Why it arrives as a PDF, and why that is not going to change
A premium schedule is a statement of record. It is the document that says what is covered, for how much, over what period — and it is issued in a form that cannot be edited, because the whole point is that it is the version that counts.
That is entirely reasonable and completely unhelpful for every downstream use. Allocating premium across sites, comparing this renewal with last, checking that a disposed property has actually come off the schedule: all of those are table operations, and none of them can be done in a PDF.
So the work gets done by hand. Someone types forty locations into a spreadsheet once a year, and next year someone types them again, and the two years are never quite in the same shape — which is exactly the shape you need them in to see what changed.
This page is about removing the typing. What comes out is a table with one row per schedule line, each row carrying the file and page it came from, which is what makes it checkable against the document that remains the record.
What is actually in a schedule
Layouts vary enormously between insurers, but the content converges. Almost every commercial schedule carries a header block describing the policy and a body listing what is covered.
| Part | Typical contents | What makes it awkward |
|---|---|---|
| Header | Insured, policy number, insurer, period | Insured name differs from the legal entity |
| Schedule lines | Location or asset, cover, sum insured | Runs over several pages with repeated headers |
| Premium block | Premium, IPT, fees, total | Sometimes per line, sometimes only in total |
| Excesses | Per cover, sometimes per location | Stated as prose rather than in a column |
| Conditions and warranties | Requirements attaching to cover | Free text, often on a separate page |
| Endorsements | Mid-term changes to the above | Issued separately and easily forgotten |
The last row is the one that quietly causes trouble. An endorsement issued in month seven changes the schedule and arrives as its own document, and a spreadsheet built from the original at renewal will not know about it. Reading endorsements into the same table as the schedule is the only reliable way to keep the two in step.
The fourth row is the reason a naive table extraction disappoints. Excesses are frequently written as sentences underneath the table rather than as a column in it, and a tool that only understands grids returns a schedule with the excesses missing.
Four things you can only do once it is rows
Allocate the premium
Charge insurance out to sites, departments or entities on a basis you can show — usually sums insured, sometimes headcount or floor area.
Compare renewals
Put last year and this year in the same columns and see what moved: rates, sums insured, locations added or dropped.
Check the asset list
A schedule is a list of what you told the insurer you have. Comparing it with your fixed asset register finds both directions of error.
Feed the accounts
Premium, IPT and fees as separate figures, ready for the prepayment schedule and the ledger rather than retyped from a PDF.
The third one earns its place more often than people expect. A schedule and a fixed asset register are two lists of the same things maintained by two different people for two different reasons, and they drift. Property sold two years ago still being insured is the entertaining version; a new site nobody added is the expensive one.
Convert a schedule now
Upload the PDF — or a scan of it — and see what comes back. The free tier needs no registration, and the useful test is the schedule with the most locations rather than the tidiest one.
Upload a document
Drop your documents here
or browse to upload
Free: 10 pages / month · Upgrade for more
Files encrypted in transit · auto-deleted after processing · GDPR compliant
The fields that come out
| Field | Where it comes from | Why it is worth having separately |
|---|---|---|
| Insured and policy number | Header block | The key that joins renewals and endorsements together |
| Insurer | Header block | Groups often carry several insurers across one programme |
| Period of cover | Header block | Drives the prepayment and the renewal diary |
| Location or asset | Schedule line | The unit everything is allocated by |
| Cover type | Schedule line | Buildings, contents, stock, BI — each behaves differently |
| Sum insured | Schedule line | The basis for allocation and the number that drifts |
| Premium | Premium block or line | The cost, before tax and fees |
| IPT | Premium block | Not VAT, generally not recoverable, must not be merged |
| Excess | Line or prose below | Needed at claim time, missing from most spreadsheets |
| Source file and page | Added by us | Turns any row back into the document it came from |
Where a field is not on the document it stays empty. That is deliberate: an empty cell is something you can fill in, while an invented one is something you do not know to check. On a document that governs what happens after a loss, that distinction is not academic.
