What echoUX delivers, and why

What a finding is, what one delivery contains.

What you receive is a finding: one thing described in several separate pieces of feedback, written to a fixed structure, with the customers' own words that support it attached. On the product's own screens a finding is called a discovery artefact.

A topic is the area of your customers' experience you chose to cover. Each finding is delivered to every destination you chose for that topic: the Slack channel, Notion page, Linear team or webhook address you connected.

a delivered Problem statement in Linear · the title and the first section with its quotes under it. Enough to show the shape of one delivery, not the whole issue

The four types of finding

Every finding is one of four types, and each type has its own sections. Every finding of one type has the same sections in the same order, so a finding delivered this month can be compared with one delivered last year.

A Problem statement describes a difficulty someone has: who they are, how they feel, what the difficulty is about, what they need to do and why, and what stands in their way.

A Need statement describes something someone wants to be able to do: who they are, when the need arises, what they want to do, and what that lets them achieve.

A How might we describes an improvement to explore: who it is for, what could be better for that person, and what that lets that person achieve.

A Job to be done is a short statement of what someone is trying to do, in three parts: a Verb, the Object the verb acts on, and a Clarifier that limits it.

What one delivery contains

Each finding is its own delivery: one Linear issue, one Slack message, one Notion row, or one request to your webhook address.

A delivery opens with the finding's title, which says what the finding is. A How might we is titled with its question, for example How might we let a traveller move the dates of a stay they have already paid for, without calling anyone. A Job to be done is titled with its statement, for example Split a booking between two payers. A Problem statement or a Need statement is titled with the name of the situation it describes. In Slack the title has its type in front: JTBD: for a Job to be done, Problem statement: or Need statement:. A How might we has nothing in front, as its title already begins with How might we.

Then come the sections, and the quotes that support them. In Linear each quote sits under the heading of the section it supports, and a horizontal line follows the finding. Under the line come the facts about it, such as its theme, how many pieces of feedback back it, and its scores. In Slack the sections come first, one per line, followed by the Confidence and Does not cover lines. The facts come next in small grey text, then a dividing line, then a list headed Evidence, in which the quotes are grouped by section.

In Notion, the page of the row is the finding. Each section is a heading with its text under it, and a toggle under the text, such as 2 quotes, holds the quotes that support it. The toggle Grounding holds the Confidence note: how far the evidence reaches, what it does not cover, and any other reading that was considered and why it was set aside. The toggle Why these scores holds the scores. The fields of the row are the facts about the finding, such as 02 Theme, 04 Pieces of feedback and 09 Strength. Each name begins with a number, so Notion lists the fields in order. Article 8 lists them all.

Each quote is in the person's own words. It is marked first-person when the person describes their own experience, or third-person when the person describes someone else's, and it carries the source it came from and its date. No quote names the person who wrote it.

In Linear and in Slack, a line about the theme states how often it has been seen, for example Seen in 3 batches of feedback. In Notion, the field 16 Batches of feedback holds that number. A batch of feedback is one imported file, or all the feedback from your sources that had arrived and not yet been read when its reading began. A large batch can take several cycles to read. The number counts separate batches, so one file counts once however many cycles it takes to read.

A Confidence line states how far the evidence reaches, and a Does not cover line states what the evidence does not cover.

Under Why these scores, each measure has a score out of 100 and the reason for it.

A delivery ends with one link, Change this topic, which opens the topic so you can change its wording.

NoteWhen later feedback refines a finding, a new finding of the same type is delivered, carrying the line Refines an earlier finding with the date of the earlier delivery.

When findings arrive

A cycle is one run over up to 150 pieces of feedback, and starts when about twenty pieces have arrived. Feedback you import starts a cycle without waiting for twenty, and a large import is read in several cycles. An import's findings arrive together, once the whole file is read. The first findings usually take between several weeks and two months, depending on how much feedback arrives. To see findings sooner, import feedback you already have or connect another source.

Related articles

Last verified:

23 October 2026