Skip to content

The product

What Ettore actually does.

Not a list of benefits. This is the work it performs on a request, in the order it performs it — and where a person still decides.

The actions

Each of these is a step the system takes on its own. What differs between customers is how many of them run without a review.

  1. 01

    Watches your shared inboxes

    Connected mailboxes are polled continuously — Outlook and Gmail — so nothing waits for someone to open the tab.

  2. 02

    Separates real requests from the rest

    Every message is classified: a request for quote, an order, a supplier offer, or noise. Only the first two enter the queue.

  3. 03

    Reads what came attached

    Bodies, PDFs, spreadsheets, images and archives. A photographed list and a formatted purchase order both end up as structured lines.

  4. 04

    Identifies the customer

    The sender is resolved to a customer record in your ERP, so everything downstream uses their terms rather than defaults.

  5. 05

    Matches every line to your catalog

    Free text, a competitor’s code, a discontinued reference. Each line resolves to a real SKU in your ERP — or is honestly marked as not resolvable.

  6. 06

    Applies that customer’s price

    Their pricelist, their discount structure. The draft carries the price they actually pay, not the list price someone then corrects.

  7. 07

    Checks stock at the moment of drafting

    Availability is read live, so a rep is not promising something that left the warehouse last week.

  8. 08

    Drafts the quote

    Priced, structured and complete, with anything uncertain already flagged rather than buried.

  9. 09

    Enters the order

    The same path in reverse for order entry: prose in, structured lines out, nothing rekeyed.

  10. 10

    Writes approved work back to the ERP

    Once approved, the quote lands in your ERP as a record. Nobody retypes it.

  11. 11

    Flags what needs a person

    Price below floor, stock short, an unrecognised item. The exceptions reach someone instead of being discovered later.

  12. 12

    Answers questions about your products

    Internal team or customer: technical questions get an answer in seconds, from your own catalog data.

How much of that runs without you

Autopilot

For request types you have chosen. A clean match against your catalog and the customer’s pricelist goes out as-is.

Copilot

Ettore does all the work and stops before sending. A person reviews the draft and approves what goes out. This is the default.

Human

Ettore organises the request and shows what it found. Your rep writes the reply.

What stays yours

Your ERP remains the source of truth

Products, customers and pricing live where they already live. Ettore reads them and writes back quotes — it never becomes the system of record.

Your mailbox is read-only

It watches the inbox that receives requests. It does not send from your account unless you configure that deliberately.

Nothing uncertain goes out alone

An unrecognised line never defaults to Ready. Anything the system is not sure about routes to a person, whatever mode you are in.

Your data stays yours

Tenant isolation is not a setting. One customer’s catalog, customers and pricing are never visible to another.

How it gets things right

Three mechanisms a buyer is entitled to ask about, described as they actually work. None of them is a number we can put on a page — they are disciplines, and the discipline is the reason to trust the output.

Quality

It is graded against quotes you already sent

The test set is not synthetic. It is real requests from your own inbox, scored against the proposal your team actually sent back — the only ground truth that counts, because it is what a customer accepted.

  • Every case is re-run rather than remembered. Verdicts are recomputed against the live pipeline each time, never read back from a stored result.
  • Runs happen with the learned layers deliberately emptied, so a score cannot be flattered by an answer the system was handed earlier.
  • A change to matching is measured against the whole corpus before it ships. One that does not beat what is already there does not go out.
  • Cost per request is measured in the same run as accuracy, so a more accurate pipeline that costs ten times more is a visible trade rather than a surprise on an invoice.

Completeness

It proposes the parts that go with the part

A luminaire needs a driver. A run of profile needs connectors and end caps. Customers rarely list them, and a quote that omits them comes back — or worse, ships incomplete. Ettore derives them from the rules of your catalog and computes the quantity.

  • Each proposal lands as one of four things: a required line that blocks delivery until it is resolved, an optional line the rep decides on, a flag that warns without adding anything, or a plain statement that nothing needs adding and why.
  • Quantities are computed rather than guessed — a formula over the line’s own facts, not a rule of thumb.
  • When a fact it needs is missing, it abstains and says so. Unknown is never treated as zero.
  • The rules are written against your catalog. Where you publish a price table carrying dependency rules your ERP does not hold, that is where they come from.

Memory

It becomes where the answers live

The knowledge that currently sits in one person’s head — which competitor code maps to which of your products, which spec is right when two sources disagree — accumulates in one place instead. That is the part that compounds.

  • A confirmed match is remembered. Next time that reference arrives, from any customer, it resolves immediately instead of going through matching again.
  • Equivalences mined from your own history arrive as suggestions, never as facts. Nothing learned changes an answer until a person activates it.
  • Where two sources disagree about a spec, both are kept side by side and the conflict is queued for someone to settle — rather than one silently overwriting the other.
  • Every correction a rep makes is an entry. The work of fixing a match is never thrown away.

No accuracy figure appears on this page. Any single number would need so much context to be honest that it would stop being informative — ask us for the eval report against your own requests instead.

How it is bought

The commercial shape, without the numbers — those depend on your volume and belong in a conversation, not on a page that cannot see your inbox.

You pay for volume, not seats

A monthly investment that tracks the number of requests handled. Putting more of your team on it costs nothing extra, so nobody has to ration access to the tool that does their work.

The cost follows the work, not the headcount

A quiet month costs less than a busy one. That is the right way round: the expense moves with the thing it is supposed to be replacing.

It starts with one mailbox

You do not buy the whole back office on day one. One inbox, real requests, and a decision about the second one afterwards.

Nothing is written back until you say so

Write-back to the ERP is switched on when your team trusts the drafts. Until then it costs you review time and nothing else.

We do not publish figures. Bring your request volume and we will put a number against it.

Questions about the product

Do we have to change how our customers write to us?

No, and that is rather the point. Requests keep arriving as email, in whatever shape your customers already send them.

How long before it is useful?

It drafts from real requests as soon as the mailbox and catalog are connected. The first weeks are mostly your team correcting matches, which is also how it learns your catalog.

What if it gets a line wrong?

A rep changes it, and that correction feeds back. Wrong matches are visible in the draft rather than discovered by a customer.

Can we start with one mailbox?

Yes. Most do. Adding the rest is configuration, not another project.

See it run on a real request.

Bring one from your own inbox — the messier the better. That shows you more than any list of actions can.