How it works
1 · Upload
The schedule as a PDF, a scan or a photograph. Up to 100 files in one pass for a full renewal cycle.
2 · Read by meaning
No per-insurer templates. An unfamiliar layout works on the first document, and a redesigned one keeps working.
3 · Arithmetic checks
Line premiums against the total, IPT against the premium, sums insured against any stated aggregate.
4 · Confidence flags
Anything read with low certainty is marked, so review is targeted rather than a full re-read.
5 · Export
Excel, CSV, JSON or XML, with the source file and page on every row.
6 · Use it
Allocate, compare, reconcile against the asset register — the things the PDF was blocking.
Step three matters more on schedules than on most documents. A single mis-read digit in a sum insured is invisible in the row and obvious in the total, and a sum insured is the one number on the page where an error changes the outcome of a claim rather than the cost of a month.
Multi-location schedules, where the hand-typing lives
A single-site schedule takes five minutes to retype and nobody looks for a better way. The pain starts at around fifteen locations and becomes acute at a hundred, which is ordinary for property portfolios, franchise groups and anything with vehicles.
Three things make these documents awkward beyond their length. Rows continue across page breaks, with the column headers repeated at the top of each page and the totals appearing only at the end. Locations are described inconsistently — a trading name on one line, a postal address on the next. And sub-limits often apply to groups of rows rather than to individual ones.
Following rows across pages rather than restarting at each header is the difference between a complete schedule and one that quietly stops at page four. It is also the failure most likely to go unnoticed, because a schedule with three of four pages looks entirely plausible.
For vehicle schedules the same logic applies with a different key: registration rather than address, and often several hundred lines with values that change every renewal. The extraction is identical; only the column you sort by differs.
Comparing renewals, which is the point for most people
The question at renewal is never just “has the premium gone up”. It is which part went up, and why — because a ten per cent increase made of three different movements is a very different conversation with your broker than a flat rate rise.
Rate movement — same sums insured, higher premium. This is the market, and it is the part a broker can negotiate.
Sums insured movement — you insured more, so you paid more. Usually intentional, occasionally an indexation nobody noticed.
Scope movement — locations or covers added or removed, which changes the total without changing the rate at all.
Excess movement — a higher excess for a lower premium is a transfer of risk to you, and it does not show in the premium comparison.
Fee and IPT movement — not the same as premium, and worth separating before anyone argues about a percentage.
Separating those five is a spreadsheet exercise that takes twenty minutes — once both years exist as rows in the same columns. Which is precisely why it usually does not get done: the twenty minutes sits behind two hours of retyping, and the retyping never gets scheduled.
The fourth line is the one most often missed. An excess that doubled is a real change in your exposure and appears nowhere in a premium comparison, which is why the excess column is worth extracting even when nobody asks for it.
Allocating premium across the business
Insurance is usually paid centrally and consumed locally, which makes it one of the standard recharge headaches — and one where the basis matters more than the arithmetic.
Sums insured is the most defensible basis for property cover, because it is stated on the schedule by the insurer rather than chosen by you. Headcount works for employer’s liability, revenue for public liability, and vehicle count for motor. What all of them require is the schedule broken into lines you can group.
Two practical points. Allocate the premium and the IPT separately, because they behave differently in the ledger and merging them means unpicking the total later. And keep the basis in a column next to the allocation: in six months, when somebody queries their charge, the answer should be visible rather than reconstructed.
For groups with a policy per entity, the same table serves both purposes — the per-entity rows for the statutory accounts, and the aggregate for the group. A hundred schedules in one pass turns a fortnight of renewal admin into an afternoon.
IPT is not VAT, and merging them causes real problems
Insurance premium tax looks like a tax line at the bottom of a document, and it gets treated like VAT by ledgers, spreadsheets and occasionally by people. It is not VAT.
The practical consequence is that it is generally not recoverable as input tax, so a schedule that folds IPT into a VAT column produces a reclaim that is wrong and a cost line that is understated. Neither error announces itself.
