How to Write an AI Acceptable Use Policy Employees Will Actually Follow

/by

An IT leader at a mid-market manufacturer told us he was racing to get his company’s first AI policy to a working version within the week. He had a borrowed template from his managed service provider open in one window and a blank page in the other. Then his infrastructure partner asked a simple question: do you have any data privacy requirements around AI, like no training on your data, no retention, no PII? His answer was honest. Not yet.

That moment plays out in mid-market companies everywhere. Your people adopted AI first, and the rules are trying to catch up. An AI acceptable use policy is how you close that gap, but only if you write one that people can actually follow.

Why most AI policies fail before they launch

Two failure modes kill most first attempts. The first is silence. At a town hall we heard about, an employee asked leadership directly what the company policy on AI was, and the answer was that one did not exist yet. In one of our operations workshops, a leader admitted the same thing plainly: no governance policy, nothing. Employees in that vacuum make up their own rules, and the most common question they carry is whether it is even okay to use AI at all.

The second failure mode is overcorrection. One leader we work with borrowed a policy from a friend at a cybersecurity firm, and it was so restrictive it effectively banned AI outright. A policy that reads like a ban does not stop AI use. It just stops honest AI use, and pushes everything underground where you cannot see it.

A policy that reads like a ban does not stop AI use. It just stops honest AI use.

The numbers say most companies are still stuck between those two failures. Only about 15 percent of organizations have updated their acceptable use policies to cover AI specifically, according to research compiled by JumpCloud, even as unsanctioned AI use shows up in nearly every workforce.

Decide what your policy is for

At a recent executive roundtable we hosted, the sharpest question of the session was deceptively simple: is this policy a structure for the organization to operate with AI, or a defensive legal and HR document? The answer shapes everything. A defensive policy protects the company from its people. An operating policy protects the company through its people, by giving them a safe, clear way to work.

Write the operating version. It should answer, in plain language, the questions your employees are already asking: which tools are approved, what data can go where, and what to do when they want something new.

The six sections your AI acceptable use policy needs

Keep it to one page if you can. Here is the structure we help clients build:

  1. Approved tools: Name the sanctioned tools and accounts. If AI use must run through IT approval on enterprise licenses rather than personal accounts, say exactly that.
  2. Data rules: Spell out what can never enter an AI tool, such as customer records, financials, and regulated data, and where protected work can happen safely.
  3. Verification duty: If you get output from AI, you verify it before it ships. Define what verification means for high-stakes work like quotes, contracts, and customer communication.
  4. Accountability: Every AI workflow and agent rolls up to a named human owner. When an agent produces a bad quote, the accountability trail should already exist.
  5. The request path: Give people a fast, judgment-free way to propose new tools. This turns shadow AI into a pipeline instead of a threat.
  6. Ownership and review: Name who signs the policy, who manages it, and when it gets revisited. A policy nobody owns goes stale in a quarter.

Roll it out like you mean it

Publishing a PDF is not a rollout. Two moves make the difference. First, put adoption metrics on managers, not just individuals. In one client rollout, training participation jumped significantly the moment completion became a management team metric rather than an individual one. Second, pair the policy with real training so people understand the why behind each rule; the gap between publishing rules and building skills is covered well in our piece on the hidden cost of not training your team on AI.

Your policy also does not live alone. It is the front door of a broader governance posture, and our guide on what every executive needs to know about AI governance covers the structures behind it. And once rules exist, adoption becomes the next battle, so keep getting your team to actually use approved AI on your radar.

From policy to activation

The gap between AI aspiration and AI that works is rarely about the technology. It is about whether your people know the safe path and trust it enough to take it. Across our client work we have seen what happens when they do: leaders standardizing on governed platforms so company information stops flowing through personal accounts, teams coordinating so three people are not quietly building the same agent, and manager-led rollouts that measurably lifted training completion.

That is Company Intelligence working alongside Security Intelligence: rules that amplify your people instead of restraining them. Diagnose where AI is already carrying weight in your business, write the one-page policy that makes it safe, and stand behind it in production.

Put AI to work for your people, with rules they can actually follow.

Frequently Asked Questions

What is an AI acceptable use policy?

A short document that tells employees which AI tools are approved, what data can and cannot enter them, how AI output must be verified, and how to request new tools.

How long should an AI acceptable use policy be?

One page is the goal. A forty-page policy nobody reads is worse than a one-page standard everyone follows.

Who should own the AI acceptable use policy?

Assign a named owner, typically IT or security leadership with HR input, plus a review cadence. Every AI workflow in the policy should also roll up to a named human owner.

Should the policy ban public AI tools?

Usually not. Over-restrictive policies push AI use underground. Provide approved alternatives that are genuinely better, then restrict only the highest-risk behavior.

Shadow AI: Your Team Is Already Using AI. You Just Don’t Know Where

/by

A technology leader at a mid-market services company put it bluntly in a working session with us: everybody was doing their own thing with AI. People were pasting company spreadsheets into free personal ChatGPT accounts, and that data was leaving the building for good. As someone in the room quipped, the public models were getting free training content, courtesy of the company.

That is shadow AI. Your people are not waiting for permission. They have deadlines, and a free AI tool is one browser tab away. The question is not whether unsanctioned AI use is happening in your company. It is how much, with what data, and what you will do about it.

What shadow AI actually is

Shadow AI is any AI tool your employees use for work without approval, oversight, or protection from IT. It is the AI-era version of shadow IT, with one difference: these tools actively process your data at the point of use. A personal ChatGPT account, a free transcription app, a browser extension summarizing contracts. Each one is a door your data walks through, and most lead outside your control.

The scale is bigger than most executives expect. Verizon’s 2026 Data Breach Investigations Report found shadow AI detections quadrupled in a single year, with 45 percent of employees now using AI regularly on corporate devices and roughly two thirds accessing AI through personal accounts that company controls never see.

Why your best people are the biggest users

Here is the uncomfortable part: shadow AI is not a discipline problem. It is a demand signal. The employees using unsanctioned tools are usually your most motivated people, trying to move faster than your official systems allow.

