How HobbyHQ works
What our numbers mean, where they come from, and what they cannot tell you. If any of this reads as hedging, that is the point — the alternative is stating things more confidently than we can support.
Each activity in the catalogue is described by a fixed set of fields: what it costs to start and to keep doing, whether it is indoors or out, how much space and equipment it needs, how physically demanding it is, how steep the learning curve is, how rewarding the first session is, how much mess and noise it makes, how portable it is, how much the weather matters, whether children can join in, what level of competition exists, how risky it is, and what you come away with.
These are the same fields the hobby finder matches on and the browse surfaces filter on. That is deliberate: it means the description on an activity page and the belief the recommendation engine holds about that activity are the same object, and cannot drift apart.
They are assessments, not measurements. A cost band is an editorial judgement about what a sensible entry setup costs in the UK, not a survey of prices. A learning-curve rating is a judgement about how long competence takes, not a study. We record them as bands and five-point scales rather than as precise figures because a band is honest about its own precision and a decimal is not.
The finder is deterministic and inspectable. There is no model involved, nothing is learned from other readers, and the same answers always produce the same results. Every suggestion shows you which of your answers it matched, which it did not, and which we have not assessed for that activity.
Your answers fall into two kinds, and the difference matters. Some are constraints: if you say you have £25 and no space, an activity that needs £400 and a workshop is not shown at all, however well it fits otherwise. Others are preferences: they change the order but never exclude.
Constraints — these exclude
- Indoors or outdoors
- Budget to start
- Space available
- Equipment needed
- Mess
- Noise
Preferences — these rank
- Physical demand
- Alone or with others
- Time per session
- Competition
- All-weather
- Portable
- Who it is for
- What you get out of it
- Rewarding from the start
The score is the weighted proportion of applicable criteria that matched, so it is a measure of fit against what you told us and nothing else. A 100% fit on two answers means much less than an 80% fit on nine, which is why the number of answers you have given is shown alongside.
Unassessed is never treated as a match. Most of the catalogue is classified but not written up, and those entries carry only the core attributes. A criterion that depends on something we have not assessed reports as unknown and scores zero — and, importantly, never excludes the activity, because the gap is ours rather than the activity’s.
HobbyHQ is intended to grow indefinitely, so the order we write things in is the most consequential choice we make. We score every subject against one idea: how many people want to know about something is not the same as how badly they are served.
A subject can attract enormous interest and still be the wrong thing to write next, because it is already answered well by publishers with decades of authority. What we look for is the combination of real interest and information that is scattered, thin or out of date. The scoring reflects that directly — fragmentation and coverage weakness carry 40 points between them out of 100, more than demand’s 18.
Every assessed subject falls into one of four classes, and the first two are the reason the model exists at all:
- Underserved opportunity
- Strong or rising interest, information scattered across forums, video and retailer pages, and no single good answer. This is what HobbyHQ is built for.
- Saturated mainstream
- Plenty of interest, but already covered well by established publishers. Worth having; not worth leading with.
- Niche depth
- A smaller audience, but one that is genuinely badly served. Cheap to serve properly and unusually loyal once served.
- Catalogue only
- Belongs in the catalogue so the finder can recommend it, but does not yet justify a deep build.
There are no search volumes in this model. Nobody has run a keyword tool against these subjects, and a fabricated figure would be indistinguishable from a measured one and would corrupt every decision downstream of it. So we record ranges instead, each with how confident we are, how we arrived at it and when — all 1538 of them are currently our own editorial estimate, and every one is marked as such, so measured figures can replace them plainly when we have any.
A comparison only exists if it corresponds to a decision somebody genuinely faces, and each one states that decision at the top. Padel versus pickleball is a real choice people make; padel versus crochet is not, and could not be written honestly.
We never declare a winner. There is no overall verdict, no score and no “our pick”, because a single answer for everybody is either wrong for most readers or describes a decision nobody had to make. Instead each comparison sets out the dimensions on which the two genuinely differ, marks which option has the advantage on that dimension, and says plainly who each option suits.
Costs. Bands reflect typical UK prices for a sensible entry setup at the last review date. They will be wrong for any specific purchase, they do not know what you already own, and second-hand availability varies enormously by region. Use them to compare activities, not to budget one.
Time. Session lengths are typical rather than minimum. Almost everything here can be done in less time than we state, less well.
Difficulty. Learning curves and beginner friendliness are judgements about the median newcomer. Prior experience in a related activity changes both substantially, and some people find the “hard” thing easy.
Risk. Our risk ratings describe an activity practised sensibly with appropriate instruction. They are not a substitute for the governing body’s safety guidance, a qualified instructor, or the manufacturer’s instructions — and where an activity involves blades, kilns, heat, chemicals, open water, heights, traffic or weapons, that guidance takes precedence over anything on this site.
An activity is only published once it computes to at least substantial depth, and that is measured rather than declared: word count, number of sections, number of distinct kinds of section, number of questions answered. A writer cannot mark their own work substantial: an activity that does not meet the bar cannot be published at all, whatever anybody thinks of it.
We are working towards a launch threshold of 30 substantial subjects with at least 8 in each of hobbies, sports and crafts, so that no strand launches empty. Currently 121 are written up out of 1538 in the catalogue. The rest are classified, searchable and usable by the finder, and are labelled as not yet written up wherever they appear.
HobbyHQ is not sponsored by any club, governing body, retailer or manufacturer, and no organisation has paid for coverage or for a favourable assessment. Where we cite a governing body it is because it owns the rules, not because it has any relationship with us.
The site is designed to carry advertising through a central platform, and there are deliberate limits on that: no page carrying a safety or legal warning is commercially eligible, and neither is any page we have not written up properly. Nothing commercial is requested until a reader agrees to it. How that works.
Found something wrong? Tell us — corrections matter more than completeness. See evidence standards for how we handle sources, and about for who we are.