Home / Blog / Discovery vs Starter

September 7, 2026 Commercial Plans AI / IT

Discovery $99 vs Starter $2,500: Which Zion Path Fits Your First AI/IT Win

Buyers ready to spend on AI or IT automation still stall on the first commercial choice: pay for a diagnostic, or pay for a scoped pilot. Those are not the same product with different stickers. Discovery maps the problem. Starter ships one scoped win. Mixing them up creates either a $2,500 project with unclear success criteria — or a $99 session that never becomes a decision.

This article is the decision write-up for CIO, CTO, and ops leads comparing Zion’s published entry SKUs. Service framing lives on AI consulting services and managed IT services. The commercial ladder sits on plans. Board-facing ROI language (without invented percentages) is in AI consulting ROI for the board.

We will not invent savings percentages, guaranteed timelines, or client logos. The test is fit: do you still need a map, or is the map already clear enough to fund Starter ZG06?

What Discovery is

Discovery is a paid diagnostic. It is not a free lead magnet, not a monthly retainer, and not an implementation.

For $99 you book a focused session and receive a written map: systems in play, owners, gaps, risks, quick wins, and a recommended next track. That map is what a later SOW should quote against — Consulting, Starter, Growth, or a Consulte engagement when the shape is still enterprise-messy.

What Discovery is designed to answer:

• Is the pain an automation candidate, an ops coverage gap, a FinOps envelope problem, or a governance/control problem?
• Who owns identity, systems of record, and the exception queue today?
• What belongs in a one-time project versus ongoing coverage?
• Are you ready for Starter, or do you still need Consulting / Consulte before money hits a pilot?

What Discovery is not: a promise that automation will “pay for itself,” a certified audit, or a substitute for your security review. It is the cheap, honest scoping step so you do not buy the wrong SKU. Book it on Discovery.

If Discovery shows you do not need Starter — only a process fix, a FinOps track, or a Consulte conversation — that is a successful diagnostic. Paying $99 to avoid an under-scoped $2,500 pilot is the point of the offer.

Treat Discovery like any other paid scoping step you would defend to finance: fixed fee, written output, clear non-goals. If a vendor offers “free discovery” that somehow becomes a multi-month commitment without a map you can reuse, you do not have a diagnostic — you have a sales process. Zion prices Discovery so the map is yours either way.

What Starter is

Starter is a scoped pilot / one automation live. The published entry is Starter — $2,500 (ZG06): implement one agreed improvement with clear inputs, owners, and a definition of done.

Starter fits when you already know:

• The problem statement in one paragraph (not a wish list of five “AI initiatives”).
• The systems involved and who can grant access.
• What “done” looks like in operational terms — a live workflow, a governed helper, a desk routing path, a spend alert — not a slide deck.
• Who accepts the deliverable on your side.

Starter is not a blank check for exploration. If the kickoff still debates which process to touch, you are not on Starter yet — you are still on Discovery or Consulte. Product lanes that often feed Starter include autonomous AI agents (governed helpers, not unsupervised agents), desk and ops automation under managed IT, and spend control paired with FinOps consulting.

Pay path and package detail live on plans. Prefer a human-scoped conversation before any package? Use Consulte / contact.

Signals you still need Discovery first

Buy Discovery first when clarity is the missing product — not labor.

• Multiple candidate problems. Leadership wants “AI somewhere,” but ops, finance, and IT name different pains. A $2,500 pilot will pick the loudest voice, not the best fit.
• Unclear ownership. Nobody can name who owns the system of record, the API keys, or the exception queue after hours.
• No written definition of done. Success is still “explore options” or “see what’s possible.” That is diagnostic language. Price it as Discovery.
• Stack is documented only in people’s heads. Vendors, SaaS tenants, and shadow tools are half-known. Starter access requests will stall.
• Budget exists but SKU does not. You can fund a pilot, but you cannot yet defend which pilot to finance in a board or ops review. Map first: see ROI framing for the board.
• AI and IT are tangled. Model spend, agent tooling, and classic ops coverage are mixed in one conversation. Separate the surfaces on a map before you buy implementation.

