Are you AI-ready? A four-week assessment framework for mid-market analytics

Most AI readiness talk stays vague. Ours is a four-week, fixed-price assessment that scores your data foundations against AI-readiness and hands you two or three costed use cases, or an honest recommendation to wait.

Ask ten mid-market leaders whether their business is ready for AI and you will get ten different answers, most of them guesses dressed up as confidence. The honest answer usually depends on things nobody has measured yet: how consistent the underlying data actually is, whether the same metric means the same thing in two different reports, and whether anyone would notice if a model quietly started making a bad call. AI readiness is not a feeling. It is a set of conditions that either exist in your data estate or do not.

That is the gap our AI Readiness Assessment closes. It is a four-week, fixed-price engagement that scores your data foundations against a structured maturity model, benchmarks you against comparable organisations, and ends with either two or three costed AI use cases or an honest recommendation to spend the next six months on foundations instead. This post covers how the assessment works, how the scoring model behind it holds together, and who tends to get the most out of it.

Why most AI conversations start in the wrong place

Most AI conversations start with a tool. Someone has seen a demo, read a case study, or been pitched a copilot, and the conversation becomes about which product to buy rather than whether the organisation is in a position to use it safely. That ordering causes most AI failures. A model trained on inconsistent data will produce inconsistent output with total confidence. A chatbot given access to reports nobody has reconciled will happily hallucinate a number that looks exactly like every other number in the report. The tool is rarely the problem. The foundation underneath it is.

Our assessment starts in the opposite place. Before any conversation about which use case or which platform, we score where the organisation actually sits today, using the same maturity model we use across every engagement, then layer an AI-specific lens on top of it.

The four-week format

The assessment runs as a fixed-scope, fixed-price engagement over four weeks. Week one is discovery: we look at your data foundations and, just as importantly, the decisions AI is actually supposed to inform. There is no point scoring data quality in the abstract if nobody can say what a model would be used for. Week two is gap analysis, comparing current state against what AI-readiness actually requires. Week three produces two or three costed AI use cases, each with a Model Trust One-Pager already drafted, covering what the model would do, what it would need, and how you would know if it was wrong. Week four delivers a written recommendation. Sometimes that recommendation is to proceed. Sometimes it is to spend six months on foundations first. There is no obligation to continue with Hopton afterwards either way.

How scoring works, and why the five profiles matter

The underlying maturity model scores five dimensions of your data estate, each on a scale of one to three. Be honest when scoring: if you sit between two levels, take the lower one. An inflated score gives a false picture and leads to the wrong decisions. The total, somewhere between five and fifteen, tells you roughly which stage you are at. The shape of your individual scores, not just the total, tells you what to fix first.

In practice, most organisations land in one of five recognisable profiles, from the spreadsheet-driven business running on manual exports at the low end, through tool-rich but governance-poor setups where dashboards exist but nobody trusts them, up to organisations with genuinely governed, well-documented data that are ready to layer AI on top with confidence. Naming the profile matters more than the number, because it tells you what kind of problem you actually have. A low score caused by weak governance needs a very different fix to a low score caused by fragmented tooling.

We also benchmark your score against comparable organisations, because the most expensive analytics decisions are rarely the ones made with too little ambition. They are the ones made without knowing how far behind, or ahead, you actually are compared to the market you compete in.

Who gets the most out of it

The assessment is built for senior leaders, IT directors, finance directors, and anyone else who owns data and analytics decisions. You do not need a technical background to complete it. It works particularly well when two or three people from the same organisation complete it independently and then compare answers, since the gaps between their scores are often as revealing as the scores themselves. A finance director who rates data quality highly and an IT director who rates it poorly are both telling you something true.

Readiness is not a one-off badge. Foundations shift as systems change, teams turn over, and new data sources get bolted on, so we recommend re-taking the assessment periodically rather than treating the first score as permanent.

What this means for you

If you cannot yet say, with evidence rather than confidence, how consistent your data is and what decisions AI would actually inform, that is the assessment talking. It is a four-week, fixed-price way to find out before you commit budget to a platform or a use case that your foundations cannot support. If AI does not fix bad data, it only makes bad data faster, so the assessment exists to tell you honestly which side of that line you are on, and what to do about it. Get in touch to arrange one.