Your data, fit for purpose.
Before anything useful can be built on your data — a report, a dashboard, a submission — it has to be clean. CostCraft™ finds what's wrong, fixes what can be fixed, and documents what can't.
Three parts of every cleanup
Extraction
Getting your data out of the form it arrived in — PDFs, supplier files, lab sheets, legacy exports, inconsistent formats — and into something a system can work with.
Error identification
Finding what's wrong before it becomes a problem downstream. Duplicates, mismatched units, out-of-range values, broken references — surfaced and documented.
Verification
Confirming that what your data says matches what it should say. Cross-checking against source records, reference lists, or the requirements the data has to meet.
A dataset you can build on.
Every cleanup engagement ends with a documented record of what was found, what was changed, and what was left for your team to resolve. No silent fixes.
Before 8 rows
| # | SKU | Supplier | Item | Qty (as sent) |
|---|---|---|---|---|
| 1 | 1042 | Supplier A Ltd. | Hex bolt M8 x 40, zinc | 100 ea |
| 2 | SKU-1042 | SUPPLIER A LTD | HEX BOLT M8X40 ZN | 1 box/100 |
| 3 | SKU-2210 | Supplier B | nitrile gloves L | 12 dozen |
| 4 | sku_3305 | supplier b | Safety glasses - clear | 24 pcs |
| 5 | SKU-4410 | Supplier C Inc | Cable tie 200mm | 1,000 EA |
| 6 | SKU-5120 | Supplier C Inc. | Duct tape 48 mm | 2 cs |
| 7 | SKU-6?01 | Supplier D | Work light, LED | 6 ea |
| 8 | SKU-7002 | Supplier-A | Shop towels | 6 rolls |
After 5 rows
| SKU | Supplier | Item | Qty | Unit |
|---|---|---|---|---|
| SKU-1042 | Supplier A | Hex bolt, M8 × 40, zinc | 100 | ea |
| SKU-2210 | Supplier B | Nitrile gloves, large | 144 | ea |
| SKU-3305 | Supplier B | Safety glasses, clear | 24 | ea |
| SKU-4410 | Supplier C | Cable tie, 200 mm | 1000 | ea |
| SKU-7002 | Supplier A | Shop towels | 6 | roll |
Exceptions 3 rows
| Row | SKU | Status | Reason |
|---|---|---|---|
| 2 | SKU-1042 | Flagged | Duplicate of row 1 once the box of 100 was converted to 100 ea. Kept once. |
| 6 | SKU-5120 | Flagged | Quantity given in cases ("2 cs"), and the case size isn't on the item record. Not guessed. |
| 7 | SKU-6?01 | Flagged | SKU not in the item master. Not guessed. |
Rules first. Every exception documented.
Tablify™ is the extraction tool behind CostCraft's reporting systems, packaged for any workflow where messy inputs are costing you clean outputs.
- Deterministic rules first
- Every exception documented
- Nothing resolved silently
When the data is clean but the shape is wrong: transformation
Questions
How is this priced?
Hourly, with an estimate before we start. The free review is what makes that estimate worth anything, because we look at your actual data first. If the work is heading past the estimate, we stop and tell you before carrying on. The first you hear of it won't be the invoice.
What do you need from us to start?
Four things: a sample of the real data, not a tidied version; an example of what you have to produce; a conversation with whoever does it by hand today; and a signed agreement, in place before any data moves.
What if our data turns out to be worse than we thought?
We tell you before doing more work. You get what we found and a revised estimate, and you decide whether to carry on. Finding that out early is most of what the review is for.
What happens to our data afterwards?
It's deleted at the end of the engagement unless you ask us to keep it. If we're running your monthly reporting, it stays while that's running.
What happens to rows you can't fix?
They're listed, with the reason. We don't guess, and we don't quietly drop them.