 

We see this pattern constantly in mid-market companies. At one executive AI session we ran, an employee had asked leadership at a town hall what the company policy on AI was. The honest answer: we have not come up with one yet. In another workshop, the most common fear in the room was simpler than job loss. Is it okay to use AI? If I put something in here, will it end up on the internet? Will my competitor get it?

When leaders leave those questions unanswered, employees answer them alone. Experimentation goes underground, and one operations leader we work with described the result perfectly: a Wild West approach to applying AI. The energy is real, but nobody is protecting the data, checking the output, or capturing what works.

The risks are specific, not hypothetical

The costs now show up in breach data, not just policy debates. IBM’s research links shadow AI to roughly one in five data breaches, adding about $670,000 to the average breach cost. Three risks matter most for mid-market leaders:

  1. Data leakage: Customer records, pricing, financials, and source code pasted into free tools may be retained, and you cannot pull them back.
  2. Compliance exposure: Regulated data moving through consumer AI tools creates violations your team never intended and your auditors will eventually find.
  3. Lost leverage: When fifty people solve the same problem fifty different ways in fifty personal accounts, nothing compounds. One construction industry leader told us they did not want three different people building the same agent. Coordination is where the value is.

Bring it into the light without killing the momentum

Banning AI does not work. It just pushes usage deeper underground and punishes your most ambitious people. The companies handling this well treat shadow AI as a map of unmet needs, then replace risky tools with better sanctioned ones. Here is the path we walk with clients:

 

  1. Start with an amnesty audit: Ask every team what AI tools they already use and what for. Treat the answers as intelligence, not evidence. This single step usually surfaces your best AI use cases for free.
  2. Give people a safe place to work: One manufacturer we partner with had real privacy concerns about public tools, so the answer was AI built inside their existing infrastructure, where sensitive data never leaves their environment. When the sanctioned option is also the best option, shadow use fades on its own. If you are weighing how to do this with your own data, our guide to using your own data with LLMs securely covers the governance fundamentals.
  3. Write rules people can actually follow: A one-page acceptable use standard beats a forty-page policy nobody reads. Pair it with the broader governance thinking in what every executive needs to know about AI governance.
  4. Make managers accountable for adoption: In one client rollout, participation in AI training jumped significantly the moment completion metrics moved from individual employees to their managers. Governance sticks when leaders own it.
  5. Keep listening: Shadow AI never fully disappears. New tools launch weekly. A quarterly check on what people are reaching for tells you where to invest next, and pairs well with getting your team to actually use the AI you approve.

The gap between AI aspiration and AI that works

Shadow AI is what happens when your people’s AI ambition outruns your company’s AI foundation. The fix is not less AI. It is AI done on purpose: diagnosed, built for your environment, and backed by someone who stands behind it in production.

That is the work of Security Intelligence, and it pays off beyond risk avoidance. Across our client work, governed environments have unlocked what unsanctioned tools never could: standardized platforms that let whole departments share what works, secure builds that finally let privacy-conscious leaders say yes, and manager-led rollouts that measurably lifted training completion. Companies that govern AI well do not slow down. They get to speed up safely.

Your team is already using AI. Put it to work for them, on your terms.

Your Meetings Cost $29K a Year, and Most of It Evaporates

/by

 That’s $29k per employee. Per year.

In a recent Augusto leadership meeting, the point was made that our client roundtables generate a massive amount of intelligence worth sharing across the company. Nobody in the room treated that as a nice sentiment. We treated it as an operating requirement, because we had already built the system that makes it true: every conversation we have is captured, mined, and turned into knowledge the whole company can use.

Most companies run the opposite way. Their best thinking happens out loud, in meetings, and then most of it disappears. The decisions, the commitments, the client saying something that belongs in a case study, the early warning sign in a customer’s tone. It all happened, and a week later almost none of it exists anywhere. That gap between what gets said and what gets kept is the whole argument for AI meeting intelligence.

The Evaporation Problem

The research on this is brutal. Without follow-up notes, 70 percent of decisions made in meetings are forgotten within 24 hours, and 47 percent of action items discussed are never captured at all. This is not a discipline problem. People speak faster than anyone can reliably record, and the person taking notes is also trying to participate.

Now put a price on it. The average employee spends 392 hours a year in meetings, at a cost of roughly $29,000 per employee per year, and executives average around 23 hours a week. Estimates of what unproductive meetings cost US companies run from $37 billion to $399 billion annually. Your company is already paying full price for these conversations. The only question is whether you keep what you paid for.

What Meeting Intelligence Actually Means

Most leaders hear this topic and picture an AI note-taker that emails a summary nobody reads. That is the shallow version. AI meeting intelligence means the conversation becomes structured, searchable business data. We know because we built this for ourselves before offering it to clients.

The capture layer comes first, and ours is standardized: every call at Augusto is recorded with Fathom, our company-wide AI note taker. We like it because it does the unglamorous part flawlessly. It joins every meeting automatically, produces an accurate transcript with speaker labels and timestamps, and delivers a summary with action items before anyone has left the call. Just as important for what comes next, every moment carries a deep link, so any quote or decision can be traced back to the exact second it was spoken in the recording.

From there, every call flows into a system we call our Second Brain, and this is where Fathom’s clean transcripts become compounding value. AI reads each Fathom transcript and extracts the action items, with owners. The system identifies ROI signals when clients describe measurable outcomes in their own words, detects relationship signals such as appreciation, concern, or renewal risk, and maps attendees, their connections, and trending topics across every engagement.
Before a client meeting, anyone on our team can pull a prep brief built from the entire history of the relationship. When we write a case study, the client quotes are real and verbatim, because the system kept them. This is the same agent workflow pattern we apply to any document-heavy process, pointed at conversations instead. It even feeds our marketing: the proof points in our content pipeline, described in how Augusto automates content creation, come out of these meetings.

What Changes When Conversations Stop Evaporating

The first shift is follow-through. Teams that adopt AI-generated meeting summaries see action-item completion rates rise from the 50 to 60 percent range to 85 to 95 percent, largely because a searchable record kills the follow-up meetings scheduled to clarify what the last meeting said.

