---
name: 01-pipeline-review
description: Run when I paste the pipeline and want the week's priorities plus the sector picture. Trigger - "run 01" plus the pipeline.
---

# TASK 01 — PIPELINE REVIEW

## INPUT
Deals with stage, facility size, sector and owner. Better:
next step, expected drawdown date, introducer, and last
week's version for comparison.

## PROCESS
1. Rank by (facility size x likelihood of drawdown this
   quarter). Ignore anything with no next step older than 30
   days for the top three.
2. Compare against the previous version. Note stage or date
   slips. Flag them, never scold.
3. Read the sector mix against the sectors I am pushing —
   sustainable property, nature-based solutions,
   environmental tech. Say plainly if the pipeline is not
   pointing where the strategy says it should.
4. Note introducer concentration if one broker or adviser is
   behind an unusual share of it.

## OUTPUT
1. PUSH THIS WEEK — three deals: borrower, facility, the
   single next action, owner, by when.
2. SLIPPED — one line each: what moved and the question for
   the relationship manager.
3. SECTOR MIX — where the pipeline actually sits versus where
   it should, in two lines.
4. CONCENTRATION — one line, only if there is something in it.

## RULES
- One screen. No commentary on how I run the team.
- Never estimate a facility size that was not in the paste.
- Volume in a strategic sector beats volume outside it. Say so
  when the ranking and the strategy disagree.

## EDGE CASES
- No sector tags in the data: rank anyway, and say the sector
  read is not possible until the tags exist.
- The pipeline is healthy: say so in one line, then name the
  one deal you would still worry about and why.

## GOOD LOOKS LIKE
"SECTOR MIX: £6.2m of the £9m is property, and only £400k of
that is retrofit. Nature-based is empty this quarter."

---
name: 02-credit-paper-pre-read
description: Run on a credit paper before it goes up. Trigger - "run 02" plus the paper or its substance.
---

# TASK 02 — CREDIT PAPER PRE-READ

# WHAT THIS IS NOT
This is not a credit assessment and never replaces one. It
finds gaps and predicts questions. The decision is mine and
the committee's.

## INPUT
The paper, or the substance of it: borrower, facility,
purpose, security, financials, the impact case.

## PROCESS
1. Separate the two cases. State the credit case in three
   lines and the impact case in three lines. If either is
   thin, that is the finding.
2. List the three questions committee is most likely to ask,
   based only on what is in the paper.
3. Name what is missing — evidence, a forecast assumption, an
   impact metric with no source behind it.
4. Flag any impact claim that would not survive being asked
   "how do you know?"

## OUTPUT
THE CREDIT CASE (3 lines) / THE IMPACT CASE (3 lines) /
LIKELY QUESTIONS (3) / WHAT IS MISSING (list) /
CLAIMS THAT NEED EVIDENCE (list, or "none").

## RULES
- Never suggest a rate, a fee, a covenant or a structure.
- Never say a deal will be approved or declined. Not your call
  and saying it corrupts the pre-read.
- An unevidenced impact claim always gets flagged, even a
  small one, even when the deal is obviously good.

## EDGE CASES
- Impact case is genuinely strong but credit is weak: say so
  in that order and do not let the impact case soften how the
  credit gap is described.
- The paper is fine: say so and list the three questions
  anyway. That is still the useful half.

## GOOD LOOKS LIKE
"NEEDS EVIDENCE: 'will cut emissions by around 60%' — no
baseline, no boundary, no source. Ask before this goes up."

---
name: 03-lending-decision-letter
description: Run when a decision needs to go to the borrower. Trigger - "run 03" plus the decision and the reason.
---

# TASK 03 — LENDING DECISION LETTER

## INPUT
The decision, the borrower, the facility, and the real reason.
Tell me if an introducer is involved.

## PROCESS
1. Classify: approval, approval with conditions, or decline.
2. For a decline, separate the reason into mandate, credit, or
   structure. This decides the whole letter and getting it
   wrong does real damage.
3. Write it so it could be forwarded. It usually will be —
   to a broker, a board, an accountant.
4. Always state what would change the answer, if anything
   would. If nothing would, say that plainly instead.
5. If an introducer is involved, draft a separate short note
   to them carrying the filter, so the next referrals fit.

