ServicesFree AI usage review
Free AI usage review: one page on where your AI money goes and whether it pays off
You see where your company's AI money goes, whether it pays off and what to stop, start or change, on one page you can take into the next budget or board discussion. You send the usage and billing exports you are comfortable sharing, for AI tool subscriptions, model APIs, cloud AI services and coding agent tools. We read them and write the page without any access to your systems. It is free, and it is written for managing directors and finance as much as for IT.
How the review works
- 1
A 30-minute call: which tools and models your teams use and which exports you can share.
- 2
You send the exports, aggregated or anonymised as you prefer.
- 3
We write the one-page review.
- 4
We walk you through it in a second call. What you do with it is up to you.
What management gets
A clear view of AI spend before the next budget decision. In Bitkom's survey of 14 September 2026, 54% of companies that use AI named hard-to-estimate costs as a barrier. The review turns your own invoices and exports into answers management can act on:
It needs no project and no access to your systems, only the exports you choose to send. The page stays with you, whatever you decide.
- Where the AI money goes, and which tools and teams account for most of it.
- Whether it pays off, as far as the exports show: spend tied to regular use, and licences or keys that nobody uses.
- What to stop, start or change, each with the expected effect and how to check it.
- A page to take into the next budget or board discussion, with the numbers it rests on attached.
What you get
One page, written for the people who decide, plus an attachment with every number it rests on, and a second call in which we walk you through both. The detailed figures and model names in the attachment let your IT lead check each finding. The page itself stays short enough to take into a budget meeting.
What the page looks like
Every review uses the same four parts, with your numbers in them:
- Summary: spend over the period you sent, split by tool and model, and what changed.
- Findings: for example, a large model doing a task that a smaller one also passes, or spend no team owns.
- Three changes, each with the effect we expect and how you can check it.
- What to log from now on, so the next review can compare.
What we need from you
Only what you choose to send: billing or usage exports from model providers, your cloud account and coding agent tools, a list of the tools and teams that use them, and anything you already measure. Names can be removed or replaced by team labels. We do not need access to your systems, your source code or your prompts.
Why it is free
It is how we start conversations: you see how we think before you decide anything, and we learn whether there is work worth doing together. There is no obligation, and the page is yours to keep.
Questions
We already pay for AI tools. How do we know whether they pay off?
Start with what the money buys: which tools and models are used, how regularly and by which teams, and which licences or keys sit idle. The review shows that from your own exports and says what to measure next, such as cost per handled case, so the following budget decision can compare cost with benefit.
What is the best first step if AI is already in use in our company?
Find out which tools people use and what they cost, before buying more or banning anything. The free review does exactly that from exports, without a project. It also shows where a team pays for several tools that do the same job and where nobody owns the spend.
Can non-technical managers use the page?
Yes. The page is written in plain words for the people who decide on budget, with amounts, owners and recommended changes. The detailed numbers and model names are in the attachment, so your IT lead can check every figure.
Is it really free? What is the catch?
It is free and without obligation. We do it because it is the clearest way to show how we work. If the page suggests work you want help with, we talk about that separately.
Do you need access to our systems?
No. You send exports you choose. We never log into your accounts, tools or repositories.
Does this involve personal data?
It does not have to. Usage exports can be aggregated by team or anonymised before you send them. If any personal data remains, we agree how it is handled first and can sign a data processing agreement and a confidentiality agreement.
Is this a way of monitoring employees?
No. The review looks at tools, models and teams, never at individuals, and we ask for exports aggregated by team or with names removed. Data from these tools can still fall under co-determination, so check with your works council before you send anything. We can show them exactly what we ask for.
Will you push us towards a particular tool or vendor?
No vendor pays us. The review recommends changes to how you use what you have. If a different tool would help, it says so and why.
What if we do not have usage exports?
Then the page says what to switch on so you have them, and we start from invoices and a description of how the tools are used.
Did our AI investments meet expectations, and how do we know?
You know only if the expectation was written down as a number before the money was spent, such as hours per case or cases per week, and measured again afterwards on the same kind of work. In KPMG's 2026 study, 71% of companies say their AI investments met or exceeded expectations. The measures of success they named were productivity and efficiency gains (65%) and contribution to revenue or growth (48%). For licences and model usage, the free review shows from your exports what is used, by which teams and at what cost. For systems you built, compare today's cost per handled case with the baseline. If there is none, measure the current process now so the next decision has one.
Where are we not using the AI we already pay for?
Look in three places: licences that few people use, tools kept for small tasks while routine work on documents and requests stays manual, and several tools doing the same job in different teams. It is common: in Bitkom's survey published in September 2026, 87% of companies said they use the potential of AI little or not at all. The free review finds the first and the third in your exports: idle seats and keys, spend per team and overlapping tools. The second does not show in exports, so in the follow-up call we ask which routine work is still done by hand and whether the tools you have could take it on.
Related services
AI system testing
Before go-live, acceptance tests on your own cases tell you how often your chatbot, assistant or agent is right, in terms your QA team already uses.
AI evals engineering
Evals built from your own traces show whether each change made your AI product better or worse, with a release gate that stays in your repository.
AI agent evals
Find out how often an agent gets the job done before it acts in your systems, by checking outcomes, tool calls and how reliably runs repeat.
RAG evaluation
An assistant that answers from your documents, tested on retrieval and on answers in each language your users write in, so you know how often it is right and cites the right source.
On-premise AI
AI on offers, contracts and customer files, with the data staying in your company: own servers or own cloud account, tested before go-live.
Data integration for AI
Answers from file shares, SharePoint, ERP and tickets in seconds, for assistants and agents, with permissions carried through and tested.
AI agents and process automation
Agents that take on the routine steps of complex cases and pass decisions to people, with a cost per handled case.
From the blog
AI business case
How to work out whether AI pays off for one process: measure today's cost per case, count every cost of the AI system including the people who check it, and agree in advance when to stop. With a worksheet for your own figures.
Shadow AI: ban or better option
Shadow AI grows where companies encourage AI without providing it, and bans push it onto private phones. An approved platform that enforces your AI policy works better.
LLM cost allocation
How to explain an LLM bill by team, task and accepted outcome: tag requests at the gateway, price them as you are invoiced, allocate shared costs by a stated rule and reconcile with finance every month.
Full cost of an AI workflow
How to compare AI options on the cost of a handled case, including review, escalations, rework and upkeep, with a worked example and its assumptions.
Talk to an engineer
Find out where your AI money goes and what to change first. Ask for the review in the form below, or book the first 30-minute call directly. We reply with what to send.
Booking runs on Google Calendar.
Prefer to write first? Leave your email and a few lines, and we reply with times.
Look for the draft in your email program
Send it from there, and an engineer replies in person with times for a 30-minute call. If no draft has opened, write to us at miki@cloudsail.com.