Everyone models the order. Nobody models the run.
The figures below are the vendors’ own published prices and documentation, re-read on 5 September 2026. Every one of them links to the page it came from. Check them yourself; we cite nothing we cannot show.
What each one solves, and where it stops
Categories, not scorecards. Every product below does its own job better than we would.
Craft inventory tools
Solves well: cost of goods for a maker selling on marketplaces, with real material costing and tax-ready reports.
Stops at: it records what already happened. There is no live floor, and a run in progress is not a thing it has.
Cloud MRP
Solves well: planning and purchasing across products and orders, with a clean picture of what to make next.
Stops at: it assumes the recipe is followed. A batch cannot be split mid-run, and lot traceability is a paid add-on on top of the base plan.
Full ERP
Solves well: everything, eventually. If you have the implementation budget and eighteen months, it will fit.
Stops at: the shop with twelve people and no IT department, which is the shop this is for.
Paper and a whiteboard
Solves well: more than software vendors admit. It is free, it never goes down, and the supervisor already trusts it.
Stops at: the recall, the audit, and the question of why this month cost more than last.
scalebatch
Solves well: what happens between the store room and the finished product, with the stop, the scrap and the extra input recorded in the phase where they occurred.
Stops at: it does not ship. Four of the products above will run your shop this month and we will not.
The rows that decide it
Mechanism by mechanism
| Mechanism | Elsewhere | Here |
|---|---|---|
| Split a batch mid-run | Refused by design | The base model |
| Costing method | Moving or weighted average | Actual lots consumed |
| See the entry before it posts | Not available | Preview, then approve |
| Reason on every stop | Free text, or nothing | Seeded list, sortable |
| Lot traceability | Top tiers only | Every plan |
| Runs with no accounting attached | Needs the cloud ledger | Standalone or connected |
| Shipping today | Yes | No. In development |
The last row is the one that matters most right now, and it is why this page exists in this shape. Better you find that out here than three weeks into a trial.
Eleven questions, asked of four products
Three competitors and not five, because a six-column table is unreadable on a phone. Everything in their columns was read from their own public pages on 5 September 2026, and the numbered link says which page. Where a fact could not be verified today, the cell says that instead of guessing.
| Question | scalebatch | Craftybase / Stocksmith | Katana | MRPeasy |
|---|---|---|---|---|
| What happens during the run | ||||
| Split or merge a batch mid-run, with genealogy | Available Both are endpoints of the lot, and each side keeps its parent | Not documented 15 | “You cannot split one MO into multiple batches” 2 | Partial completion makes separate stock lots; splitting or merging a lot is not documented 10 |
| Scrap and extra material recorded in the phase where they happen | Available The scrap, the stop reason and the extra input belong to the phase, not to the close | Not documented 14 | Actual quantities are reported when a task is finished, in the Shop Floor App add-on 41 | A booking can be increased or returned to stock, and consumption can be reported in real time 9 |
| Material leaves the store room when the phase consumes it | Available The declared consumption writes the WIP balance before the lot closes | Not documented 14 | Stock movements are recorded only when the order is completed 3 | Booked material is consumed as production is reported, and automatically at the last operation 9 |
| Cost taken from the lots actually consumed | Available FIFO, FEFO or manual picking, frozen when the lot closes | Weighted average cost, by design 16 | Moving average cost only. FIFO and LIFO are not supported 5 | Actual cost of the stock lots, using FIFO or FEFO 11 |
| Hold, release and rework as events of their own | Partly available Hold, release, rework and the recall report are built. Certificates of analysis and dual signatures are not | Not documented 14 | Not listed on the pricing page 1 | Quality control from the Professional plan 8 |
| Software that watches the run and raises it to a person | Partly available The stall detector and its proposals run on a five-minute clock. The push to the screen and the email are not built | Not documented 14 | The vendor’s own AI page describes an intention: “Katana envisions a future where AI does more than automate” 6 | Not documented 8 |
| What it costs, and what the price includes | ||||
| Unlimited users, with no charge per seat | Available Yes. The meter is production orders per month | Yes. Priced by order lines per month 14 | Yes. Unlimited users on every tier 1 | No. Every plan is priced per user, $49 to $149 a month 8 |
| What lot traceability and quality add to the bill | Available Nothing. Both are in every plan | Traceability from the Indie plan at $99 a month. Quality is not documented 1415 | $249 a month for the Traceability add-on, on top of Core at $299. Quality is not listed 1 | Quality control needs Professional, $69 per user per month 8 |
| Where they are ahead of us today | ||||
| Ready-made connectors to a store and to accounting | On the way The channel and accounting providers read today and do not write back. The journal preview is built; the posting is not | QuickBooks, Xero, Shopify, Amazon, Etsy, WooCommerce, Square, Faire, Wix and Squarespace 17 | QuickBooks Online, Xero, Shopify, WooCommerce, BigCommerce, Amazon and eBay, with stock synced back to the store 7 | QuickBooks Online and Xero, journal entries included, plus Shopify, WooCommerce, BigCommerce and Amazon 13 |
| Scheduling that respects workstation capacity | Not planned Not on the plan. The board forecasts from the median phase time; it does not level a workstation | Not documented 14 | Not verifiable in a public source today | “Follow production schedules in dynamic calendars or Gantt charts”, with capacity and utilisation in real time 12 |
| You can buy it and run your shop on it this month | Not planned No. Early access, and nothing on this site takes money | Yes 14 | Yes 1 | Yes 8 |
Two rows are worth reading against us. MRPeasy costs from the lots actually consumed, so on costing method this is a tie and not a win; and it is the only one of the four that schedules against workstation capacity, which is not on our plan at all. The last row is still the one that decides this year.
- katanamrp.com/pricing/
- support.katanamrp.com/en/articles/5966899-assigning-batch-numbers-to-manufactured-products
- support.katanamrp.com/en/articles/5914331-completing-a-manufacturing-order-mo
- support.katanamrp.com/en/articles/5967154-tracking-the-actual-quantity-of-ingredients-used-in-manufacturing
- support.katanamrp.com/en/articles/5966125-understanding-moving-average-cost-mac
- katanamrp.com/artificial-intelligence/
- katanamrp.com/integrations/
- mrpeasy.com/pricing/
- mrpeasy.com/resources/user-manual/production-planning/manufacturing-orders/details/
- mrpeasy.com/resources/user-manual/stock/lots/
- mrpeasy.com/resources/user-manual/stock/items/cost/
- mrpeasy.com/production-planning-software/
- mrpeasy.com/resources/user-manual/integration/
- stocksmith.io/pricing
- help.stocksmith.io/article/1217-about-traceability
- help.craftybase.com/article/812-does-craftybase-use-lifo-fifo-or-weighted-average-cost
- stocksmith.io/integrations/
If your recipe is always followed, buy one of the others.
This exists for the shop where a run gets blended, split, held and reworked, and somebody has to explain the cost afterwards.