Loupe—the content design value framework
Loupe Beta The content design impact we never see
This is a demo run—explore it freely. Every step is filled in for an invented product, and every answer is yours to change. Move a weight or an answer and watch the scorecard follow.
Nothing you enter is stored. This is a front end only—no database, no backend, no account. Reloading the page clears everything, so export the PDF before you leave.

The state of content

Demo
What you know, step by step
{{ stepKicker }}

{{ stepTitle }}

{{ stepNote }}

✳ {{ stepEffect }}
The product Optional—it names the scorecard
{{ urlHint }}
Add one screen at a time below: its image, then its name. The files stay in your browser—nothing is sent anywhere, and nothing is read or analysed.
✳ The product comes first because everything later needs something to point at. It is not what Loupe reads first—the structural question in the next step sets the ceiling on what any screen-level number can mean.

Name at least one screen and select what do you want to weight it against. The scorecard answers your questions.

{{ sc.n }} {{ sc.peekLabel }}
What kind of work is this screen's content doing? Select one or more.
{{ k.label }}
{{ k.hint }}
No screens listed. That is a valid answer—the scorecard will judge the product as a whole instead of screen by screen.
How much arithmetic do you want to do?

You can change this per concern later. Symptoms alone are a legitimate answer—most first passes have no reliable figures to hand.

✳ {{ depthEffect }}
{{ g.title }}
{{ g.note }}
{{ g.total }}
%
Seen on
Can the people writing this content name the metric it moves?

And have they defended it to someone holding budget? This is the difference between a practice that can make the argument and one the argument is made for.

{{ literacyNote }}
{{ weightFooter }}
{{ p.name }}
{{ p.hint }}
{{ p.refLabel }}
How the score is worked out
Each practice scores 3 if it is in use beyond content, 2 if it exists but is mostly unused, 1 if you do something like it informally, and 0 if it is new to you. {{ maturityMath }}
Add another practice {{ extraNote }}
✳ {{ switchNote }}
{{ b.title }} {{ b.scope }}
I can answer with
Numbers
{{ f.label }}
{{ f.hint }}
{{ b.formula }}
Symptoms
{{ b.signal }}
On this page
  • {{ c.label }}
    •
What Loupe sees One reading of the whole run
{{ arch.kicker }}
{{ arch.name }}

{{ arch.reading }}

Why this one: {{ arch.rule }}
Does this describe you? {{ archVoteNote }}
What you told it Click any line to go back and change it
Where each concern sits

{{ shapeIntro }}

Priority you gave it
100%50%0%
A · Instrument firstMatters, can't yet prove it B · Measure nowMatters, evidence in hand C · WatchLow weight, no evidence D · Report as contextLow weight, evidence in hand {{ pt.n }}
NoneAll five
Evidence you can reach today
{{ pt.n }} {{ pt.label }} {{ pt.mode }}
Carries a money range Read from symptoms
{{ s.label }} {{ s.right }}
{{ s.note }}
What to do next, in this order
Each move is chosen by your answers. Do them in sequence—later ones assume the earlier ones happened.
{{ p.n }}
{{ p.title }}
{{ p.body }}
Why: {{ p.why }}
On this page
  • {{ c.label }}
    •
Who do you have to convince?

Three are ticked to start with—all seven are available. Pick the rooms you will actually walk into: {{ audienceAdvice }}

{{ c.title }}
Counts in {{ c.unit }}

{{ c.paste }}
{{ l.cue }} {{ l.text }}
The ask—{{ c.ask }}
Don't say—{{ c.dontSay }}
How the cards relate

The same events, counted in different units. These are alternative framings, not separate values—there is deliberately no total on this page.