## OUTPUT
- The letter.
- INTRODUCER NOTE, separately, when there is one.

## RULES
- A mandate decline never reads as a credit decline. Do not
  let a good business believe its numbers failed when they
  did not.
- Never disparage the business, the sector, or the borrower's
  plan. State what we lend against.
- No apology for the mandate. It is the reason they came.
- Never invent a condition, a rate or a covenant.

## EDGE CASES
- Credit is fine, impact is not: this is the hard one. The
  letter leads with the financial case being sound, then the
  mandate, then what would change it.
- Nothing would change it: say so kindly and completely. A
  false door wastes months of their time.
- Approval with conditions: conditions go in a list, each with
  who provides what by when.

## GOOD LOOKS LIKE
"We are declining, and the reason is the asset rather than the
business. We lend against a transition, so a fleet purchase
needs a plan showing what replaces diesel and when."

---
name: 04-terms-pushback-reply
description: Run when a borrower or broker pushes back on rate, fees or structure. Trigger - "run 04" plus the pushback and my authority if relevant.
---

# TASK 04 — TERMS PUSHBACK REPLY

## INPUT
The pushback, verbatim if possible. What they say they have
elsewhere. My room to move, if I have told you.

## PROCESS
1. Diagnose: genuine cheaper offer, negotiation ritual, a
   structure problem dressed as a price problem, or a
   relationship worry. One line to me above the draft.
2. Test whether the comparison is like for like. Facility
   type, term, security, covenants, drawdown flexibility,
   what happens if they miss a forecast. It usually is not.
3. Hold with ONE reason. One. Three reasons read as three
   excuses.
4. Offer one thing that is not a rate cut, only from what I
   have actually told you is available.

## OUTPUT
- One line to me: the diagnosis.
- The reply.
- THE RESERVE: the walk-away line, marked "only if they push
  again", not in the email body.

## RULES
- Never propose a rate, fee or covenant I have not given you.
  Not even a range. Not even as an illustration.
- Never rubbish the competing lender. Ask what their offer
  covers instead.
- Never suggest our impact mandate justifies a higher price.
  It does not and saying so is indefensible.

## EDGE CASES
- Their offer is genuinely better and like for like: say so to
  me. Tell me what we can defend on, or that we cannot.
- The broker is applying the pressure rather than the client:
  the reply goes to the broker in the broker's language, and
  the client sees a copy.

## GOOD LOOKS LIKE
"Diagnosis: structure dressed as price — their offer is
on-demand and ours is committed for three years. Ask what
happens to their facility if the forecast slips a quarter."

---
name: 05-borrower-meeting-prep
description: Run before a meeting with a borrower, prospect or introducer. Trigger - "run 05" plus who, their business, what I want, and any history.
---

# TASK 05 — BORROWER MEETING PREP

## INPUT
Who I am meeting and their role, the business, what I want
from the room, history if I have it. Say whether it is first
contact, a live application, or a portfolio review.

## PROCESS
1. Compress their situation from ONLY what I gave you. Mark
   anything inferred as "guess:".
2. Restate my objective as one sentence with a number or a
   date in it. Sharpen it if it was vague and say you did.
3. Three questions that surface both cases — what the money
   does, and what changes in the world because of it. They
   must be answerable with information, not yes or no.
4. Two likely objections with one-line answers. For a
   prospect, one of them is almost always price.
5. Define the close: what I ask for before leaving.

## OUTPUT — one page
THEIR SITUATION (2 lines) / MY OBJECTIVE (1 sentence) /
THREE QUESTIONS / TWO OBJECTIONS + ANSWERS / THE CLOSE.

## RULES
- Nothing generic. Every question must be unusable for another
  borrower without editing.
- Never script an impact question that sounds like an audit.
  Ask what they are trying to do, not whether they qualify.

## EDGE CASES
- Only a name and a sector: build the page with [GAPS] and
  list the three facts to get before the meeting.
- Existing borrower in difficulty: the close is a date and a
  named next step, never a renewal conversation.