The second shift is institutional memory. When a veteran team member leaves, their client context does not leave with them, because two years of their conversations are searchable. When someone new joins an account, they read the relationship’s history instead of asking five people to repeat it. And leadership stops flying blind between quarterly reviews, because concern and satisfaction signals surface from calls as they happen, not months later in a churned account. None of this requires anyone to change how they run a meeting. The meetings you already hold become an asset instead of an expense, which is the cleanest ROI case we know. Our guide on measuring AI ROI before you invest shows how to baseline it.

How to Start

The pattern is simple. Record by default, with clear consent and sensible exceptions. Pipe transcripts into a system that extracts action items, signals, and people, rather than a folder of files nobody opens. Connect the output to where work actually happens, your CRM, your project tool, your team channels. Keep a human reviewing anything that drives a decision. Start with client-facing calls, where the evaporating value is highest, and expand from there.

Your meetings already contain the intelligence. The only thing missing is the system that keeps it. If you want to see what a Second Brain would look like on your own calls, start a conversation with our team.

Monthly LLM News September 2026

/by

August gave us a first: the most capable model of the year got pulled off the field by its own team, one week after its greatest win. Add the biggest leadership shakeup in Google’s AI history, an 80 percent price collapse, and one billion people now using AI weekly, and the month reads like an industry growing up in public. Here is your LLM news roundup, and what each story means for how you run your business.

OpenAI’s Astra solved 80-year-old math problems, then got benched

In the first week of August, OpenAI announced that an internal version of Astra, its next major model, solved ten previously open problems in mathematics and theoretical computer science, publishing formally verified proofs for roughly $2,000 in compute. These were not benchmark tricks; some had resisted mathematicians for decades.

Five days later, the same model earned a very different distinction. Astra became the first system OpenAI classified as potentially Critical for cybersecurity under its Preparedness Framework, meaning it may be capable of finding zero-day exploits or executing novel attacks on hardened systems on its own. OpenAI paused every internal Astra activity that did not meet a strengthened set of security controls, and Axios later reported that its largest planned frontier training run remains on hold.

The smartest model of the year is currently benched by its own maker. That is the whole story of AI in 2026 in one sentence.

What it means for your business: capability is still climbing steeply, and so is the caution around it. Do not build Q4 plans around a promised model release, because safety pauses are now part of the calendar. And take the dual-use warning seriously: if a frontier model can attack hardened systems, your security posture is the differentiator. Our guide on AI governance for executives is the place to start.

Frontier AI prices collapsed 80 percent, but read the fine print

On July 30, OpenAI cut the price of GPT-5.6 Luna by 80 percent, from $1 and $6 per million tokens down to $0.20 and $1.20, with Terra down 20 percent. A week later Luna became the free ChatGPT default. The pressure is real: Chinese open-weight models captured roughly 46 percent of US enterprise token usage on OpenRouter, and pricing is the counterpunch.

One caution before you celebrate: some of the new low prices are promotional. Google’s newly released Gemini 3.7 Flash, for example, carries pricing that is locked only through 2026, with the list price set to double on January 1, 2027. Teaser rates are now part of the AI market.

What it means for your business: re-run the math on every high-volume AI workload, because tasks too expensive in June may be nearly free now. Then diary a pricing review for every model you depend on, so an expiring promo does not quietly double a line item. AI budgeting is where enthusiasm meets the CFO, and this month moved the numbers in your favor if you act.

 

Google’s AI brain trust walked out the door in one day

On August 5, Google announced that Demis Hassabis is stepping back from running Google DeepMind day to day, becoming its Chair and Alphabet’s Chief Scientist, while Jeff Dean, employee number 30 and a 27-year veteran, left to co-found the startup Discovery Loop with fellow senior researchers. Alphabet fell about 4 percent, and reporting has since tied the shakeup to stalled models, missed deadlines, and staff burnout, with Gemini’s next flagships months behind.

What it means for your business: vendor stability is now a diligence item. July showed what happens when a model disappears overnight; August showed even the biggest labs can wobble organizationally. The answer we build with clients is model-agnostic plumbing, so critical workflows reroute between providers without a rewrite.

One billion people now use AI weekly, and they work for you

Alongside the price cuts, OpenAI disclosed that its models now reach more than one billion active users and over two million businesses, with usage deepening over time: six months in, people send about 50 percent more messages and use AI for twice as many kinds of work. The shift OpenAI describes, from asking to doing, matches what we see in mid-market companies weekly: employees are not waiting for an AI strategy, and leaders in our workshops still call their environment a Wild West.

What it means for your business: your customers now experience AI-grade responsiveness everywhere else, so their expectation bar for you just moved. And your employees are already among that billion, which makes governed, sanctioned AI a retention and security issue at the same time.

Meta doubled down on open weights, and that helps you

In mid-August Meta shipped its Muse Glimmer model alongside a 6,500-word Zuckerberg manifesto arguing for more open-weight releases plus a $1 billion fund for data center communities. The practical effect: credible open-weight options keep improving, strengthening your hand at every renewal and keeping self-hosted AI viable for cost-heavy or privacy-sensitive work. The client teams that standardized on governed platforms with an open-model fallback are sleeping through months like this one.

Four moves for September, with owners

  1. Re-run your model pricing (CFO and CTO, 30 days): Test Luna-class models on your highest-volume tasks and diary every promotional rate’s expiry date.
  2. Kill your single point of failure (CIO, 60 days): Add fallback routing so no one vendor pause, shakeup, or shutdown can stall a production workflow.
  3. Raise your customer experience bar (CEO, this quarter): A billion weekly AI users are recalibrating what responsive feels like. Pick one customer-facing workflow and close the gap.
  4. Harden before the agents arrive (Security leader, now): Astra’s Critical rating is a preview of the threat curve. Verify your AI use is governed and your defenses assume AI-assisted attackers.

The pattern across all five stories is the same one we see in the field: the technology is sprinting, and the winners are the companies whose foundations let them change direction without falling over. Treat AI as infrastructure, deploy it against high-value workflows, and scale what delivers measurable returns.

