|Kith Consulting
← All documentsPortfolio edition · fictional client
01What this is

The data under every page, and how to read it

Every figure on the other pages comes from the files listed here. They are plain tables you can open in Excel or Numbers. This page says what each one shows and which columns matter, so you can go further without help.

11
data files
10
companies in the connected dataset
22
simulated baseline values, kept apart
02Guide to the files

What each file shows

A CSV opens as a grid with technical column names. Open any card for what one row is, why you would want it, and what the main columns mean. The files hold the workings; the other pages are the readable view.

How to open one: download the file and open it in Excel or Numbers; turn on filters on the top row; sort by the column you care about. Dollar columns are whole dollars; percentages are shares of the company's EBITDA unless the name says otherwise.

Company factsconnected-dataset-v4.csv
One row isOne row per company (ten rows).
Download CSV

Start here when you want a company's size, margin, multiple, hold or footprint.

  • Columns worth knowing
  • priority_company: yes for the priority four
  • scope1_t_co2e / scope2_t_co2e: tonnes of CO2e from fuel burned on site / from purchased electricity
  • scope12_t_co2e_base: the footprint total the model uses
  • ltm_revenue_usd_m / ltm_adj_ebitda_usd_m: revenue and adjusted EBITDA for the twelve months to January 31, 2026, in $ millions
  • exit_multiple: the multiple of EBITDA a buyer is assumed to pay
  • remaining_hold_n_years: years from the report date to the modelled exit
  • ..._basis: where that value came from: portfolio spine, register or a company document
Where each company fact came fromconnected-dataset-lineage-v4.csv
One row isOne row per value in the file above.
Download CSV

Open it when you want to trace a number to its source.

  • Columns worth knowing
  • field / value / unit / period: what the value is and the period it covers
  • file / line: the data-room file and line it was read from
  • data_grade: A, B or C: see the grade key on the exposure model
  • basis / note: how it was obtained, and any caveat
Simulated water and site baselinesimulated-extended-data-v1.csv
One row isOne row per plant and parameter (four Harvest Ridge plants).
Download CSV

Open it to see the value, the range and how each simulated figure was anchored.

  • Columns worth knowing
  • parameter: what is estimated: water used, peak summer month, share used for washing, site area, paved share, distance to water
  • value: the single simulated value
  • central / range_low / range_high: the range it was drawn from
  • anchor: whether it is tied to a reported figure or is a judgment
  • label: always SIMULATED
Exposure linesexposure-lines-v6.csv
One row isOne row per company and risk (sixty-two rows: six per company, plus two extra water lines for Harvest Ridge).
Download CSV

The full workings behind the exposure model. Many columns; the ones that matter are listed.

  • Columns worth knowing
  • risk / status: which exposure, and whether it was quantified
  • impact_usd / impact_pct_ebitda: dollars lost in the year it happens, and as a share of EBITDA
  • likelihood_band / p_hold_midpoint: how likely, and the chance over the remaining hold
  • expected_loss_mid_usd: the average loss a year: impact times the yearly chance implied by the chance over the hold
  • grade: A, B or C
  • indicator_fact / note: the fact it rests on, in words
Exposure totals by companyexposure-summary-v6.csv
One row isOne row per company.
Download CSV

A quick way to compare companies by size of expected loss.

  • Columns worth knowing
  • lines_quantified: how many of its six lines have a dollar figure
  • ..._expected_loss_usd: expected loss a year by type of exposure
  • largest_line: the exposure that matters most there
Materiality matrix cellsd4-matrix-v5.csv
One row isOne row per company, risk and case.
Download CSV

Open it to see which cell each exposure sits in.

  • Columns worth knowing
  • case: base, or everything lost / half lost for the customer data requirement
  • likelihood_band / severity_band / priority: where it falls on the matrix
  • impact_usd / expected_loss_mid_usd: its size
Matrix totals by companyd4-summary-v5.csv
One row isOne row per company.
Download CSV

Company totals with the cost and saving of its measures beside them.

  • Columns worth knowing
  • expected_loss_mid_usd: total expected loss a year, everything lost
  • expected_loss_half_case_mid_usd: the same at half lost (an illustration)
  • measures_cost_usd / measures_saving_usd: cost and annual saving of its efficiency measures
  • lines_not_quantified: lines the data room could not size
Ranked initiativesd5-ranking-v6.csv
One row isOne row per initiative.
Download CSV

The ranked list behind the initiatives page.

  • Columns worth knowing
  • rank / status: its place in the company, or why it is not ranked
  • cost_low_usd / cost_high_usd: cost range
  • savings_low_usd / savings_high_usd: saving a year range
  • payback_years / payback_exceeds_n: years to pay back, and whether that is after the modelled exit
  • exit_value_effect_usd: annual saving times the exit multiple
  • risk_avoided_usd: expected loss avoided where the measure addresses a risk
Initiative sourcesd5-initiatives-v6.csv
One row isOne row per candidate initiative.
Download CSV

Open it to see where each cost and saving came from.

  • Columns worth knowing
  • source_kind: company plan, company measure or Kith estimate
  • derivation: how the cost and saving were worked out, in words
  • risk_addressed: which exposure it speaks to
Responding to customers' data requestsd5-value-protected-v6.csv
One row isOne row per company with customers asking.
Download CSV

The cost to respond beside the revenue it protects.

  • Columns worth knowing
  • revenue_type: earns today, or hope to win (revenue not yet won)
  • effort_days_* / cost_*_usd: days of work and cost, low to high (a Kith estimate)
  • value_protected_per_year_*: revenue protected a year if all, or half, of the exposed revenue were lost
  • payback_days_*: days to pay back against the half-lost value
  • components_in_base / not_covered: what the cost includes, and what it leaves out
Published figures behind the Kith estimatesrevision-benchmarks-v1.csv
One row isOne row per published figure used.
Download CSV

Open it to see each outside figure, where it came from, its date and the sentence it rests on.