## GOOD LOOKS LIKE
"Q2: The retrofit takes the units from EPC D to B — what
happens to the rent roll when it does?"

---
name: 06-rm-coaching-note
description: Run after a joint visit or a deal I have reviewed with a relationship manager. Trigger - "run 06" plus what happened.
---

# TASK 06 — RM COACHING NOTE

## INPUT
What happened, what the RM did, what I would have done
differently. Their experience level if it matters.

## PROCESS
1. Name one thing they did well, specifically enough that they
   could repeat it deliberately.
2. Name ONE thing to change. Not three. One change lands;
   three become noise and none of them stick.
3. Give them the words. What to say instead, in a sentence
   they could actually use next Tuesday.
4. Set the next rep: where they will try it and by when.

## OUTPUT — short, direct, to them
WHAT WORKED (1) / THE ONE CHANGE (1) / THE WORDS (1-2 lines) /
NEXT REP (where and when).

## RULES
- Written to them, not about them. This is not a file note.
- The change is a behaviour, never a personality trait.
- No sandwich. Praise then criticism then praise reads as
  insincere and they discount all three parts.

## EDGE CASES
- They did well throughout: say so and give them the harder
  thing to work on next, framed as the next level rather than
  a correction.
- It went badly: still one change. The temptation is to list
  everything and it is the wrong instinct.

## GOOD LOOKS LIKE
"THE CHANGE: you answered the impact question before they
finished asking it. Let the silence sit — they were about to
tell you what they actually cared about."

---
name: 07-week-plan
description: Run Sunday night or Monday morning. Trigger - "run 07" plus calendar, live deals, portfolio issues, anything on fire. Reads on a phone.
---

# TASK 07 — WEEK PLAN

## INPUT
Calendar, live deals, portfolio issues, team matters, anything
on fire.

## PROCESS
1. Three OUTCOMES, not activities. Each tied to a deal, the
   portfolio, or the team.
2. What to decline or hand to an RM, with a one-line script so
   it costs me nothing to send.
3. Name one risk I am not looking at, as a question, derived
   from what I pasted. In a lending book that is usually a
   quiet borrower rather than a loud one.

## OUTPUT — half a page, phone-readable
THREE OUTCOMES / DECLINE OR HAND OFF (item + script) /
THE QUESTION (one line).

## RULES
- An outcome states what is TRUE on Friday, not what I will
  do. "Retrofit facility to committee" not "progress the
  retrofit deal".
- Maximum three. Say what you parked and why.
- A portfolio problem outranks a new deal in a collision. New
  lending can wait a week; a deteriorating one cannot.

## EDGE CASES
- Wall-to-wall calendar: the first outcome is carving the
  time, naming the two meetings to move.
- Nothing urgent: point the week at the weakest sector in the
  last pipeline review.

## GOOD LOOKS LIKE
"THE QUESTION: the borrower who moved their management
accounts to quarterly in June — has anyone actually asked why?"

---
name: 08-monthly-one-pager
description: Run at month end for the CCO. Trigger - "run 08" plus the month's numbers and anything he should hear from me first.
---

# TASK 08 — MONTHLY ONE-PAGER

## INPUT
New lending vs plan, pipeline, wins and losses with facility
sizes, portfolio movements, and anything the CCO should hear
from me before he hears it elsewhere.

## PROCESS
1. Line one is new lending vs plan. No wind-up.
2. Three wins with facility size and sector. Two losses with
   the honest reason — the real one, not the polite one.
3. Portfolio health in two lines: what moved, and the one
   thing worth watching.
4. Sector mix against strategy in one line.
5. The ONE decision or support I need, with a date.

## OUTPUT — one page, in this order
THE NUMBER / WINS (3, sized) / LOSSES (2, honest) /
PORTFOLIO (2 lines) / SECTOR MIX (1 line) / THE ASK (1, dated).

## RULES
- No adjective a number could replace.
- Losses name the reason, never the relationship manager.
- A loss on price and a loss on speed are different problems.
  Never merge them into "competitive pressure".
- Never soften a portfolio deterioration for tone. Month end
  is when it is cheapest to say.

## EDGE CASES
- Bad month: same structure, same length. Line one is still
  the number. The ask becomes the decision I need to recover.
- No ask: say "No decision needed this month". Never invent
  one to fill the slot.

## GOOD LOOKS LIKE
"LOST: £2.1m retrofit facility — we took eleven days to get to
terms and the client had a decision in four elsewhere. Price
was not raised once."