Nothing below is waiting to be built. Each one is something people go looking for, that was investigated and then decided against, and several were built and working before they were taken out again. The reason is given in each case, along with what to do instead, because a refusal without an alternative is just an apology.
It is published for a plain reason: it is the most useful thing we can tell you about the other tools on this site. Anybody can claim their figures are right. A site that will name what it refused to guess at has given you something you can actually check it against.
It would be a specification, and nobody here can sign one
A tool that prints a section size against a span and a load is not information, it is a design. Somebody orders that beam. A specification belongs to whoever signs it and there is nobody here to sign it, so a disclaimer under the number does not fix it: the number is what gets used and the paragraph is what gets skipped. This is the single largest category and it was decided by removing tools that already worked.
A beam sizer, or anything that picks a steel section for you
Eleven tools were removed from this site in one go on 12 August 2026, having already been built: a beam sizer, weld capacity, bolt group, baseplate, purlin spacing, lintel span, slab thickness, pad foundation, bearing pressure, padstone and rafter span. Each printed a size or a capacity. The argument against a warning note is simple and it decided it: the number is what somebody orders from, and the paragraph underneath is what nobody reads.
Instead. Work out what you need to ask for, then ask somebody who can put their name to the answer. The engineer brief tool exists for exactly that conversation, and going into it knowing your spans, loads and constraints is most of the value an engineer would otherwise spend an hour extracting from you. What to tell your engineer.
The document costs money, and we do not guess
Where being correct requires a standard that has to be bought, the tool stays out rather than being approximated. Nothing on this site is behind a payment and nothing on it was paid for, which means a paid standard is not a purchasing decision here, it is a wall. Reproducing a table from memory or from a secondary source is the failure mode this rule exists to prevent, and it is worse than an absent tool because it looks identical to a correct one.
A joist span table
The span tables Approved Document A points at are published by TRADA and are a paid document. Reproducing them from memory is exactly the thing this site does not do, and a span table that is nearly right is indistinguishable from one that is right until a floor moves.
Instead. The joist span tool here works the span out from the deflection equation rather than from a table, which is legitimate because the equation is public and the published tables were built from it. It agrees with TRADA to within 0.3%. It is deliberately a COSTING tool: it leads with how many joists and how many metres, and it will not tell you which one to buy. Joist span table & count.
Rebar lap lengths, rebar cover, or soakaway sizing
Lap lengths come from Eurocode 2, cover from the BS 8500 exposure classes, and soakaway sizing from BRE Digest 365. All three are paid documents, and Approved Document H points at BRE 365 rather than reproducing a method, so there is no free primary source to report from either. These were scoped on 16 August and cut for that reason rather than for want of interest.
Instead. The rebar weight tool covers the arithmetic that needs no standard: bar masses are calculated from diameter and density, so a schedule, a weight and a bar count are all exact. For cover and laps, the specification on your drawing is the authority and it is there because somebody was paid to decide it. Rebar weight & schedule.
Australian steel sections, timber framing, masonry or concrete grades
AS/NZS 3679.1 for steel sections, AS 1684 for timber framing, AS 3700 for masonry and AS 1379 for concrete are all Standards Australia documents and all cost money. It is the same wall the American structural tools hit with the ICC codes, and it is why the structure hub is thin for Australia in a way that is not neglect.
Instead. Where the maths is universal rather than tabulated, the tool works in Australia already: volumes, weights, quantities and areas do not have a nationality. What is missing is the reference data, and your supplier publishes the section properties of what they actually stock.
US notching and drilling limits, or strip foundation widths
IRC R502.8 for notching and R403 for footings are International Code Council documents. ICC offers free read-only access behind an account, and this site has no accounts, so that route is closed by the same rule that keeps the tools free. Searched on 20 August across awc.org, fema.gov, pnnl.gov, energy.gov, usda.gov and hud.gov: none carries the section text.