Put AI to work for your people, and keep your footing while the industry finds its own.

Is Your ERP Ready for AI?

/by

Every mid-market operations leader we meet eventually asks this question, usually with some dread behind it. The ERP runs the business. It has for fifteen or twenty years, and it works. Yet every AI conversation now seems to end at the same wall: our system was never built for this. In one recent working session, the concern was stated plainly. If AI hits the production database in real time, all day long, what happens to the performance of the system the whole operation depends on?

It is the right instinct and the wrong conclusion. The instinct is right because the fear is earned. The conclusion is wrong because ERP AI integration does not require your ERP to be ready at all. It requires your data to be reachable, and those are very different problems. The second one is far easier to solve.

Why the Hesitation Is Earned

Caution around legacy systems is not resistance to change; it is pattern recognition. Gartner estimates that by the end of 2026, six out of ten AI initiatives will be scrapped because the underlying data was not prepared for AI. Decades-old ERPs scatter customer, product, and supplier records across modules, and only 43 percent of collected manufacturing data is used effectively, which leaves most of its value locked away. Meanwhile the pressure keeps building: Rootstock’s 2026 manufacturing survey puts AI adoption at 94 percent, so your competitors are not waiting for their systems to feel ready either.

Faced with that, leaders see a forced choice: bolt AI onto a system that cannot support it, or replace the ERP entirely. Both options are expensive, and the second is the kind of project that puts careers at risk. There is a third path, and it asks nothing of your ERP.

The Coexistence Pattern

Here is what ERP AI integration looks like in our client work, using a real example. An industrial components manufacturer takes quote requests by email all day: part numbers, pricing, stock, lead times. Behind every answer sits an ERP with more than 1,100 tables, and a fourteen-person customer service team worked each request up by hand against it. The automation case was obvious. The hesitation was just as obvious, because a system that complex, sitting underneath quoting, inventory, and scheduling, is the last place anyone wants experimental AI processes running.

So we did not touch it. We connected AI to just the data the quoting workflow needs, let it read each incoming request, pull the price, stock, and lead time, and tee up a draft quote beside the ERP rather than inside it. The system of record stays exactly as it is, a person approves every quote before it reaches a customer, and today roughly half of quote requests are handled with no manual work-up at all. The pattern generalizes: leave the ERP alone, automate next to it, and route results back through the same interfaces humans already use. It is the same discipline we describe in how to automate manual processes, applied to the most sensitive system you own.

This is also why coexistence usually beats waiting for an ERP replacement. Modernization can happen in parallel, one module at a time, while the automations deliver value now.

People Stay in the Loop

A worry usually surfaces at this point: can we trust AI output flowing near our system of record? The honest answer is that you should not, blindly, and the companies getting real value do not. McKinsey’s research shows 71 percent of organizations now use generative AI in at least one function, and the ones capturing the most value clearly define when human validation is required. In the quoting workflow above, every draft passes through human eyes before it goes out, review takes minutes, and the complex, configured requests still route to the people who know the product best. The team felt the difference. As the leader who owns that workflow told us:

“My customer service team, which is about 14 people, happens to love the idea now, so that’s a huge win for us.”

That is the shape of every durable automation we build: AI does the repetitive reading and lookup work, and agent workflows escalate anything uncertain to a human who knows the business.

Three Questions Before You Start

You do not need an AI strategy document to begin. You need answers to three questions about one workflow. First, where does this workflow’s data live, and can it be reached through a standard connection? Almost always, yes, even on the oldest systems. Second, who will review the output, and what should trigger their attention? Third, what is the measurable change, in hours or dollars, that proves the case? Mid-market companies already allocate 3 to 5 percent of annual revenue to integration and ERP systems; starting small makes sure the next dollar of that budget returns something visible. Our framework for measuring AI ROI before you invest walks through the scorecard.

Your ERP is not the obstacle. It is the system of record, and it can keep doing that job untouched while AI takes over the manual work happening around it. If you want to pressure-test one workflow against this pattern, start a conversation with our team.

Frequently Asked Questions

Does ERP AI integration require replacing our ERP?

No. The most reliable pattern for mid-market companies is coexistence: connect to the data one workflow needs, automate in a secure environment beside the ERP, and leave the system of record untouched.

Will AI slow down or destabilize our production ERP?

Not if it never runs against production in real time. Working from a synced copy of the data removes the performance risk, which is usually the biggest and most legitimate objection.

Our system is decades old. Is it too old for this?

Almost never. Even ERPs that predate the cloud expose standard connections that let data be reached safely. Age affects how you connect to the data, not whether AI can work beside the system.

Where should we start?

Pick one workflow where people manually move information in or out of the ERP, such as quote preparation, order entry, or document matching. Those workflows have clear volume, clear hours, and a measurable before and after.

The 3 Questions That Actually Decide Your Automation Platform

/by

When automation comes up in a leadership meeting, the first question is rarely about features. On a recent call, an executive at a manufacturer asked us directly: are you working in environments we already have, or are you bringing a whole new platform we will have to staff and pay for? That is the right question, and it is the lens for the n8n vs Zapier vs Make decision. All three can move data between your systems. They differ on what you pay at volume, where your data lives, and who has to maintain the thing.

Here is how we walk mid-market teams through the choice, including why our own AI work runs on one of them.

Three Platforms, Three Pricing Models

The sticker prices look similar. The models behind them do not. Zapier bills per task, Make bills per operation, and n8n bills per workflow execution, meaning one full run counts once no matter how many steps it contains. As of mid-2026, Zapier’s Professional plan starts at $19.99 a month for 750 tasks, Make starts at $9 for 10,000 credits, and n8n Cloud starts around $20 for 2,500 executions, with the self-hosted edition free.

At low volume, any of them is affordable. The divergence shows up as workflows grow. At 10,000 tasks a month, Make runs roughly 70 percent cheaper than Zapier, and self-hosted n8n roughly 95 percent cheaper. A ten-step workflow that runs constantly is ten billable tasks per run in Zapier and one execution in n8n. If AI is in your plans, volume is in your future, because agent-style automations run all day. That math is a cousin of the consumption pricing we unpack in our explainer on AI costs, tokens, and credits.

