ABOUT · LOEPWHAT THE INSTRUMENT DOES · AND WHAT IT REFUSES

The numbers are public. The instrument is not.

A county clerk recorded the deed. The company filed the 10-K. The Census published the rent distribution for that ZIP, and FEMA published the expected annual loss for that county. None of it is secret, and none of it is behind a password.

What costs money is the instrument: the desk that reads all of it at once, cross-checks one record against another and prints the result on a single sheet. What reaches everybody else is a headline that has already decided, an afternoon across a dozen tabs, or a box with a score in it that will not show its work.

Loep is that instrument, self-serve and monthly. The underwriting model and the filings reader here were built for a private investment firm and used on real deals, where a wrong number costs money and a missing number is a decision nobody can make. Porting them changed the surface and not the arithmetic: a map you click instead of a parcel number, a ticker box instead of a query, the same refusals underneath.

Nothing here is a feed to watch. Give it one address or one ticker, it surveys that one thing, and it stops.

THE SHEETSFOUR · AN ADDRESS, NOT A RANKING

Four sheets, and what each one hands back

A sheet takes one thing you already know — an address, a ticker, a chart — and returns everything that can be read about it, with the reading shown beside the result.

SHEET 01 · PROPERTY SCANNER: In any US address, typed or clicked as a roof on the map.
Out for-sale status and asking price, value and rent estimates with the comparable sales behind them, the county record, tax and assessment history, Census ACS gross rent by bedroom count, the county tax rate, energy rates, schools, expected natural-hazard loss, demand, supply, and where that market sits in its cycle.
Underneath the survey, a full underwriting model runs in the browser: a cost stack, loan sizing, DSCR, cap rate, cash-on-cash, IRR, stress tests, and a verdict with the thresholds it cleared printed beside it.
SHEET 02 · MARKET RESEARCH: In one ticker.
Out 10-K and 10-Q figures read from SEC EDGAR, the three statements, the segments, the fundamentals, a mechanical signal with its rule printed on the same screen as the word, insider and institutional positions, congressional disclosures, put/call, analysts’ estimates labeled as theirs, peers, and headwinds and tailwinds in three tiers that are never merged.
Beside the report, a pulse of posts and headlines for the tickers people are searching.
SHEET 03 · INDICATORS: In a TradingView username.
Out invite-only scripts you add to your own charts, so the reading happens where your charts already are.
TDPS is on the bench today, with its method written out and its thresholds withheld by design. Five more are in final testing and stay unnamed until they land.
SHEET 04 · BUSINESS ANALYST: Reading a private company the way sheet 02 reads a public one: its books, its market, its comparables. It does not exist yet. There is a waitlist, one email when it opens, and no screenshot of a thing that has not been built.

Open the scanner Read a report

WHERE THE FIGURES COME FROMPRIMARY RECORDS FIRST · PROVIDERS SECOND

Every figure arrives with its source attached

Primary records first, providers second, and the sheet says which is which. Filings come from SEC EDGAR, by form, accession number and date. Rent by bedroom count comes from the Census American Community Survey, with its margin of error. Expected annual loss from natural hazards comes from the FEMA National Risk Index. Tax rates and assessment histories come from county rolls, energy rates from the EIA. The listing record, prices, options, estimates and sentiment scores come from two licensed providers on a delayed clock, each named with what it supplies on the disclosures page.

A figure on a sheet is one of three things, and the sheet says which.

Retrieved

Taken from a record somebody else published: a parcel’s assessed taxes, a state’s energy rate, a company’s filed revenue. Accurate as of when it was read and no more current than its source, which is stamped beside it with a date.

Modeled

Loep’s own arithmetic on retrieved figures and assumptions: the energy a building of that size uses, a coverage ratio. A reasonable estimate, and an estimate.

Assumption

A value you set: vacancy, the loan terms, the exit year. The default is printed beside the input with the reason it is the default, and changing it changes the result.