{{ m.fact }} {{ m.who }} {{ m.units }}
The one sentence they can all sign
{{ agreedLine }}
{{ progress }}
{{ nextBlockedNote }}

What this will ask you

Demo
Two questions, {{ orientCount }} steps: whether your organisation can describe its content before writing it, and what specific screens are costing you. The scorecard below fills in as you answer.
✳ Loupe reads structure first, screens second. A team that cannot describe its content keeps paying the same costs after every fix, so the structural read sets what any screen-level number is allowed to claim.
{{ o.phase }}
{{ o.claim }}
{{ i.n }} {{ i.label }}—{{ i.asks }} {{ i.effort }}
{{ orientTime }}. {{ orientNote }} How Loupe reads content value

The scorecard

Product {{ productLabel }}
✳ Nothing below is anchored to a screen yet. Everything still works, but the rows describe your organisation rather than a place you can measure. Name the product
Content thinking maturity
{{ maturityNote }}
{{ maturityLevel }}
{{ maturityScore }}
{{ row.dim }} {{ row.priority }}
{{ row.measure }}
Evidence: {{ row.source }}
{{ p.t }}
{{ row.value }}
{{ row.confidence }}
Annual value in play
{{ totalNote }}
{{ total }}
{{ confidenceNote }}
Underlined figures are Loupe's assumptions, not measurements—starting points drawn from how these costs usually break down. Everything else is what you typed. If you know your own, change them and every row recalculates.
{{ g.label }}
✳ {{ k.term }}
{{ k.what }}
Use it to—{{ k.use }}
Your weights set priority and reading order only. They never multiply a money range—weighting a guess does not make it a measurement.
Take the scorecard with you
{{ exportNote }}
The download is an exact picture of the sheet. Print version opens your print dialog—choose Save as PDF there for a smaller file whose text can be selected, searched and quoted.
Two ways to stay in this
Share your Loupe feedback—help shape how we measure the value of content, including anything you wanted to tell the tool that it never asked you.
Subscribe to updates—occasional notes on this tool. No spamming or sales pitch.
Share the content love
Start over?

This clears everything from this session—your screens, weights, numbers, symptoms and answers. Nothing is stored anywhere, so it cannot be recovered.

Download the scorecard first if you want a record of this run.
Loupe The loupe to see the shape of work we do © 2026 Vinish Garg. All rights reserved.
Nothing you enter leaves this page.
Loupe is Vinish's idea of building a shape of our work, proudly designed with Claude Design.
Loupe
{{ printTitle }}
Content value scorecard
{{ printDate }}
{{ maturityHeading }}
{{ maturityScore }}
{{ totalHeading }}
{{ confidenceNote }}
{{ row.dim }} {{ row.priority }}
{{ row.measure }}
Evidence: {{ row.source }}. {{ row.formula }}
{{ row.value }}
{{ row.confidence }}
Measurement path
{{ p.n }} {{ p.title }}—{{ p.body }}
{{ c.title }} counts in {{ c.unit }}
{{ c.paste }}
The ask: {{ c.ask }}
What Loupe sees—{{ arch.kicker }}
{{ arch.name }}
{{ arch.reading }}
Why this one: {{ arch.rule }}
How to read this scorecard
Every row is one way content earns or costs money. Some rows carry a range; some deliberately do not. Read the confidence label before you quote the figure.
{{ g.term }}—{{ g.body }}
Example, a priced row. "Deflection—{{ egRange }}, Corroborated" means: from your own contact volume and cost, and your own estimate of how many contacts are content-caused, a fix in this range is defensible—and the symptoms you ticked independently point the same way. Say the range, then say the assumption. Never say the midpoint as though it were a forecast.
Example, an argued row. "Comprehension and trust—No price, Argued" means the tool refuses to invent a number here. Make this case with a five-user comprehension test people can watch, not with currency. A number you cannot defend costs you the rest of the scorecard.
The order of rows is the weighting you gave, not an assessment of importance. Lead with the top row unless the room you are in counts in different units—the audience notes above say which.
{{ printAssumptions }}
Loupe—the loupe to see the shape of work we do © {{ printYear }} {{ authorName }}. vinishgarg.com