a defined record, qualification, routing, and acknowledgement path
- Revenue-producing channels can be identified and compared where source and sales data are available
- First-response time is measured against a baseline and reduced where routing is the constraint
- Selected repetitive steps run automatically, with alerts and a manual fallback
- A new hire gets a documented process instead of an apprenticeship
- Marketing spend can be increased or cut based on evidence
- The system can handle more qualified demand without requiring the same proportional increase in manual coordination
What does not change
You still have to be good at the work. Growth engineering builds the machine that brings customers to a good business. It cannot make a mediocre service sell.
In practice
What this looks like in practice.
These are illustrative scenarios drawn from common patterns, not client case studies. Real case studies will appear in Work as engagements complete.
The referral-dependent trades business
Twenty staff, strong reputation, pipeline entirely word of mouth. The constraint is not demand; it is that no channel exists that they control.
- Diagnosis
- Diagnostic checkpoints 1 and 3.
- Build
- Service pages targeting the jobs people search for, unified intake across phone and web, and attribution so the first paid experiment is measurable.
- Deliberately skipped
- CRM replacement, because the existing one was adequate once intake fed it correctly.
The consultancy that generates leads and loses them
Eight staff, healthy inbound from content and referrals, and a close rate that drops the busier they get.
- Diagnosis
- Diagnostic checkpoints 4, 5, and 6.
- Build
- Routing with a response standard, a follow-up sequence that runs whether or not anyone remembers, and a pipeline with stages someone other than the founder can operate.
- Deliberately skipped
- SEO, because the demand already existed.
The service business drowning in admin
Twelve staff, growth constrained by delivery overhead rather than acquisition.
- Diagnosis
- Diagnostic checkpoint 8, with a measurement gap at checkpoint 7.
- Build
- Quote and document generation, scheduling and reminder automation, and reporting that makes the remaining bottleneck visible.
- Deliberately skipped
- The website rebuild they originally asked for.
Fit
Who this is for.
A good fit
- Owner-led service business, roughly 3 to 25 staff
- One new customer is worth at least a few thousand dollars
- The sale starts with a quote, consultation, assessment, or booking
- Leads arrive through more than one channel and live in more than one place
- The founder is still personally involved in most deals
- Marketing spend exists but attribution does not
Not a fit
- Pre-revenue, with no existing demand to route
- Wants a cheap website and nothing after the form submit
- Ecommerce or DTC. Different mechanics, different stack
- Expects Ariadne to act as legal or compliance counsel, replace an EMR or professional practice system, or handle regulated data without the appropriate agreements and client-reviewed constraints
- Anyone whose opening request is “run our ads”
These boundaries keep the work focused on systems Ariadne can responsibly design and build. How this compares to a marketing agency
Straight answers
Common questions.
If yours is more specific, put it in the audit form.
support@ariadne.fyiIs growth engineering the same as growth hacking?
No. Growth hacking is commonly associated with tactic-led experiments in a channel, often for product companies. Growth engineering is system-led and infrastructural, and it treats operations and marketing as possible locations of the constraint.
Do I need to be technical?
No. You need to be able to describe how a customer currently reaches you and what happens after they do. The engineering is my job.
How is this different from hiring a fractional COO or CMO?
A fractional executive usually focuses on judgement, leadership, and management. Ariadne’s scope includes implementation: writing code, configuring the CRM, and wiring attribution. The roles can complement each other, but they are not the same engagement.
How long before it shows up in revenue?
There is no defensible universal timeline. Routing and follow-up can affect demand that already exists soon after launch, while conversion and organic work need enough traffic and sales-cycle data to evaluate. The audit establishes the baseline and measurement window; it does not promise a revenue date.
What if I only need one part?
Then buy one part. The audit exists to find out which.
Can you work with our existing agency?
Yes. They can generate demand while Ariadne improves how the business captures, routes, and measures it. Shared attribution can give both parties a clearer basis for evaluating the relationship.
Start here
Start with the diagnosis, not the build.
A growth system audit reviews nine diagnostic checkpoints across the seven-layer system, ranks the two or three highest-priority constraints supported by the available evidence, and gives you a costed plan you could execute with anyone. $950 CAD, five business days, credited in full against a build.