HOW A FIGURE IS CHECKEDFOOTNOTES · STAMPS · A DELAYED CLOCK

A claim that cannot be walked back does not go on the sheet

Every sourced figure carries a mono footnote marker that resolves to the sources line at the foot of its own sheet: the accession number of the filing, the vintage of the census table, the release of the hazard layer, the date the county record was read. Checking is not a step at the end. It is the shape of the page.

An answer served from cache is stamped with when it was retrieved, not with now. A source that could not be read says so in a sentence, and never looks like a source that had nothing to report; those are two different failures and they are printed differently. A company with too little history to measure says how much it has.

Nothing on Loep is real time. The market half runs on a delayed clock and carries that stamp in the header of every sheet that touches it. A live quote would be a different instrument at a different price, and implying otherwise would be the cheapest lie available.

Read the disclosures

THE RULE, PRINTEDCOMPUTED FROM PUBLISHED PRICES ALONE

The rule is on the same screen as the answer

The market research sheet prints one word for a security: BUY, NEUTRAL or SELL. It cannot print that word without the rule that produced it, because the rule renders as part of the same component. Three conventional indicators are computed from daily closes: where the close sits against its 200-day simple moving average and whether the 50-day sits above or below it; a 14-day Wilder RSI read as mean reversion at 30 and 70; and the sign of the histogram of a 12/26/9 MACD. Each contributes −1, 0 or +1. They are added. A total of +2 or more reads BUY, −2 or less reads SELL, and everything between reads NEUTRAL, which is the reading about two thirds of the time.

The rule does not change between tickers, and it does not change because a reading is inconvenient. It is a textbook combination documented publicly since the 1970s, published in full precisely because it is nobody’s secret. Where there is too little price history to run it, the sheet says how much it has and how much the rule needs, and prints no signal at all. It knows nothing about the business under the price, so it reads a company whose fundamentals are deteriorating as BUY for as long as the price holds up — which is why the rule printed beside the signal says, in its own words, that it is mechanical, computed from published daily prices alone, and not Loep’s view of the security.

The same discipline governs the evidence on a report. Headwinds and tailwinds arrive in three tiers, under three headings, on three different bases, and they are never merged, netted or scored against one another.

01
Arithmetic on filed figures.
Revenue, margin, share count and cash flow, measured by Loep from what the company filed with the SEC, each claim carrying the accession number of the filing it came from.
02
The company’s own risk-factor headings.
Taken verbatim from the latest annual report and left unweighted. These are the company’s words about itself, not a reading of them.
03
A vendor’s sentiment score on the news.
Carried as the vendor classified it, which Loep neither verifies nor endorses.

THE REFUSALS10 REFUSALS · WHAT EACH ONE COSTS

What Loep will not do, and what each refusal costs

Most of the work went into the refusals. Each one makes a sheet look emptier, slower or less decisive than it could be, and each one is kept for the same reason: an instrument that guesses once cannot be trusted afterwards.

Read them before paying, because they are what the money buys.

It will not print a figure it did not read.Where a source is silent the sheet shows a gap and names what went quiet. Silence from a provider is not a finding, and unmeasured is not the same as steady. The cost is sheets with holes in them.
It will not net the three tiers of evidence into one score, because the weakest would otherwise borrow the credibility of the strongest.The cost is that there is no single number to quote.
It will not run the underwriting until the price is confirmed.The price a provider holds and the price on the listing in front of you are frequently different, and choosing between them is your judgment rather than the instrument’s. The cost is one more step before any result.
It will not hide an assumption.Every input sits on the screen with its default and the reason that default is what it is, including the defaults that flatter nothing: rent growth starts at zero on purpose. Two people underwriting the same address with different assumptions get different results, and both are correct arithmetic.
It will not publish a price target, a fair value or a forecast of its own for any security.Estimates on a report are other analysts’, labeled as theirs, with the count behind each figure and how many revised up against down. The cost is a report that ends without a number to act on.
It will not show performance of its own.No track record, no hit rate, no backtest offered as an expectation of returns.
It will not treat the written note as a source.The note under a survey or a report is written by a language model from the confirmed figures alone, labeled as such with the time it was written, and nothing in it is read back into a figure on the sheet. If the model stops before it finishes, no note appears; half a note is refused rather than shown.
It will not claim to be live.Delayed is printed, not implied, and the only thing on the site that claims freshness is a timestamp changing.
It will not bill quietly.At the limit the meter stops and says when the window resets, and nothing is charged for going over, because nothing can go over.
It will not tell you what to do.Loep does not know your holdings, your tax position, your horizon or your objectives, and every reader who opens the same address on the same day sees the same sheet. The cost is the one thing people most want from research.