Rule of thumb: if a one-page SOW would still need three discovery workshops inside it, buy Discovery separately. Do not hide scoping inside Starter and call it a pilot.

Another signal: your internal Slack thread about “the AI project” still has three competing owners and no single paragraph of scope. That is not readiness for ZG06. Map the owners and the pain first.

Start with Discovery — $99 — systems, owners, gaps, and the next honest track.

Signals ready for Starter (ZG06)

Buy Starter ZG06 when the map is already clear enough that implementation is the bottleneck.

• One named process or automation with a single owner who can say yes/no on scope changes.
• Systems and access path known — tickets, CRM, identity, cloud billing, or the agent tool surface are identified, not guessed.
• KPI or operating signal agreed — queue time, exception rate, spend alerts, ticket routing quality — without inventing a savings percentage. The signal is observable after kickoff.
• Security and data boundaries drafted — what the helper may read, what it may write, and who reviews exceptions. For agent work, that gate belongs before production; see also autonomous AI agents.
• Acceptance owner named — someone on your side will sign that the deliverable matches the SOW.

If those boxes are checked, paying $99 again just to restate the same map is optional, not required. Many buyers still run Discovery as a cheap confirmation before ZG06 — that is fine. What is not fine is skipping the map when none of the bullets above are true.

Ready path: review plans for Starter ZG06, or Consulte if you need the SOW shaped with a human first.

When Consulte /contact/ beats both

Discovery and Starter are published packages. They are not the right first move for every buyer.

Consulte via /contact/ beats both when:

• Enterprise / multi-vendor maze — several MSPs, cloud accounts, or SI relationships already in flight; a $99 map will not replace a stakeholder workshop, and a single $2,500 pilot will not land without political design.
• Regulated or high-audit environments — you need a scoped consulting conversation on controls, evidence, and ownership before any Pay path.
• Multi-team RACI — IT, security, finance, and a product org all must agree before a helper touches production data.
• Unclear which Zion lane applies — AI consulting, managed IT, agents, FinOps — or a blend. That is strategy work; start on AI consulting or contact, not a self-serve pilot.
• You need a custom SOW — larger than Starter, different from Growth, or spanning multiple workstreams.

Consulte is not “the expensive version of Discovery.” It is the path when package SKUs are the wrong unit. Discovery can still be a cheap pre-read before a Consulte call; Starter can still be the first implementation after Consulte scopes it. The commercial honesty is: do not force a messy enterprise shape into ZG06 just because it is published.

If procurement requires a conversation before any card payment, Consulte is also the practical door — then route to Discovery or Starter once the commercial path is allowed. Do not invent a fourth SKU; use the published ladder on plans.

How /plans/ maps the ladder

Zion publishes a staged ladder so you do not buy Growth coverage or a pilot before the map says you need it. Live detail: plans.

• Discovery — $99 (one-time): paid diagnostic map — gaps, risks, quick wins, next track.
• Consulting — $499 (project): operating model, RACI, roadmap, pricing-fit recommendation when you need more than a map and less than a pilot.
• Starter — $2,500 (ZG06) (project): implement one agreed automation or ops improvement.
• Growth (ongoing): managed / co-managed coverage under agreed scope — only after the operating model is clear.

How to read the ladder for this article’s decision:

1. Unclear problem / unclear owners → Discovery $99.
2. Map clear, need RACI / roadmap / multi-stakeholder design → Consulting $499 or Consulte.
3. One scoped win ready to build → Starter ZG06 on plans.
4. Enterprise / regulated / multi-vendor → Consulte first; packages later.

Related lanes after the ladder choice: AI consulting services, managed IT services, autonomous AI agents, FinOps consulting.

Proof chips only: Discovery $99 · Consulting $499 · Starter $2,500 (ZG06) · Response within 24h. Prefer email first? kleber@ziontechgroup.com

FAQs

See also: Discovery $99 · Plans · Consulte / contact · AI consulting services · Managed IT services · Autonomous AI agents · FinOps consulting · AI consulting ROI for the board · Pay vs Consulte