  • Columns worth knowing
  • value / unit: the figure and what it measures
  • source / url: who published it and where
  • vintage / retrieved: the period it describes and the date it was read
  • quote: the sentence that supports it
  • use: how it is used, and its limits
03Connected dataset

Carbon and financials, side by side, for all ten

Revenue and EBITDA are for the twelve months to January 31, 2026. This is the readable view of the first file above.

CompanySectorLTM revenue, $MLTM adj. EBITDA, $MGross marginExit multipleHold left, yearsFootprint, t CO2eFootprint basis
Sterling Precision★Industrial Technology168.030.232.5%9.0x1.005,114Inferred
Harvest Ridge Foods★Food & Agriculture152.022.833.1%8.5x3.235,607Inferred
Meridian Business Solutions★Business Services225.038.330.0%9.5x1.001,722Reported
Clearview Environmental★Environmental Services195.031.238.8%9.5x1.233,102Inferred
Pinnacle TestingIndustrial Technology115.020.146.7%8.5x2.732,475Inferred
Ridgeline Property ServicesReal Estate Services95.017.441.8%8.5x3.23515Inferred
NovaCare Health PartnersHealthcare Services175.028.034.0%10.0x1.001,613Inferred
Trident IT SolutionsTechnology Services142.024.942.0%10.5x1.00701Inferred
Keystone Behavioral HealthHealthcare Services138.024.251.3%10.0x2.73929Reported
Patriot StaffingBusiness Services310.023.819.8%8.5x2.481,661Reported
04Simulated baseline

A simulated water and site baseline for Harvest Ridge Foods, and where it is used

Harvest Ridge's four plants have no metered water or site data in the data room beyond a total, so this baseline is simulated: values drawn within stated ranges around what the company did report. Every value is labelled simulated.

How it was built

Total water was fixed at the 28 million gallons the company reported, with Bowling Green at 10 million. The other plants split the rest by headcount, varied by plus or minus 25%; site values vary by plus or minus 50%. The ranges and the random seed were written down before any value was generated, so a re-run gives the same numbers.

Where it is used

In one line of the exposure model, and nowhere else. Drought curtailment is the line that uses it; the water price line beside it uses only the company's reported 28 million gallons at a benchmark price. The drought line uses Bowling Green's share of the water (10 of the reported 28 million gallons) and its simulated peak-month use to size what a month of restriction would cost. It is grade C, marked †, and tagged as using simulated values wherever it appears. By rule no figure may normally rest on simulated data; this is a specific exception made by the project owner for this one line.

What would make it firm

Meter readings for the four plants, and a survey of the Bowling Green site. With those, the simulated values drop out and the drought line becomes measured. The totals do not move much on the water price line (it is small); the drought line is the one that matters, and it rests most on the plant split.

PlantWhat it isSimulated valueUnitRangeAnchored on
NashvilleWater used in a year7,398,391gallons a year5,760,776.0 to 9,601,293.0judgment
MurfreesboroWater used in a year4,878,281gallons a year3,607,759.0 to 6,012,931.0judgment
Bowling GreenWater used in a year10,000,000gallons a year10,000,000.0 to 10,000,000.0reported
ChattanoogaWater used in a year5,723,328gallons a year4,131,466.0 to 6,885,776.0judgment
all plantsHow much busier a summer month is than the average month1.0ratio of summer-month use to the annual monthly average1.0 to 1.6judgment
Bowling GreenShare of the plant's water used for clean-in-place washing0.5share of plant water used for clean-in-place0.5 to 0.8judgment
NashvilleWater used in a peak summer month631,589gallons a month, derived from the two values above616,533.0 to 963,332.0judgment
MurfreesboroWater used in a peak summer month416,451gallons a month, derived from the two values above406,523.0 to 635,193.0judgment
Bowling GreenWater used in a peak summer month853,684gallons a month, derived from the two values above833,333.0 to 1,302,083.0judgment
ChattanoogaWater used in a peak summer month488,591gallons a month, derived from the two values above476,944.0 to 745,225.0judgment
NashvilleSite area23.3acres10.0 to 30.0judgment
NashvilleShare of the site that is paved or roofed72.5percent of site35.0 to 100.0judgment
NashvilleDistance to the nearest surface water1,059metres400.0 to 1,200.0judgment
MurfreesboroSite area25.8acres10.0 to 30.0judgment
MurfreesboroShare of the site that is paved or roofed51.2percent of site35.0 to 100.0judgment
MurfreesboroDistance to the nearest surface water769metres400.0 to 1,200.0judgment
Bowling GreenSite area23.6acres10.0 to 30.0judgment
Bowling GreenShare of the site that is paved or roofed90.3percent of site35.0 to 100.0judgment
Bowling GreenDistance to the nearest surface water303metres200.0 to 600.0judgment
ChattanoogaSite area16.8acres10.0 to 30.0judgment
ChattanoogaShare of the site that is paved or roofed99.7percent of site35.0 to 100.0judgment
ChattanoogaDistance to the nearest surface water830metres400.0 to 1,200.0judgment
05Basis

Notes

Notes
  • The files are the same ones every page is built from, copied unchanged.
  • Footprint figures are the signed inventory from the earlier carbon work, reported or inferred by company.
  • Estimates built from published benchmarks are listed with their sources in the benchmarks file above.
  • Simulated values carry their seed and the decision that fixed their ranges before they were generated, so a re-run reproduces them.
  • Where a company fact is a Kith estimate it carries grade C in the lineage file.