Why Our AI Work Runs on n8n

We are opinionated here, and transparent about why. Nearly every client automation we build is orchestrated with n8n. It is open source, so there are zero licensing costs when self-hosted. It runs on a lightweight server, which means clients host it either on a small VM in their own Azure environment or on a local machine inside their firewall. Every one of our Microsoft-shop manufacturing clients has ended up doing exactly that.

The contrast that surprises IT leaders is with cloud-native automation stacks. Spinning up equivalent capability with native cloud services tends to auto-create a collection of resources that quietly bill every month. A single inexpensive server running an open platform keeps the cost visible and flat. Ongoing AI usage costs on top of that are typically modest, not thousands of dollars a month. And since n8n’s 2026 release added native LangChain integration and more than 70 AI nodes, it has become the strongest of the three for the agent workflows we build, the kind we describe in our guide to agent workflow automation.

The honest tradeoff: someone has to stand it up, secure it, and maintain it. n8n rewards teams with a technical partner or in-house engineering. It punishes teams that have neither.

When Zapier or Make Is the Right Call

Zapier earns its premium in one scenario: non-technical teams automating at low volume. With more than 9,000 app integrations and the most polished builder in the category, a marketer or office manager can connect tools in an afternoon with nobody from IT involved. If your whole automation footprint is a dozen simple workflows, pay for the ease and move on.

Make is the middle path. It delivers roughly ten times the operations per dollar of Zapier and a visual builder capable of genuinely complex logic, without requiring you to host anything. For a mid-market team with moderate technical comfort and growing volume, but no appetite for managing a server, Make is often the best value.

The Decision in Three Questions

Strip away the feature grids and the choice comes down to volume, data, and people. First, how many runs per month will you hit within a year? Under a few thousand, choose on ease. Above ten thousand, per-task pricing will hurt. Second, does your data need to stay inside your walls? n8n is the only one of the three you can fully self-host, which settles it for regulated or security-sensitive teams. Third, who maintains it? Be honest. The platform matters less than the operator, and an unmaintained automation fails silently until it fails loudly. Whichever way you lean, start by mapping the process itself, the discipline we cover in how to automate manual processes.

Our approach is to fit into what you already have rather than sell you a new platform, and to stand behind what we build after it ships. If you want help pressure-testing the choice against your actual workflows, start a conversation with our team.

Frequently Asked Questions

Which is cheapest: n8n vs Zapier vs Make?

At low volume the differences are small. At scale, Make is typically around 70 percent cheaper than Zapier, and self-hosted n8n approaches free on licensing, with costs limited to a small server and maintenance time.

Is n8n hard to set up for a mid-market company?

It requires a server and someone to secure and maintain it, either in-house or through a partner. In our client work it typically runs on a small VM in the client’s existing Azure environment or a local machine behind the firewall.

Which platform is best for AI automation?

n8n currently leads for AI agent workflows thanks to native LangChain support and a large library of AI nodes, plus the option to keep data on your own infrastructure. Zapier and Make both added AI agent features, which are fine for lighter use cases.

Can we use more than one of these platforms?

Yes, and some organizations do: Zapier for simple departmental automations and n8n or Make for high-volume or sensitive workflows. Just assign ownership, because ungoverned tools multiply quietly.

Your On-Prem Servers Are the Real Bottleneck to AI

/by

Ask about the benefits of cloud computing in 2019 and you got a familiar list: lower costs, scalability, remote access. Ask in 2026 and the honest answer is different. We recently sat with the IT team at a mid-market manufacturer where on-prem hardware is what gets approved, largely for CapEx reasons. They have a stable network, a dependable ERP, and no cloud footprint to speak of. They also have a growing list of AI ideas, and every single one stalls on the same question: where would this even run?

That is the 2026 reality. The benefits of cloud computing are no longer mostly about cost. Cloud is now the prerequisite for putting AI to work, and companies still weighing the old pros and cons are answering a question the market has moved past.

Show Image

The Classic Benefits Still Hold

None of the traditional advantages went away. You still trade capital expense for operating expense, scale up without buying hardware, recover faster from failures, and give your team access from anywhere. The market has voted: 94 percent of enterprises now run workloads in the cloud, and public cloud accounts for 45 percent of enterprise IT spending, up from 17 percent in 2021. For a mid-market company, the classic case was already strong. But if that case has not moved your leadership team yet, the next one should.

What AI Changed

AI workloads need three things most server rooms cannot provide: elastic compute, managed data services, and direct access to frontier models. That is why the money is moving so fast. Gartner forecasts worldwide AI spending to grow 47 percent in 2026, and data center systems spending is growing over 55 percent as providers race to host those workloads. AI-related spending is now roughly a fifth of all cloud spending, more than double its share three years ago.

For a mid-market leader, the strategic point is simpler than the numbers. McKinsey puts trillions of dollars of generative AI value across everyday business functions, and nearly all of it is delivered through cloud services. If your data and workflows live only on hardware inside your firewall, that value is hard to reach. Not impossible, just slow and expensive at exactly the moment speed matters.

A Story From the Plant Floor

Here is what this looks like in practice. A manufacturer we work with runs their business on an IBM AS400-era ERP. Compliance documents arrive by email all day, and a team of seven people spends about a person and a half of combined effort manually matching them into the system. The automation opportunity was obvious. The constraint was infrastructure: running AI against the production ERP in real time risked the performance of the system the whole plant depends on.

The answer was not a rip-and-replace migration. We moved a copy of the relevant data into a secure cloud environment, built the pipelines there, and applied AI tools to do the matching. The ERP stayed untouched. The workflow we scoped is worth about $70,000 a year in recovered capacity. That is the modern benefit of cloud computing in one sentence: it is the place where your data can finally meet AI tools without endangering the systems that run your business. The same pattern shows up in most of the projects covered in our guide to how to automate manual processes.

The Objections That Used to Win

Three objections stall most cloud conversations, and all three have aged.