Instead. The British notching tool states the Approved Document A zones, which are free to reproduce, and the geometry of a notch is the same everywhere even where the permitted zone is not. Your local building department will tell you the adopted code edition, and that is the answer that actually governs, because the model code is amended state by state. Notching & drilling limits.
The source is free and will not let us read it
A different problem from a paid one, and a more frustrating one: the document is public, it is authoritative, and the server refuses automated access. Where every route is closed, the figures cannot be checked, so the tool does not ship. This category can change without warning, in either direction.
The NIOSH lifting equation
The revised equation is a genuinely good instrument, free, and published by a US federal agency. On 24 August 2026 every route to the Applications Manual was refused: cdc.gov returns 403 to a fetch, stacks.cdc.gov is denied in a browser and returns 403 to curl. The frequency multiplier is a fifteen row table and the coupling multiplier is another, and neither will be written down from memory.
Instead. NIOSH publishes a free NLE Calc app for iOS and Android that does exactly this, built by the people who wrote the equation. Use theirs. Our manual handling tool covers the British guideline figures, which is a different instrument answering a nearby question. Manual handling limits.
One person cannot keep it right
Some tools are correct on the day they are written and wrong within a year, in fifty places at once. A fifty-state table or a thirteen-province one is not a build problem, it is a standing commitment, and this site is one person. A quietly stale page that looks current is worse than no page, so the honest answer is not to start.
Anything that needs a fifty-state or thirteen-province table
US unemployment tax rates, sales tax nexus thresholds, workers compensation figures and state paid family leave schemes are all fifty separate answers that move independently. Canadian employment standards are thirteen. Getting them right once is a week. Keeping them right is a job, and this site is one person who has said so on every page that carries a figure.
Instead. Where a federal or national answer exists, the tools here anchor on it and say so: Canadian tools name the Canada Labour Code or say explicitly that they are Ontario. For a state or provincial figure, the regulator publishes it free and it will be more current than any aggregator.
It is somebody else’s work
Question banks and examination material are written, owned and often sold by the body that sets the test. Practice questions written from the published syllabus are fair and are what this site does. Reproducing the real bank is not, however easy it would be.
Driving theory practice outside the UK
The British theory test has a published syllabus, which is what the mock tests here are written from. American learner permit tests are per state and Canadian ones per province, and the question banks are owned by those authorities. That is fifty or thirteen sets of somebody else’s material, which fails on ownership and on upkeep at the same time.
Instead. Most DMVs and provincial licensing bodies publish their own free practice tests and an official handbook. Theirs will match the real questions because theirs are the real questions.
The real CSCS or DVSA question bank
The live banks are not published, and where revision material exists it is sold by the body that sets the test. Every mock question on this site is written from the published syllabus and from the source documents behind it, which is legitimate and is also why they will not be word for word what you sit.
Instead. Use these to find out what you do not know, then read the source. The tools name the document behind each question for that reason rather than as decoration. CSCS mock test.
What would change any of this
Three of the categories above could move, and it is worth saying which, because a permanent refusal and a temporary one are different things.
A blocked source can open. Those rows record the date they were tried for exactly that reason. A government server that refuses automated access today may not next month, and the tool goes in the day it can be checked.
A paid standard could be reproduced by somebody with the right to do it. Where a national laboratory or a regulator republishes a table under an open licence, that is a legitimate source and it has already unlocked work here: the US insulation figures came off a Department of Energy laboratory rather than from the code body, because they had published the table themselves.
A design tool would need somebody who can sign it. If a chartered engineer ever wanted to author and own one of those eleven, it could exist with their name on it. That is a real route and it is not one that can be taken by writing a longer disclaimer.
What will not change is the fifty-state problem. That is not about access or money, it is about one person keeping fifty answers current, and no amount of effort at the start fixes what happens in year two.
If something is missing that should not be
This list covers what was considered and declined. It does not cover what nobody has thought of, and that is a much longer list. If there is a tool you keep wanting and cannot find a decent free version of, ask for it. Small, unglamorous tools with no competition are the best things on this site, and they mostly exist because somebody mentioned needing one.