So IPT comes out as its own field wherever the schedule states it, alongside the net premium and any broker or administration fees. Three separate figures, because they land in three different places.
Where a schedule states only a gross total with no breakdown — which happens, particularly on smaller policies — the field stays empty rather than being derived by applying a rate. A derived figure that happens to be right is indistinguishable from one that is wrong, and this is not a number to guess at.
Sums insured drift, which nobody notices annually
A sum insured set five years ago and uplifted by an index each year is a number nobody has actually looked at since it was set. Whether it still reflects rebuilding cost or stock value is a question for your broker or a valuer — but noticing that it should be asked is a question of having the numbers side by side.
Once several years of schedules are rows in the same columns, the pattern becomes visible in a way it never is in a stack of PDFs: which locations have moved with the index and which have moved because someone made a decision, and which have not moved at all while the business behind them grew.
The same view catches the opposite error. A site closed eighteen months ago that still carries a sum insured is premium being paid for nothing, and it survives precisely because nobody reads a forty-page schedule line by line at renewal.
To be explicit about the boundary: this shows you what the schedules say and how they have changed. Whether a figure is adequate is advice, and that conversation belongs with your broker — who will have it far more usefully when you arrive with the comparison already made.
What this cannot tell you
Whether you are adequately covered
Reading a schedule tells you what it says, not whether the sums insured or covers are right for the business. That is advice.
What the policy actually covers
The schedule summarises; the wording governs. Exclusions and conditions live in a document this page does not attempt to interpret.
Whether a claim will be paid
That depends on the wording, the facts and the circumstances of the loss, in that order.
Whether a premium is competitive
A comparison needs a market view. What you get here is your own position, clearly, which is the input to that conversation.
The field-level detail of how a schedule is read — and what happens with ambiguous or missing values — is on policy field extraction.
What comes out
One row per schedule line, with the header fields repeated on each row so the table stands alone. Excel to work in, CSV to import, JSON for anything programmatic, XML where a system requires it.
Every row carries the source file and page. On a document that governs a claim, being able to get from a spreadsheet cell back to the page it came from is not a convenience — it is the difference between a working file and a transcription nobody trusts.
Amounts come back with the currency as stated, never converted. Where a programme spans countries, that is the only safe behaviour: the rate and the date are your decisions, made with your accountant.
Who tends to use this
Finance teams at renewal. The premium has to be allocated, accrued and prepaid, and the schedule is the source for all three.
Multi-site operators. Retail groups, care providers, franchises — anyone whose schedule runs to pages and whose sites change during the year.
Property and facilities managers. Keeping the insured list and the actual estate in step is a permanent job that is easy while the data is rows and impossible while it is PDFs.
Brokers preparing a market presentation. Client schedules from four insurers in four layouts, needed in one format, on a deadline.
Anyone mid-claim. When a loss happens, the excess and the sum insured for one specific location need to be found in seconds. That is the argument for having done this before anything went wrong — the rest of the sequence is in how to evidence a business interruption claim.
Six mistakes worth avoiding
Merging IPT into VAT. Produces a wrong reclaim and an understated cost, silently.
Dropping the excess column. It is the field you need at claim time and the one most often omitted, because it is written as prose rather than as a column.
Ignoring endorsements. The schedule at renewal is not the schedule at the loss if anything changed mid-term.
Retyping only the totals. A total cannot be allocated, compared or reconciled. Keep the lines.
Assuming the schedule stops where the table stops. Long schedules continue past page breaks, and a truncated one looks completely normal.
Keeping it only in the broker’s portal. Portals change, brokers change. An exported table is yours.
Where to start
Take the current schedule and last year’s, and convert both. Two files, two minutes, and you have the comparison that usually does not get made because it sits behind two hours of typing.
Then look at three things: locations present in one year and not the other, sums insured that did not move at all, and any excess that changed. Those three columns produce most of the questions worth taking to a broker.
If you have endorsements from during the year, convert those too and append them. A schedule plus its endorsements is the actual position; the schedule alone is the position as at renewal day.