We intend to publish measured data on how small businesses actually run. We have not published any yet.
NoxOrigin sits on the operating records of a small number of Indian businesses: their quotes, invoices, payments, projects, and stock. In time that is a source of original data that nobody else has, about the unglamorous middle of running a company — how long money takes to arrive, how often scope moves, which reconciliation breaks first.
That data has not been collected. There is no index, no benchmark, and no finding on this programme. This page exists so the first real result lands somewhere honest, with its method attached, rather than arriving as a blog post with a number nobody can check.
What this programme wants to measure
These are the measures we intend to define, define properly, and only then collect. None has been collected yet. Each entry names the open question that has to be settled before the measure means anything, because a statistic is only as good as the decision about what to count.
Invoice payment delay
How long money actually takes to arrive after an invoice is issued, rather than after its due date. Open questions: which clock starts the measure, how credit terms are grouped, and how invoices that never arrive at all are treated.
Partial payments per invoice
How often an invoice is settled in more than one instalment, and how many. Open question: whether an instalment is defined by the payment record or by the invoice allocation, and how advance payments are handled.
Lead response distribution
The spread of time between an enquiry arriving and a first human reply, rather than a single average. Open question: which channels count as a response, and how out-of-hours enquiries are handled.
Quotation-to-invoice conversion
The share of issued quotations that become invoices, and how long that takes. Open question: what happens to a quote that is revised repeatedly, and whether expiry is a conversion or a loss.
Frequency of scope change
How often agreed work is changed after delivery starts, and how those changes are recorded. Open question: the difference between a formal change request and informal extra work, which is the harder half.
Reconciliation discrepancy
The most common reason a day-end or a bank reconciliation does not balance on the first pass. Open question: whether the measure counts reasons, events, or time spent resolving them.
Stockout frequency
How often an item the business expected to sell was unavailable, and for how long. Open question: how a stockout is distinguished from a deliberate decision not to reorder.
Project margin distribution
The spread of realised margin across delivered projects, rather than a single figure. Open question: how internal time and rework are valued, which moves the number more than anything else.
There is nothing to read here yet
This page is the programme’s front door, not its output. We are not publishing a provisional number, an illustrative example, or a range borrowed from elsewhere and presented as though we had measured it. If a figure appears on this site without its sample, its collection window, and its consent basis printed beside it, it is a mistake — and we would rather leave this page empty.
What we can tell you today is the gate. That is a real commitment, and it is checkable against anything we publish later.
The publication gate
A measure becomes publishable only when every condition below is satisfied and stated on the same page as the number. The conditions are written before any data exists, so that the standard cannot be quietly relaxed once a result is in hand and it is tempting.
The method is on the page, not in a footnote
Definition, population, exclusions, and collection method are readable next to the number. If the method cannot be printed in full, the result cannot be published.
The sample is stated
How many businesses, and how much of their activity, the figure covers. A median over an unstated number of records is not a result, it is an impression.
The collection window is stated
The period the data came from, and confirmation that it was long enough to cover the cycle the measure describes. A payment-delay figure collected over a few days is a season artefact.
The consent basis is stated
Which businesses agreed, on what terms, and under what permission. No permission, no figure — the consent basis is a condition of publication, not a footnote.
The disqualifiers are stated
What was excluded and why. A measure that quietly drops its weakest records is not a measurement.
The number is reproducible
A second person can take the underlying extract and arrive at the same figure. If only the person who collected it can compute it, it is an anecdote with decimals.
Why this page exists before there is data on it
It is worth being direct about the motivation, because a page with no findings on it can look like an attempt to borrow credibility that has not been earned.
It is discoverability, not findings
A research programme needs a permanent address. When the first finding clears the gate, it needs a home where the method is already published, rather than a last-minute page that has to retrofit a definition after the number exists.
It is entity clarity
There are a great many numbers on the internet about small business behaviour, most of them unattributed. Stating in advance what this programme will and will not do — and that it has not done it yet — is the only way to be recognisable later as the source rather than as one more site restating someone else’s figure.
It is not a ranking tactic
This page is not built to attract traffic, and we do not expect it to. A page that states plainly that it has no results is a poor landing page and a good commitment. If the only reason to keep it were traffic, the honest move would be to delete it.
No date is attached to it
We are not promising a publication date, because a date would be a guess dressed as a plan. What is committed to is the gate. When the gate is met, something is published; until then, this page stays as it is.
Useful while we wait
These are the parts of the site that answer a real operational question today, without waiting on a dataset.
Research methodology
How a measure is defined, where data may come from, what consent is required, and what disqualifies a result.
Business operations
The connected record these measures would describe: customer, opportunity, quote, project, work, invoice, payment.
Calculators and planners
Run the arithmetic on your own figures. The formula is on the page, so the result is yours to verify.
Operational guides
Practical writing on billing, collections, permissions, day-end close, and stock — argued from practice rather than from data.
Customer stories
How businesses run these workflows today. Every story is held until the customer approves its wording and its figures.
Privacy policy
What we collect, what we do with it, and the rights you have over it. Any research data collection has to sit inside this.