No paid placement, ever
No vendor can buy a position, a badge, a category, or a mention. There is no tier of listing, no sponsored row, and no arrangement where paying changes what a reader sees. This is permanent, not a launch-phase policy.
Editorial policy
We sell an ERP platform and implementation services, and we also publish comparisons and a vendor directory. That is three conflicts stacked on top of each other. This page states them, and states the rules we hold ourselves to — because a ranking whose method is private is an advertisement.
Vendors and readers can both file a correction. Tell us what is wrong and we will check it.
Most methodology pages open with rigour and mention commercial relationships somewhere near the bottom. We think that ordering is itself a tell, so here is the disclosure before anything else.
That third and fourth point are what make the first two survivable. We benefit when you buy software — sometimes ours, often someone else’s, with us implementing it — and we do not benefit differently depending on which vendor you choose. The incentive is toward a working recommendation rather than a particular logo.
Six criteria, fixed weights, applied identically to every product in a category including ours. The weights are published here before any product is scored, and they do not change between products or after results are seen.
The rules
No vendor can buy a position, a badge, a category, or a mention. There is no tier of listing, no sponsored row, and no arrangement where paying changes what a reader sees. This is permanent, not a launch-phase policy.
Every page that ranks or compares products carries an inline notice that we sell a competing product and implementation services — at the top, in the reading flow, not in a footer nobody scrolls to.
The six criteria and their weights are fixed for a category and published. We do not tune weights after seeing where products land, which is the most common way a rigged ranking looks legitimate.
Any vendor may dispute a factual claim. We correct errors, date the correction, and say what changed. We do not remove unfavourable assessments because a vendor objects to the conclusion.
Scored on the same criteria, erp.io does not lead most categories and we do not present it as though it does. It scores well on time to value, implementation risk, and extensibility. It scores poorly on support and ecosystem, because our partner network is small and the talent pool for our platform is a fraction of NetSuite’s. It does not appear at all in manufacturing or warehouse management, because we do not build those.
We publish at least one piece of research a year that is unflattering to our own product. Not as a gesture — because a directory that has never said anything bad about its owner has told you everything you need to know about its method.
Software changes continuously and any comparison starts decaying the day it is published. Every ranked and compared page carries a last-reviewed date and the name of the reviewer. Pages are re-reviewed quarterly at minimum.
Anyone can file a correction — vendors, customers, or readers. If the claim is factual and wrong, we fix it, date the fix, and note what changed at the foot of the page. We do not silently edit. What we will not do is remove an assessment because a vendor dislikes the conclusion; a disputed judgement gets the vendor’s position published alongside ours rather than replacing it.
Two honest limitations. Our coverage of non-US statutory and localisation requirements is thin, because our customer base is US-centric and we would rather say so than score a dimension we cannot assess properly. And our sample of live implementations is smaller than that of an established analyst firm, so failure-pattern claims come from a few dozen engagements rather than thousands. Both are stated on the pages where they matter.
The directory is published as an open dataset under a permissive licence, including the scores and the weights. You can download it, check our arithmetic, or use it in your own analysis. A ranking you cannot audit is a claim, not a finding.
Questions
The full dataset is downloadable — scores, weights, and review dates for every product we rank, including ours.