  1. Our data will train someone’s model: Under enterprise agreements with major providers, your data is contractually excluded from model training. This concern deserves diligence, not a veto. Teams that want maximum control can even run local models, accepting the tradeoff of losing frontier capability.
  2. CapEx is how we buy things: Buying servers feels safe because it is familiar. But hardware sized for today cannot flex for AI workloads that spike and idle, and the finance math increasingly favors paying for what you use. Understanding consumption pricing matters here, which is exactly what our explainer on AI costs, tokens, and credits breaks down.
  3. We are not ready: Readiness is built, not waited for. Microsoft’s Work Trend Index shows the workforce is already ahead of infrastructure, with two-thirds of regular AI users spending more time on high-value work. Your people are ready. The question is whether your systems are.

How to Start Without Betting the Company

Skip the grand migration plan. Pick one workflow where money or hours are visibly leaking, move only the data that workflow needs into a secure cloud environment, and build the automation there. Measure the before and after. This proves the value in one quarter, builds internal confidence, and leaves your core systems alone. Our framework for measuring AI ROI before you invest gives you the scorecard.

The benefits of cloud computing in 2026 come down to this: the cost case was always good, but the AI case is decisive. If your infrastructure cannot host AI, your competitors’ can. To find the first workflow worth moving, start a conversation with our team.

Frequently asked questions

What are the main benefits of cloud computing in 2026?

The classic benefits remain: lower capital costs, scalability, resilience, and remote access. The decisive new benefit is AI readiness. Cloud environments provide the elastic compute, managed data services, and model access that AI automation requires.

Do we need to migrate everything to get AI benefits?

No. Most mid-market wins come from moving one workflow’s data into a secure cloud environment and automating there, while core systems like your ERP stay in place.

Is the cloud safe for our business data?

Enterprise agreements with major cloud providers contractually exclude your data from model training and typically offer stronger security operations than a mid-market server room. Diligence still matters, especially for regulated data.

Is on-premise hardware ever the right call?

Sometimes, particularly for regulated workloads or when running local AI models is a requirement. The tradeoff is losing access to frontier models and elastic capacity, so treat on-prem as a deliberate exception rather than a default.

AI Automation Cost Savings: What Marketing Leaders Are Actually Banking in 2026

/by

Picture a marketing team of five people supporting a business unit measured in billions. That is a real conversation we had with a marketing leader at a large manufacturer, and it captures where most mid-market teams sit. The mandate keeps growing. The headcount does not.

AI automation is the obvious answer, and most leaders have tried something. Yet there is a wide gap between buying AI and banking savings from it. One executive we work with put it bluntly: his marketing workflows were running at about 5 percent of what they should be. The tools were there. The dollars were not.

Here is where AI automation cost savings are real for marketing teams in 2026, where they are hype, and how to capture them without adding a hire.

Where the Money Is Actually Leaking

Before chasing savings, find the leaks. The same three show up in almost every mid-market marketing team we work with.

  1. Follow-up that never happens: One manufacturer we work with was spending $60,000 a year on trade shows, and the leads went nowhere. In their words, the team would come home with relationships and then nothing happened. That pattern is the industry norm: up to 80 percent of trade show leads never receive any follow-up. Another team was manually entering more than 2,000 event contacts a year into their CRM by hand, with no tracking and no sequence behind them.
  2. Content that costs too much to sustain: Every leader knows consistent publishing matters, and almost none have the capacity for it. Posting stalls, and the brand goes quiet exactly when buyers are looking.
  3. Leads that sales quietly ignores: When marketing hands over a name and a phone number, a busy sales team works only the obvious wins. The rest die in the queue.

None of these are tool problems. They are process problems, which is why buying more software rarely fixes them. If your processes still run on manual handoffs, it is worth reading how to automate manual processes without breaking what already works before adding anything new.

The Savings That Are Real

So where do the dollars actually show up? McKinsey ranks marketing and sales among the four functions with the most generative AI value at stake, and in our client work the savings land in three places, consistently.

  1. Content production: This is the most dramatic and most measurable saving. We rebuilt our own content pipeline with AI agents grounded in our voice and past work, and the production cost per article dropped from around $1,000 to well under a dollar, with a human reviewing every piece before it ships. Teams we work with are now targeting 4 to 6 times their previous content output with the same people. We documented the full system in how Augusto automates content creation from keywords to conversations.
  2. Lead capture and enrichment: Automating intake from events, forms, and outreach campaigns means every contact lands in the CRM with research already attached. An AI agent takes a bare name and email and returns company, role, and fit before a human ever looks at it. The saving is not the data entry hours. It is the pipeline that stops leaking.
  3. Qualification before the first call: Routing prospects through an intelligent qualification flow lets poor-fit leads opt themselves out. Sales conversations start faster because reps only spend time where there is real alignment.

What these have in common is that the savings come from redesigned workflows, not from a subscription. That is the difference between AI aspiration and AI that actually works in production.

The Savings That Are Mostly Hype

Two claims deserve skepticism. The first is that buying licenses equals saving money. Industry data consistently shows a large share of paid AI seats going unused, and 35 percent of marketers say they juggle too many overlapping AI tools that do not connect. An unused seat is a cost, not a saving. The second is headcount reduction. The teams getting real returns are not cutting marketers. They are moving them up the value chain. One client saw the opportunity clearly: automate the routine content work so their junior marketer could take on search strategy and AI optimization instead. Recent research from Microsoft’s Work Trend Index backs this up, with 66 percent of regular AI users reporting more time on high-value work. The saving is capacity, and capacity compounds. Cutting the people removes the judgment that makes the automation trustworthy.

How to Capture the Savings Without Adding Headcount

The pattern that works is discovery first. Map one marketing process end to end before automating anything. Then automate the repetitive, rules-based steps, keep a human reviewing everything that carries your brand or touches a customer, and measure the before and after in hours and dollars. Start where the leak is biggest, prove the number, and expand. If you want a framework for the measurement side, our guide on how to measure AI ROI before you invest walks through it.

