FAQ

We extract structural metadata — the configuration layer of your system. This includes module and field definitions, workflow rules, blueprint stage definitions and transitions, custom and scheduled functions, Zoho Flow connections, and the relationship structure between modules.

We do not extract transactional business data — individual records, contact details, deal values, or anything of that nature. The goal is a complete map of how your system is built, not a copy of what's in it.

No. Every engagement is strictly read-only. We query your Zoho environment through its APIs — nothing is created, modified, or deleted. Your system is untouched throughout the entire process.

Currently, Zoho CRM, Zoho Recruit, and Zoho Flow (including a dedicated Connections audit). The Starter package covers CRM and Flow; Standard and Advanced add Recruit.

Zoho Desk and Zoho Analytics extraction are on the roadmap. If those are important to you, mention it when you get in touch — we'll let you know where things stand.

There are two options — you choose whichever you're more comfortable with.

Option 1 — Zoho Assist (remote session). You establish a remote connection to your browser using Zoho Assist, and we log into your Zoho environment through that session. You never share a password, and your MFA setup stays entirely on your side. We do what we need to do, and the session ends.

Option 2 — Credentials with Settings access. You provide login credentials for a Zoho user that has access to Settings. No special configuration is needed beyond that — standard admin-level visibility is sufficient for extraction.

Either way, nothing in your system is modified. We treat all client system information as confidential — full details are in our Privacy Policy and Terms of Service.

All packages are delivered within 5 business days of access confirmation. For Advanced package refresh extractions, the same 5-day window applies per extraction, starting from the date you initiate it.

All packages include interactive HTML visualizations of your system's structure, the underlying structured metadata files, and a set of AI prompt templates designed to help you get immediate value from the output using whatever AI tool you prefer.

Standard and Advanced packages also include a written findings summary — a plain-language synthesis of what we found, flagging anomalies, gaps, and items worth attention. Advanced adds two client-initiated extraction refreshes with delta analysis, valid for 18 months.

See the Sample Output page for a closer look at what's included.

Yes. The visualizations are designed to be readable without a technical background — they show you what's in your system in a structured, navigable format. The AI prompt templates are written for non-developers: you paste your metadata into an AI tool, run the provided prompts, and get plain-language answers to questions like "where are my workflows most likely to conflict?" or "what fields are never populated?"

Standard and Advanced clients also receive a written findings summary that interprets the output for you directly.

Both. We deliver a structured, AI-ready package: the metadata, the visualizations, and a set of prompt templates covering common diagnostic and documentation use cases. From there, you can take the output into any AI tool — Claude, ChatGPT, Gemini — and query it yourself.

For Standard and Advanced clients, we also perform our own analysis pass and deliver a written findings summary — so you're not starting from zero even if you never open an AI tool yourself.

Any AI with a large context window and the ability to accept uploaded files or pasted text. Claude, ChatGPT, and Gemini all work well. The prompt templates are written to be AI-agnostic — the instructions in each template explain how to load the metadata and what to ask.

Generally high, but with a known edge case worth understanding: our parser identifies field references within Deluge scripts by analyzing the code directly. Most field references are captured reliably. However, when a field's API name is constructed dynamically — for example, by concatenating string values at runtime — the parser cannot resolve the resulting name, and that reference will be missed.

This is a disclosure, not a disclaimer. If you identify a genuine gap caused by this limitation and it's something we can address, we'll rerun the relevant analysis at no charge and deliver an updated output. Just flag it and we'll take it from there.

Sometimes the most useful thing a small-system audit turns up is what's missing — fields that were never mapped to a workflow, stages that have no exit condition, automations that were planned but never built. A clean system is still worth documenting, especially before it gets complicated.

That said, if you're genuinely unsure whether the scope fits, get in touch and describe your setup. We'll give you a straight answer.

Early is actually the best time. A baseline extraction in the first 90 days gives you a clean reference point before complexity accumulates — and means that 12 months from now, when something goes wrong, you're diagnosing against a known-good starting state rather than guessing what changed.

It also means your system is documented from the start. That matters the first time a staff member leaves or a consultant hands off.

Yes. An inherited system with layers of undocumented automation, renamed fields, overlapping workflows, and logic that nobody can fully explain is exactly what this service was built for. We've seen it all — and we don't need anyone to explain it to us first. The extraction tells us what's there. The analysis tells us what it means.

No. You can engage us directly by filling out the inquiry form. We'll review your situation, respond within one business day, and scope an engagement from there — no Upwork account required.

Absolutely. The Starter package is CRM-only and includes the full extraction, visualizations, and AI prompt templates. If Recruit isn't in your environment, Standard and Advanced still deliver complete value — the Recruit component simply won't apply. You're paying for the scope of what you have, not for modules you don't use.