The HubSpot blueprint is a 20-section, implementation-ready architecture document written from your HubSpot discovery — precise enough for a certified HubSpot partner to start building without questions, and readable enough for a CEO to approve with confidence.
Twenty sections from your discovery — data model, licensing, pipelines, Breeze — audited before you see them. The portal is never a blank page.
The structure is fixed. Thin discovery still gets a recommended default, marked for confirmation — never a dropped section a partner has to invent on the day.
Every technical section opens in plain English, then gets precise enough to build from. A CEO can approve it. A certified partner can start without questions.
A second pass scores completeness, TCO arithmetic, tier gates and traceability back to discovery. Miss the bar, and the blueprint does not ship.
Hub-by-hub why this tier, seats by role, and a first-year TCO table in your currency — before anyone is invoiced for seats you will not use.
Objects, associations, pipelines and gates are decided here so records never land in the wrong place — and year two is not a rebuild of a messy CRM.
Every section is grounded in the live documentation index plus 26 expert frameworks, then track-assigned to the tier you can actually switch on.
Every technical section opens with a plain-English explanation of what the thing is and why it matters — then gets precise enough to build from.
"Your data model is the blueprint for how information lives in HubSpot — which objects store what, and how they connect." Each section explains itself before it specifies anything, and the executive summary states your 6-month success criteria as the measurable definition of done.
Real property names and types, explicit pipeline stages with probabilities and gates, workflow enrollment triggers, field-level sync mappings, seat-by-seat licensing. Nothing is left as "to be designed later."
Before you see your HubSpot blueprint, a second pass scores it against a weighted 100-point rubric — completeness, internal consistency, HubSpot tier-gate correctness, TCO arithmetic, and traceability back to what you actually said in discovery.
The generator pre-fetches the relevant HubSpot documentation for every section — property constraints, tier gates, current pricing — plus 26 expert delivery frameworks, from lead-scoring practice to migration sequencing.
A Starter HubSpot engagement gets a Starter-shaped architecture — not an Enterprise design with features you can't switch on. Seventeen industry verticals shape the HubSpot defaults.
Generated in batches (30,000–50,000 tokens of output), scored against the 100-point rubric, auto-remediated where needed, then saved to your encrypted workspace. Afterwards, AI Refine can rewrite the document, change the HubSpot tier, or apply a targeted revision — without a new engagement.
The architecture names every property, pipeline, workflow and gate. The runbook turns those decisions into twelve HubSpot phases — exact clicks, paste-ready code, and a Verify checklist on every step. Blueprint + Runbook is $2,100, including the team onboarding kit.
The structure is fixed. No section is dropped because your data was thin: the generator still writes it, states a recommended default, and marks what must be confirmed. Below is the full document, in order, and what each part actually decides.
Opens in plain English: who you are, why this project is happening now, and what success looks like in human terms — before any tables. Then a context sheet (company, industry, footprint, record counts, seats, budget, sponsor, admin) and an explicit map from your Ideal Customer Profile into the architecture: Green/Yellow/Red tiers → scoring, buying triggers → lifecycle, exclusions → suppression, churn patterns → negative scores and service design.
A “here’s what we’re going to build and why” narrative. Your explicit six-month success criteria — a quantitative target plus a qualitative outcome — become the measurable definition of done. Every later section has to trace back to it. Closes with what this means for your team, in language a CEO can approve.
Hub-by-hub why this tier, a seat-allocation table by role (Core, Sales, Service, Commerce, View-Only), and a mandatory first-year Total Cost of Ownership table in your currency. Every row shows the formula — unit × quantity × 12 — then a recurring subtotal, a one-time subtotal, and a first-year total that actually adds up. Mixed-tier seat escalation is called out when it applies. Closes with what the number means against the budget you stated in discovery.
What a data model is, in one paragraph, then the objects that will hold your information. HubSpot’s standard object set is mapped first (contacts, companies, deals, tickets, subscriptions, quotes, and the rest); a custom object is justified only when nothing native fits, and flagged as an Enterprise gate. Associations, property types, internal names, and the limits that silently break a CRM if you ignore them — association caps, dropdown internal values, currency stored as money not integer cents.
A dual Fit + Engagement score so sales is not guessing who to call. Fit is graded from your ICP Green-tier attributes; engagement from real behavioural signals; negatives from exclusions and churn patterns. Group caps stop one noisy activity from dominating the score. A sales View-Gate saved view shows only records that pass both thresholds — benchmarked, where you have it, to historical lead-to-customer conversion.
Every lifecycle stage with an owner, entry and exit criteria, and the custom status properties that sit alongside HubSpot’s defaults. Entry criteria reference the buying triggers from your ICP; Green-tier fit becomes the MQL threshold, not a generic “filled in a form.”
Every pipeline you actually run — new business, renewal, service, whatever discovery found — with ordered stages, probabilities, required-property gates and per-stage automation. Stage names follow your buying journey, not HubSpot’s demo defaults. If you quote, collect payment or track recurring revenue, a Commerce / quote-to-cash design specifies the products library, quotes, payments, subscriptions and deal splits, with the Commerce Hub and seat gates written in. If you don’t sell that way, commerce is explicitly out of scope — it is not invented.
The core workflow table: name, enrolment trigger, actions, owning hub. Nurture that addresses the emotional blocker from your ICP; routing through a single master routing workflow rather than a tangle. If reps run outbound or follow-up cadences, a Sales Sequences subsection (personal inbox, step list, delays, unenrol-on-reply) with the Professional-or-Enterprise seat gate. If they don’t, sequences are out of scope. Custom-coded action limits are stated so nobody designs a 21-second script that will never run.