This is also where a partner matters. Advice alone does not produce savings. Someone has to build the workflows, connect them to your CRM and publishing stack, and keep them running as your business changes. That is how we work with marketing teams.

 

What This Looks Like in Dollars

Pull the threads together. Content that cost four figures per piece now costs pocket change, at several times the volume. Trade show budgets in the tens of thousands finally produce tracked, followed-up pipeline instead of business cards in a drawer. And the marketers you already pay spend their hours on strategy instead of data entry, which matches the broader market: 67 percent of marketing teams now report saving 10 or more hours per week with AI. Those are the AI automation cost savings worth chasing in 2026: fewer leaks, more output, and a team doing the work only people can do.

If you want to find the biggest leak in your own marketing operation, start a conversation with our team.

 

Frequently Asked Questions

How much can AI automation save a marketing team?

It depends on the workflow. Content production savings of 90 percent or more per piece are common once a grounded AI pipeline replaces a manual chain. Measure one process before and after to get your own number.

Do AI automation cost savings require cutting staff?

No. The strongest results come from teams that keep their people, with savings showing up as capacity: the same team produces several times the output on higher-value work.

What should a marketing team automate first?

Start where money is visibly leaking. For most mid-market teams that is lead follow-up after events, or content production. Both are measurable within 90 days.

Why did our previous AI tools not save money?

Usually because tools were added on top of unchanged processes. Savings come from redesigning the workflow, grounding the AI in your brand and data, and keeping human review. A license alone changes nothing.

 

Hiring an AI Consultant: What Mid-Market Leaders Should Ask

/by

Search for an AI consultant in Michigan and you will find no shortage of options. National firms with impressive decks. Boutique shops with one specialty. Freelancers who built something clever last year. Sorting real capability from good marketing is the hard part, especially if you have never bought AI work before.

If you want the general criteria for evaluating any AI firm, we cover those in our practical guide to choosing an AI consulting firm. What follows is the local layer, plus the questions that actually predict whether an engagement works.

What makes the Michigan mid-market different

Our economy runs on companies that make things and serve customers, often with workforces that have been in place a long time. Manufacturing alone still employs roughly 580,000 people across the state, and that is before counting the distributors, suppliers, and service firms around it. That combination creates a specific opportunity and a specific risk.

The opportunity is depth of knowledge. Decades of hard-won judgment about products, customers, and edge cases live inside your team. AI can capture and extend that knowledge instead of letting it retire out the door.

The risk sits right next to it. Much of that knowledge is undocumented, and the honest answer to many questions is “it depends on the customer.” Feed that ambiguity into a system without people who understand the business, and you scale the wrong answer efficiently. Someone who has never walked your floor will not catch it.

Regional hiring pressure adds urgency. When roles stay open for months, the practical question shifts from whether to adopt AI to which work your existing team should stop doing by hand. If that pressure sounds familiar, our view on where AI helps when hiring stalls covers it directly.

Five questions to ask before you sign

These come from watching engagements succeed and stall, and each one surfaces something a capability deck will not.

  1. Who runs this after launch? Plenty of firms build and leave. Ask directly who monitors the system in month six, who fixes it when your pricing changes, and whether that sits in the contract or becomes an upsell. Ask whether a guarantee is attached. Every engagement we run includes a check-in every 90 days on results and risks, and we fix what is not working at no additional cost.
  2. Will my team be able to run it without you? One client told us plainly that he wanted his own staff to learn the build, not just receive it, so they could point to the work as their own. That is the right instinct. The aim is to amplify what your team can achieve, not replace who they are, and a partner should leave your people more capable rather than more dependent.
  3. What happens before anyone builds anything? Rework is the most expensive line item in AI work. As one executive put it, spending on the front end beats developing and then discovering you dislike half of what you got. Insist on a discovery step that names where AI goes first and what measurable result you expect.
  4. How will we know it worked? If nobody can define what “correct” looks like before launch, nobody can prove value afterward. Agree on the measure in writing. This single question filters out more weak proposals than any other.
  5. Can you show me results, not just logos? Client lists prove sales ability. Ask instead for specific outcomes: hours recovered, cycle time reduced, error rates cut. Retention is the quieter signal worth probing. More than half of our revenue comes from clients who came back for more, and our first client is still with us today.

Local or national? The honest answer

Proximity is not automatically better. A national firm with deep expertise in your exact problem may beat a nearby generalist, and you should say yes to that.

Where local genuinely wins is the early relationship. Our team sits in the Grand Rapids area, and we start engagements with real face-to-face time before settling into a remote rhythm. Sitting in a room with the people who do the work surfaces what a video call misses: the workaround nobody documented, the spreadsheet quietly running a department, the reason a process has an odd extra step.

Familiarity compounds from there. One long-running client chose us over a much larger national consultancy, reasoning that we already understood their business and could therefore move faster. Notice that the argument was speed, not sentiment.

What good looks like in the first 90 days

A healthy start is narrow and measurable. Expect a structured working session, then one contained pilot with a defined success measure, then a decision point backed by evidence rather than enthusiasm.

Be wary of the opposite pattern. Sweeping roadmaps, long strategy phases, and no working software after a quarter are the conditions in which AI pilots quietly stall. A widely cited 2025 MIT review of enterprise AI found that roughly 95 percent of pilots showed no measurable profit impact, and its authors pointed at adoption habits rather than the technology. Treat that as a caution about the process, not a reason to wait. Momentum comes from proving something small and real, quickly.

Augusto works as an AI activation partner rather than an advisory shop. We start by pinpointing where your people are spending energy that AI should be carrying, then build the solution that changes it, then keep it running as your business shifts. What your team feels is simple enough to describe: they get to do the work they were hired to do, instead of absorbing work a system should be handling.

Put AI to work for your people. The way we start is a Rumble, a two-week working session at a fixed price that produces a prioritized roadmap and one quick win ready to build. It is small enough to say yes to and big enough to show what is actually possible. If you want a straight answer about whether we fit your situation, book a short intro call with our team.

Frequently Asked Question

What does an AI consultant in Michigan typically cost?

Pricing varies with scope, though most mid-market engagements begin with a fixed-price working session before any build. Ask for a phased structure so you can prove value before committing to a larger spend.