THE OPERATORNAMED HERE AND ON THE LEGAL PAGES

Who operates it

Loep is operated by ADJL Capital LLC, a company in the United States. The operator is named here and on the four legal documents and nowhere else; everywhere else the product speaks as Loep, because the instrument is the thing being offered and a masthead is not evidence.

Loep is a publisher of research software. It manages no money, executes no trades, holds no assets for anyone and tailors nothing to a person’s circumstances. Nothing on it is offered as the service of an investment adviser or a broker, and reading it or paying for it creates no relationship of that kind.

The instruments were not designed for a website. They were built for a private investment firm’s own work and used on its own deals, and the underwriting engine, the filings pipeline and the provider clients came across with their tests rather than being rewritten for a launch. What is new is the public half: the map, the search field, the meter, the account, and a written account of what every figure is.

What is not on this page: a team photograph, a founding story, a row of logos belonging to firms that have never used it. None of that would tell you whether the revenue figure on sheet 02 is the one the company filed. The provenance line under the figure does.

Corrections, questions and indicator access go to laskowskidanny@gmail.com, the address on the legal pages. A correction that names the sheet and the figure gets a faster answer than one that does not.

WHAT IS SURVEYED · WHAT IS NOTLIVE · ON THE BENCH · NOT BUILT

Where the survey stands today

Stated plainly, because a product that hides its edges has already started lying about its middle.

Survey	Where it stands today
Property scanner	Live. The map, address search, the area scan for listings, the full property survey, and the underwriting model beneath it.
Market research	Live. The filings half reads SEC EDGAR directly and works without any paid feed. The market half and the pulse run on a delayed clock from a licensed provider, and when that provider cannot be reached the sheet collapses to one panel and says so rather than thinning out quietly. A report prints to PDF the way it renders.
Indicators	One on the bench. TDPS, with its method written out. Access is requested by TradingView username and granted by hand. Five more are in final testing. Names: —. Dates: —.
Business analyst	Not built. A waitlist, one email when it opens, and nothing else. Reserve a seat
Account, plan and meter	Live. Sign-up, sign-in, the usage meters, saved surveys, the watchlist and indicator requests.
Billing	The plan, the trial and the meter run today. Card checkout opens when the payment processor is connected; until then the pricing page and the account page say so in one line instead of pretending.
Real time	Not offered, and not planned. Every clock is printed beside the figure it belongs to.

HOW TO START7 DAYS ON LOEP PRO · NO CARD

Start with one address, or one ticker

Type an address and the scanner opens on the roof. Type a ticker and the report opens on the filings. Both work before a card exists: every new account starts with 7 days on Loep Pro without one, then lands on Explorer, the free plan, which keeps the scanner and the reports at 3 property surveys and 3 ticker reports a day for as long as you like.

Loep Pro is $150 a month. It carries the three live sheets, the indicators with TradingView access, saving, exporting and a watchlist, metered at 60 property surveys and 30 ticker reports an hour. The meter runs on a rolling window rather than a monthly allowance, so a long afternoon at the bench costs nothing extra, and at the limit it stops and says when it resets.

Cancel from the account page in two clicks. The month already paid for stays open.

Start surveying See the plan

Research, never advice.