# The Automation Audit: Four Things We Check Before We Scope Anything *Nikhai Jaysen ยท October 26, 2026* > Before we propose a voice AI agent, chatbot, or workflow automation, we run an automation audit. Here is exactly what we check, and what it changes. Source: https://www.spaceblanket.ai/blog/automation-audit-four-things-we-check Before we recommend a voice AI agent, a chatbot, or any automation, we run an audit on four things: where the time goes, what the data looks like, what's already half-automated, and who will own it after launch. Here's what that actually looks like, and what it changes. Most conversations about automation start with a tool. Should it be a voice AI agent or a chatbot? WhatsApp or email? We don't start there. Before we propose anything, we run an **automation audit**: a structured look at how the business actually runs today, not how the founder remembers it running. It usually changes the scope of the project. Sometimes it cancels the project the client walked in wanting. Most AI automation agencies open with a product pitch: here's the agent, here's the demo, here's the price. We open with four questions, and we answer them before we write anything down for the client to sign. ## What an Automation Audit Actually Checks ### Where the time actually goes We pull call volume, response-time data, and ticket or chat logs before we form an opinion. A SaaS founder tells us demo scheduling is the bottleneck. The call logs tell us the booking flow is fine and the real problem is no-shows after the demo is booked. Those are two different automation projects, and only one of them is what the founder asked for. ### What the data actually looks like We check the CRM or spreadsheet the agent would have to run on. Duplicate contact records, missing phone numbers, fields that are supposed to be populated but aren't. In one audit for a B2B SaaS client, inbound demo requests sat across three different spreadsheets before anything reached the CRM. No [automation strategy](/services/ai-consultation) fixes that on its own. The first project became a [CRM integration](/blog/voice-ai-crm-integration-before-you-build), not a voice AI agent, because a voice AI agent calling into duplicated data would have called the same lead three times in one afternoon. ### What's already half-automated Most businesses we audit already have two or three tools quietly doing part of the job: an old chatbot nobody maintains, a Zapier chain that breaks every few weeks, a WhatsApp number someone's assistant checks between other tasks. We map these before proposing anything new, because replacing a half-working tool is a different, usually smaller, project than building from nothing. We'd rather recommend [AI chatbot development](/services/ai-chatbots) that replaces a broken flow than stack a second system on top of it. ### Who owns it after launch We confirm who reads the call logs, updates the knowledge base, and handles escalations once the agent is live. This is the check clients skip when they design their own automation plan, and it's the one that decides whether the system still works in month three. We've written about what happens when [no one owns the agent after launch](/blog/ai-agent-ownership-gap), and it's rarely the model's fault. ## What an Automation Audit Decides The honest version of this: the audit sometimes tells a client to fix something before they automate anything. Gartner has [projected that organizations will abandon 60% of AI projects through 2026 because the underlying data wasn't ready for AI](https://www.gartner.com/en/newsroom/press-releases/2025-02-26-lack-of-ai-ready-data-puts-ai-projects-at-risk), and that matches what we see before a build even starts. The benefit of developing an automation strategy before you buy anything is that you find out which process is broken while it's still cheap to fix, not after an agent has been live for a month calling into bad records. This is also where an audit saves a client money they didn't expect to save: catching that the real bottleneck is a staffing or ownership gap, not a missing tool, before anyone pays for a build that would have sat unused. We'd rather have that conversation in week one than in month three. If this sounds like your [discovery call](/blog/what-our-discovery-call-actually-looks-like) is going to raise more questions than it answers, that's the audit doing its job. [Book a free automation audit](/services/ai-consultation). We'll map the highest-ROI automation for your business in 30 minutes, and tell you honestly if something needs fixing first. ## How we audit a business before scoping an automation project: the steps 1. **Where the time actually goes.** We pull call volume, response-time data, and ticket or chat logs to find out where staff hours are really going, not where the client assumes they're going. 2. **What the data actually looks like.** We check the CRM or spreadsheet the agent would run on for duplicate records, missing phone numbers or emails, and fields that are supposed to be populated but aren't. 3. **What's already half-automated.** We look for existing tools and workflows already in place that quietly break, get ignored, or duplicate work, so we don't build a second version of a problem that already exists. 4. **Who owns it after launch.** We confirm who on the client's team will read the logs, update the knowledge base, and handle escalations once the agent is live, because an agent with no owner degrades within weeks. ## Frequently Asked Questions ### What is an automation audit? A structured review of how a business actually runs today, before any tool is chosen. We look at call and chat volume, data quality in the CRM, what's already half-automated, and who will own the system after launch. It decides what should be built, not just how. ### How long does an automation audit take? Most audits run alongside the discovery call and take three to five business days, mainly waiting on CRM or call-log access from the client's side. The audit has to finish before we write a proposal, because the findings often change which channel or process gets automated first. ### What happens if the audit finds the business isn't ready to automate? We say so before any build starts. If lead data is missing phone numbers, or three tools already quietly do the same job, fixing that becomes the first project instead of the automation the client originally asked for. Automating broken data just moves the mess faster. ### Does every client get the same audit? The four checks stay the same: time, data, existing tools, and ownership. What changes is depth. A small team with one inbox needs a lighter pass than a SaaS company with a CRM, a support queue, and three existing automations already running. ## Related Service ### [AI Strategy Consultation](/services/ai-consultation) Not sure where to start with AI? Get a structured audit of your business and a clear roadmap for where AI will deliver the most ROI. ## Related Reading ### [What Our Discovery Call Actually Looks Like (And Why We Rarely Pitch on It)](/blog/what-our-discovery-call-actually-looks-like) Most businesses contact an AI automation agency expecting a pitch deck and a quote. Our discovery call doesn't work that way, and understanding the difference tells you a lot about whether an automation build will succeed. ### [From Brief to Live: How We Build a WhatsApp AI Agent in 2 Weeks](/blog/whatsapp-ai-agent-build-process-two-weeks) Most clients have no idea what a two-week WhatsApp AI build actually involves. Here's the exact process, discovery to go-live, and the three things that reliably cause delays. ### [What Happens After You Launch an AI Chatbot: The Part No One Talks About](/blog/ai-chatbot-post-launch-maintenance-guide) Most agencies talk about the build and go quiet after launch. Here's what post-launch AI chatbot operations actually look like: the monitoring, the tuning, the maintenance, and the failure mode that catches most businesses off guard. ## Related - [All articles](https://www.spaceblanket.ai/blog) - [AI Strategy Consultation](https://www.spaceblanket.ai/services/ai-consultation) - [All AI automation services](https://www.spaceblanket.ai/services) - [Contact](https://www.spaceblanket.ai/contact)