Should we hire a partner or build an internal AI team?

Often both, in sequence. A partner accelerates the first wins and trains your people along the way. We compare the tradeoffs in our piece on AI consulting versus in-house AI.

How long before we see results?

A well-scoped pilot should produce evidence within weeks, not quarters. If a proposal pushes the first measurable result beyond 90 days, ask why.

Does our data need to be clean before we start?

Not perfectly. Pick a first use case where the data you already have is good enough to prove the point.

Do we need to be in West Michigan to work with you?

No. We are based in the Grand Rapids area and work well beyond it. Local clients simply tend to get more in-person time early on.

 

When Should AI Make the Call? A Framework for Mid-Market Leaders

/by

Most AI projects do not stall because the technology is weak. They stall because nobody decided how much freedom the AI should have before it went live. That sounds like an engineering detail, yet it is a business call about risk, and it belongs to leadership.

Here is the choice in plain terms. A deterministic workflow follows steps you define, in the same order, every time. A non-deterministic agent picks its own path based on the situation in front of it. Both earn their keep. Confusing them turns a promising pilot into something nobody trusts enough to use.

The difference that actually shows up in your numbers

Deterministic means repeatable. Feed the workflow the same input on Monday and again on Friday, and you get the same result. Because of that consistency, you can test it, audit it, and explain it to a customer or an auditor without hedging. Rules-based pricing, invoice matching, and document routing all belong in this category.

Non-deterministic means adaptive. The agent reads context, weighs options, and chooses what to do next. Naturally, that flexibility is the whole point when the work is messy. One manufacturing leader we work with gets handed 5,000-page compliance manuals to review before quoting. As he put it, an agent reading those pages “may not be perfect, but it’s going to catch the 80%, 90%.” Nobody was reading all 5,000 pages before, so partial coverage beats none.

The question is never which approach is better. It is which one fits the cost of being wrong.

Start with one question: what does a mistake cost?

Before anyone writes a prompt, answer that. A wrong ticket assignment costs a few minutes of rerouting. A wrong number on a customer quote costs margin and credibility. Those two situations deserve very different designs, though teams routinely build them the same way.

We use a simple layered test with clients, and it maps cleanly to how experienced workflow engineers think about the problem:

  1. Risk of error: If a mistake is cheap and easy to catch, lean toward more automation. If the work touches regulated data or goes straight to a customer, add a human checkpoint early.
  2. Nature of the task: If the task runs on clear rules and lookup tables, and you can measure the answer against a known correct value, full automation is realistic. If the task depends on judgment or context the model may read wrong, start it as an assistant to a person instead.
  3. Measurability: If you cannot define what “correct” looks like, you cannot claim ROI later. Define it first.

Most disappointing pilots skip step three. They launch something impressive, then discover months later that nobody can prove it worked. That gap between AI ambition and AI that holds up in production is where most pilots quietly die.

Reliability is something you build, not something you hope for

Here is the part that separates a demo from a system your team will actually rely on. Even when an agent behaves unpredictably by design, the workflow around it does not have to.

On one client quoting workflow, our team built an evaluation trigger directly into the process. The workflow receives expected values, compares its own output against them, and flags anything that drifts. In practice, that means accuracy gets tracked continuously rather than spot-checked by someone with spare time. On another engagement, a second program grades the first one, reviewing whether an information extractor pulled the right fields and then suggesting specific improvements.

Neither addition is glamorous. Both are why the results hold up. The NIST AI Risk Management Framework makes a similar point: measurable, documented controls are what make AI trustworthy over time, not the sophistication of the model.

Move your people from in the loop to on the loop

Early on, a person usually checks every output. That is appropriate, and it is also slow. Over time, the goal shifts. One of our leaders describes it as moving from human in the loop to human on the loop, where the process runs continuously and people review exceptions instead of acting as a cog in the machine.

Notice what that shift does for your team. Your reviewers stop rubber-stamping routine work and start applying judgment where judgment actually matters. Capacity opens up without anyone losing a job. In one finance conversation, a CFO looked at roughly 6,000 annual hours across a three-person accounting team and saw a path to a fraction of that, with the balance redirected toward analysis rather than counting.

One catch is worth naming. Whoever validates the output teaches the system what “right” means. A CEO we work with put it bluntly: much of what a company knows is tribal knowledge, and the honest answer is often “it depends on the customer.” Feed that ambiguity into an agent without a qualified reviewer and you scale the wrong answer efficiently. Choosing your validators therefore matters as much as choosing your tools.

What this looks like when it works

The pattern is consistent across the operations and finance work we run. Lock the steps where rules are clear. Let the agent think where input is messy. Wrap both in evaluation you can show a board. Then put your people on the loop rather than inside it.

Results follow that discipline. On one quoting process, turnaround got fast enough that customers commented unprompted, and roughly 90 percent of the work behind those quotes was automated. Elsewhere, a manual pricing routine consuming three to four hours daily became a background process.

Augusto does not stop at recommending which approach fits. We build these workflows, instrument them, and keep them running as your business changes, because an agent that worked last quarter will drift once your products or pricing move. Deciding where AI goes first is the guidance half. Keeping it dependable in production is the execution half, and you need both.

If your pilot produced a strong demo and an unclear answer about value, that is usually a design question rather than a technology problem, and a short conversation with our team is the fastest way to tell which one you are facing.

Frequently Asked Questions

What is a deterministic AI agent?

It is an automated workflow that follows steps you define, in the same sequence, producing the same output for the same input. Because the behavior repeats, you can test and audit it confidently.

When should we use a non-deterministic agent instead?

Choose adaptive agents when input varies widely and no fixed rule covers it, such as reading long unstructured documents. Accept that output will vary, then design review around that.

Can you make agent workflows auditable?

Yes. Build evaluation into the workflow so it compares output against expected values and flags exceptions automatically. That record is what makes the system defensible later.

How do we know an AI workflow is actually working?

Define what a correct result looks like before launch, then measure against it continuously. Without that baseline, you have activity rather than evidence.