Lawful basis for processing, how consent is captured (double opt-in where the market requires it), forms, subscription types aligned to how your buyers actually consume content, suppression lists from ICP exclusions, and sending-domain authentication (SPF, DKIM, DMARC). Cookie and tracking decisions match the jurisdiction you named in discovery — including TTDSG and Impressum where you operate in DACH.
Why each connection exists, in plain English. A system-of-record table — one row per object and key field — names the single master and the sync direction, so no field has two writers. A Phase-1 (launch) vs Phase-2 (deferred) split, with a reason on every deferred tool, so day one is not “every system anyone mentioned in discovery.”
What migration is, and why order matters: pipelines, properties and automation are built first, so imported records land in the right stages and do not fire emails mid-cutover. Legacy data is scoped with MoSCoW, cleansed outside HubSpot in five phases (export & audit → deduplicate → standardize → validate → import & verify), then imported companies → contacts → deals/tickets, sample-first. A field-mapping workbook, owner-ID map, picklist-value map, and three validation checkpoints. Workflows stay off during import.
Executive, operational and compliance dashboards tied to the success criteria in Section 1 — including ICP fit score and Green/Yellow/Red distribution where they belong. If you run a deal pipeline with quota, a Sales Forecasting design: amount mode (weighted vs total), forecast period, stage-to-forecast-category map, and per-rep goals, with the Professional/Enterprise seat gate. Starter portals get an explicit workaround instead of a forecast tool they cannot switch on.
Teams, roles and field-level controls — who can see what, including any PHI/PII or commercially sensitive properties. Then the adoption layer: per-team index-page saved views with the exact ordered column list (not “relevant fields”), record-page layouts, and named custom cards. Each team logs in to a purpose-built workspace. Team-specific layouts are an Enterprise gate; below that, one global default is designed and said out loud.
The same twelve-phase backbone the runbook executes click-by-click — Foundation & Security through UAT, go-live QA and Breeze rollout — mapped to weeks and owners for this engagement. A “what your team will feel” column on every phase. If a prior CRM or marketing-automation rollout failed, those causes from discovery become specific hypercare and change-management mitigations, not a generic “communicate more.”
Responsible, Accountable, Consulted and Informed for every workstream — data model, scoring, pipelines, automation, integrations, migration, reporting, go-live — including who owns RevOps after the partner leaves. Later mitigations in the risk register have to name someone who appears here.
A test matrix covering every compliance gate and technical constraint the architecture depends on. Plus a formal User Acceptance Testing plan: business users (reps, CSMs, marketers, the RevOps owner — not just the implementer) validate the end-to-end revenue flow in Given/When/Then scenarios before a Go/No-Go sponsor sign-off. The click-by-click UAT script lives in the runbook; this section designs what must pass.
A role-based training plan — what admins, managers, reps and marketing actually need to learn — and the enablement assets (SOPs, internal runbooks, deal-stage definitions, handoff criteria) without which a well-built portal still fails adoption.
Each material risk with likelihood, impact and a mitigation. Every mitigation is traceable to a named RACI owner. Typical entries: no RevOps owner after go-live, over-automation before adoption, migration without a pre-cleanse, missing stage-entry SOPs — the anti-patterns discovery is designed to surface.
30 / 60 / 90-day milestones, a Phase-2 roadmap for anything deferred (integrations, agents, extra pipelines), and how the Section 1 success criteria will be tracked. An explicit Out of Scope subsection states what this engagement deliberately does not include, and why — so the build cannot silently expand.
Brand assets, sending-domain branding, and branded templates (quotes, emails, meeting links). Where logos or hex values were not supplied in discovery, placeholders are flagged “TBD by client.” The section is never omitted.
Breeze explained for this client, then an AI readiness assessment (L0–L3). Foundation configuration: AI settings toggles, who may use the Assistant vs Agent Builder, and how Breeze context is seeded from this engagement’s ICP. An agents-in-scope table with hub/tier gates and prerequisites — never an agent the recommended tier cannot unlock — plus agents deliberately out of scope. A staged flagship-agent rollout with entry/exit criteria and a rollback rule. Breeze Intelligence (enrichment mode, fill vs overwrite). Credit governance that references Section 2’s TCO rather than inventing a second number. Per-agent success metrics. If you scoped no AI, the section still says why, what the readiness path is, and when to revisit — it is never dropped.
For every justified custom object: internal name, labels, properties, association labels, and the Enterprise gate. Empty if no custom object was warranted — the appendix still exists so the implementer knows the decision was considered.
Standard HubSpot properties cited with their real internal names, types and enumeration values — never renamed. New custom properties this engagement creates, in your lowercase snake_case convention, clearly separated from the standard set so nobody duplicates a standard field under a new name.
The Section 9 master table in one place: object, field, source system, sync direction, launch vs deferred. The implementer’s single lookup when a field looks like it has two owners.
Consent model, DOI, cookie banner, data residency, field-level restrictions, audit-trail properties, and any BAA or sensitive-data stance — the controls from Sections 8 and 12 collected so a DPO can read one appendix.
The HubSpot and RevOps terms this document uses, defined once, so a business reader and an implementer are talking about the same objects, stages and scores.
The Ironwood Field Services architecture, unedited — all 20 sections plus appendices. Tiers, data model, workflows, roadmap and the rest are in the contents list.
Loading the full document…
Same 20-section document a live engagement produces. Every heading in the contents list is in the file.
The AI-led HubSpot discovery interview and your Ideal Customer Profile are free — no credit card, no sales call. Pay only if you want the HubSpot blueprint, or Blueprint + Runbook together. The runbook is not sold on its own.
