Disclaimer

Everything published on this site is for general information. That includes the guides, the homepage calculator, the sample audit report, the service and solution pages, the FAQs, and anything else we write here. It is written for a broad audience of small and mid-sized businesses in the United States. It is not written for you, your company, your industry, or your situation, because we don’t know any of them.

We write in confident, direct language because hedged writing is useless to read. Confidence about a general pattern is not knowledge of your specific case. Use what you find here as a starting point for your own thinking, not as a conclusion.

Some of our writing uses regulated settings as examples, because that is where friction is easiest to see. Patient scheduling, legal document drafting, bookkeeping and tax workflows, and payroll all show up as illustrations. We are describing operational patterns in those settings. We are not advising on clinical care, on legal matters, on tax positions, or on wage and hour compliance, and nothing we write should be read as telling you what the rules in your field permit.

Compliance is your responsibility. If a change to your process touches a regulated area, get it reviewed by someone qualified to review it before it goes live.

Reading this site creates no relationship between us. Neither does sending a contact form, booking a call, subscribing to anything, or emailing us a question. No professional-client relationship, no advisory relationship, and no fiduciary duty arises from any of it.

We take on obligations to you only under a signed agreement or a purchased engagement, and the scope of those obligations is whatever that agreement says. See the Terms of Service for what an engagement with SnagLabs AI LLC actually commits us to.

Information you send us before an engagement exists is not covered by any duty of confidentiality we haven’t agreed to in writing. We handle it carefully and as described in the Privacy Policy, but please don’t send us anything sensitive, privileged, or subject to somebody else’s confidentiality obligations in an unsolicited enquiry. We’re glad to sign a mutual non-disclosure agreement first.

The calculator on our homepage takes three numbers you choose: how many people are affected, an estimated hourly cost, and how many weeks a year they work. It returns hours back per year and what those hours are worth. It is arithmetic on your inputs and nothing more. It is not a quote, a forecast, a valuation, or a projection of savings, and its output is not a figure we are standing behind.

We would rather tell you what it leaves out than let a big number do the arguing. The calculator assumes one hour is returned per affected person per workday, and then it ignores everything that makes real savings smaller:

• Adoption. It applies no adoption rate. It assumes every affected person uses the new process, correctly, every day, from day one.
• Implementation cost. It subtracts nothing for the cost of building, licensing, or maintaining whatever produces the saving.
• Ramp-up. It assumes the full saving starts immediately. Real changes have a period where things get slower before they get faster.
• Discounting. It does not discount future hours to present value, so a dollar saved next December counts the same as a dollar saved today.
• Whether the hour is worth anything. An hour freed is only a saving if it gets spent on something useful, or if it comes off the payroll. The calculator can’t know which.

The same applies to any figure, range, benchmark, or rule of thumb anywhere on this site. Where we cite numbers in the guides, they are illustrative rather than sourced research, and you should not treat them as measured findings. Your own results depend on your data, your people, your systems, and your follow-through.

The businesses in our writing are illustrations. The dental practice, the construction contractor, the solo consultant and every other example like them are composites: patterns we have seen assembled into a readable scenario. Some are written as if they were anecdotes because that reads better than a list of abstractions. They are not accounts of identifiable clients, and no detail in them should be taken as a fact about a real business or a real result.

The sample audit report on the audit page is labelled as made up to show the shape of a report rather than a real client’s report, and that label is exact. Its findings, figures, and recommendations were written to demonstrate format. They are not a template of what we will find in your business and not a floor or a ceiling on it.

Our long-form guides are opinion pieces with frameworks in them. When one tells you to build, buy, improve, integrate, or leave a system alone, or when it argues that the right amount of AI for a given job is none, that is our editorial judgment about a general case. It is not a recommendation for your case and not a substitute for looking at your own numbers.

They also go out of date. Software changes, prices change, and what was true about a category of tools last year may not be true now. We write about a fast-moving field. A guide reflects our view at the time it was written, and we don’t undertake to review, update, correct, or withdraw any of it on a schedule. If a page still reads as current and no longer is, that is the nature of published writing rather than a representation by us.

We may add, rewrite, or remove any of this content whenever we like, without notice.

We build AI systems, and we are blunt on this site about what they do badly. An AI system produces probable output, not correct output. It can be confidently wrong. It can invent a fact, a citation, a total, or a policy that does not exist, and it will present all of them in the same fluent tone as the parts it got right. Feed it bad data and it will be wrong faster and more persuasively than a person would be.

So: a qualified human has to review AI output before anyone relies on it, especially where money, safety, employment, health, legal exposure, or a customer commitment is involved. Automation does not transfer judgment or accountability from you to a model, and it is not a substitute for professional review by someone licensed to give it.

That applies to anything we build with you as well. We design for review points and we will tell you where they belong, but the person accountable for a decision is still the person accountable for it.

How this differs from the One-Hour-Back Guarantee

We do offer one guarantee, and it is narrower than it is sometimes read to be. The One-Hour-Back Guarantee is a contractual refund term about the audit itself: if we cannot identify at least five hours per standard workweek of practical, measurable time savings, we refund the audit fee. It is a commitment about what the report finds.

It is not a guarantee that your business will realize those hours. Realizing them depends on implementing the recommendations and on your people actually adopting the change, and neither is within our control or covered by the refund. The guarantee also carries conditions and a claim window. The operative language, including your obligations under it and the deadline to claim, is in the Terms of Service, and that document controls if anything on this page reads differently.

We name other companies’ products and services throughout this site, because you can’t write usefully about business software without naming software. Those names, logos, and trademarks belong to their owners and are used for identification and commentary only, under nominative fair use.

Naming a product implies no affiliation with, sponsorship by, partnership with, or endorsement by its owner, in either direction. Where we hold a genuine partnership or certification with a vendor, we say so explicitly and name it. If we haven’t said so, assume there isn’t one.

Where we link out, it’s because the destination was useful when we linked it. We don’t control those sites, we don’t review them on an ongoing basis, and we’re not responsible for their content, their accuracy, their products, their privacy practices, or what they do with your information. A link is not an endorsement, and following one takes you somewhere our policies stop applying and theirs begin.

We try to keep this site accurate and working. We can’t promise it is either. Content may contain errors, omissions, outdated statements, or typographical mistakes, including in prices, features, and descriptions of what we do. Where an obvious error conflicts with what we have agreed with you in writing, the written agreement governs.

The site is provided as-is and as-available. We may change, suspend, move, or remove any page, guide, tool, or feature at any time without notice, and we don’t promise uninterrupted or error-free access. We reserve the right to correct any error and to decline to honor a price or term that was published in error.

This page describes what our content is. The legally operative disclaimer of warranties and the limitation of our liability, including the cap on damages and the deadline for bringing a claim, are in the Terms of Service, and they apply to your use of this site and everything on it.

Nothing on this page expands our liability, creates a warranty, or gives you a right you don’t have under those Terms. Where this page and the Terms disagree, the Terms control. Nothing here excludes any liability that can’t be excluded under applicable law.

If something on this site looks wrong, we’d genuinely like to know. Write to legal@snaglabs.ai and we’ll fix it or explain why we think it’s right.