Build it yourself, or don't
You can do this in Excel. The question is who maintains it in March.
Your CDS reports arrive as CSV files, so of course you can analyse them in Excel or Power BI. Plenty of businesses do, and for some of them it is the right answer. This page is the honest scope of that build: what the easy part is, what the hard part is, and the specific job that never finishes.
Visualisation is the easy part.
This is the thing that makes the build look smaller than it is. Getting a CDS report into a pivot table and charting duty by commodity code is an afternoon's work, and the result looks like the finished article. Power BI will do it faster and prettier.
But a chart of what you paid is not the same as knowing what you should have paid. Every useful finding in customs data comes from comparing your declaration against something external: the tariff measure that applied on that date, the preference you were entitled to claim, the quota that was open, the suspension that existed. Your CSV contains none of that. It contains what was declared.
So the real project is not a dashboard. It is an enrichment pipeline, and then the upkeep of it.
What the build actually involves.
Five jobs. The first is quick. The last one is the reason most in-house builds quietly stop being used.
1. Get the data in and joined up
Request the four report types from HMRC's Get customs data service, then stitch them together. Each report covers a maximum of 31 consecutive days, so a year is twelve requests per report type. Item, header and tax lines have to be joined on declaration reference before any of it means anything, and if you use several EORIs each one needs its own request under its own Government Gateway sign-in.
2. Join every line to the tariff that applied
This is the step people underestimate. To know whether a duty rate was right, you need the measure that was in force for that commodity code, that country of origin and that date. Not today's rate: the rate on the day of the declaration. That means holding tariff history, not a snapshot, and matching on a composite of code, origin, date, procedure and any document codes claimed.
3. Write the rules that find things
A dashboard shows you totals. Findings come from rules: this code carries a preference that was not claimed, this valuation looks inconsistent with the rest of the shipment, this origin and code combination is a classification the tariff has since changed, this entry paid the third-country rate where a quota was open. Each rule is a piece of customs knowledge encoded as logic. Writing them needs someone who knows what to look for, which is usually a different person from the one who knows Power BI.
4. Deal with the messy reality of declaration data
Commodity codes get restructured mid-year. Waivers appear against specific codes. Quotas open, fill and close. Document codes change. Your declarants make inconsistent entries that are not technically errors. Every one of these breaks a naive join, and each break produces either a false positive you waste time on or a real finding you never see.
5. Keep the tariff feed current
Here is the job that does not end. The UK tariff changes constantly, and HMRC issues corrections between releases through stop press notices. Our own public archive of those notices holds 747 records going back to 2018, and it grows every week. A model built against last quarter's tariff gives confidently wrong answers, which is worse than no answers, because nobody knows to distrust it.
The failure mode is not that the build does not work. It usually works, once. The failure mode is that it becomes a spreadsheet or a report only one person understands, that person moves on, and nobody else can say whether the numbers are still right. That is key-person risk on a compliance process, which is exactly where you do not want it.
When building it yourself is the right call.
We would rather say this plainly than pretend a subscription is always the answer.
You are doing this once. A single retrospective investigation, a specific reclaim you already suspect, or a due diligence exercise on an acquisition. Pull the reports, do the work, close the file. No tool needed.
Your volume is genuinely low. If you file a few dozen declarations a year across a handful of commodity codes, you can check them properly by hand and the errors are cheap to find.
You already carry the two skills, permanently. A data analyst and a customs specialist, both with capacity, both staying. If you have that, you can build something good. Some businesses have exactly this and should use it.
The question is not whether Excel can do it. It is whether you want to own a tariff-matching engine as an internal product, with the maintenance and the succession planning that implies.
Side by side.
The same job, done two ways.
| The job | Excel or Power BI | CAT360 |
|---|---|---|
| Getting your declaration data | You request it from HMRC, up to 31 days per report, per EORI. | Same reports. You upload them, or authorise us to retrieve them for you. |
| Tariff history to compare against | You source, load and maintain it. | Maintained by us, weekly, and published openly in the alerts archive so you can check it. |
| Finding the errors | Rules you write, based on knowledge you hold. | Checks written by practising customs consultants, run over every line. |
| Evidence for a claim or an audit | Whatever you build. Usually a filtered sheet. | Each finding cites the legal source and the declaration line it came from. |
| Several brokers or EORIs | Possible, and one of the fiddlier parts of the join. | Built in. |
| Who can run it next year | Whoever built it, if they are still there. | Anyone with a login. |
| Cost | Analyst and specialist time, ongoing. | From £499 a month. See pricing. |
Questions
Can I keep using Power BI alongside CAT360?
Yes, and some customers do. CAT360 is where the enrichment and the checks happen; if your business reports through Power BI, there is nothing stopping you pulling findings into it for internal reporting. The distinction is that the tariff matching does not live in your model any more.
Why can't I just download the tariff and join it myself?
You can, and that is exactly the build described above. The difficulty is not access, it is that you need the tariff as it stood on each declaration date rather than as it stands today, and you need to keep taking delivery of changes indefinitely. The join is the easy half of that.
We built something already and it works. Why change?
Then the question is narrower than whether to buy a tool: it is who maintains the tariff feed and the rules, and what happens when that person is on holiday during an HMRC audit. If you have a clear answer to both, keep it. Plenty of businesses do not, and only find out when they need the model to be right.
How long before CAT360 shows me anything?
As soon as your reports are loaded. The reports themselves are the wait: HMRC says a request can take up to 72 hours. The 14-day trial runs on your own data on the Enterprise tier, no card required.
Does CAT360 replace my filing software?
No. It reads what was filed after the goods cleared. Keep filing on whatever you use now. Post-clearance review is all we do.
Test it against your own spreadsheet.
Run CAT360 over the same declarations you already analysed and see what it finds that your model did not. That is a fair test, and it is the one we would want if we were buying. 14 days, your own data, no card required.