Life with a lender

The facility is live. Here's what you actually signed up for, and how to run it so the next negotiation — the upsize, the amendment, the refinancing — starts from strength.

The reporting apparatus

Take the reporting section of your loan agreement and turn it into a calendar. It will look something like:

Two operating rules. Automate to the source. Weekly loan tapes assembled by hand in spreadsheets fail within a quarter — a wrong tape is a misrepresentation risk, and a late one is (per Part 8) potentially a zero-cure default. Pipeline it from your servicing system on day one; the engineering week pays for itself immediately. Never miss the boring stuff. Reporting punctuality is the lender's highest-frequency signal about your operational competence. Fifty on-time tapes is how "waive it, they're solid" gets said in a committee someday. It's the cheapest trust you'll ever buy.

The lender update: your most underrated capital-markets asset

Beyond the contractual minimum, run a proper periodic update — quarterly as a baseline, monthly during ramp or turbulence. Over time this document is your reputation, and eventually it writes your upsize memo for you. The skeleton that works (Appendix C has the full template):

1. Executive summary — the quarter in five bullets, growth and problems.
2. Wins — new channels, partners, products; structural improvements (a risk-sharing arrangement, a better data source).
3. What's not working, with root causes and remediation — the section that builds compounding trust. "Volume missed plan by 20%; the driver was partner X's delayed launch; here's the recovery plan and the revised date." Lenders can absorb almost any miss that arrives explained, owned, and planned-for. What they cannot absorb is discovering a miss themselves, in your tape, before you mentioned it. The candor bar is: your lender should never learn bad news from the data before they learn it from you.
4. Actuals vs projections — origination volume and outstandings, monthly, by product, against the plan you gave them. Yes, showing the misses. The borrower who publishes their own variances owns the narrative around them; the borrower who doesn't gets a narrative assigned.
5. Portfolio performance — delinquency, losses, vintage curves; commentary on anything drifting; distance-to-trigger on every covenant (they're computing it anyway — compute it first).
6. Pipeline with honest stage labels — and this includes the graveyard: the channels you paused, the partners you exited, the deals you walked away from. A pipeline where nothing ever dies is a pipeline nobody believes; visible pruning is what makes the live entries credible. Define your stages with numeric gates ("ramping = $250k+/month") so status changes are facts, not vibes.
7. Corporate — runway, equity plans, key hires. Your equity timeline is covenant-adjacent information (Part 5); volunteering it beats having it extracted.

Send it to prospective lenders too — the funds that passed, the ones you'll want next year. A clean update landing quarterly in the inbox of a lender who said "come back with more data" is the come-back. By the time you formally raise again, half the diligence has read itself.

The credit-ops muscle

At institutional scale, lenders expect the machinery behind the numbers to be documented — not aspirationally, but in written procedures that diligence teams read and test against reality:

• Credit model documentation: what models exist per product, their inputs, training standards, retraining cadence; monthly monitoring of predicted-vs-actual with defined review triggers (e.g., "model performance degradation beyond X triggers formal review"); periodic independent validation.
• Loss forecasting: vintage-curve based, refreshed monthly with actuals, with scenario analysis (base / downturn / severe) and documented assumptions.
• Committee governance: a credit committee that approves model and policy changes, with minutes. Small companies flinch at this — it can be three people and a standing meeting — but "who approved this change, when, and on what analysis" needs a written answer.
• The policy suite: credit policy, servicing and collections policy, charge-off policy, fair lending / responsible lending policy, complaints handling, information security, business continuity, vendor management. Each facility's diligence asks for roughly this stack; the SPV-grade version adds flow-of-funds documentation and operational-security policies.

Build it incrementally and honestly — a thin, true policy beats a thick aspirational one, because diligence tests these against practice, and the gap is what kills credibility. Start the folder now (Appendix B lists the full set); every document in it is reusable across every future facility, forever.

Amendments are sold, not given

Sooner or later you'll need something changed: a draw window extended because ramp was slower than plan, a covenant reset after a noisy quarter, a trigger waived, an eligibility criterion widened. Understand the economics of that conversation before you're in it.

An amendment is a product the lender sells. The price is paid in terms: extensions get traded for mandatory minimum draws (use it or lose it — you will borrow on schedule now), for raised cash covenants, for new fees, for tightened eligibility screens — and, per Part 8's pattern, negotiated upside (your performance step-ups, your fee step-offs) is the first currency spent. Amendment documents also arrive wrapped in lender-protective boilerplate — "no waiver of any other rights, all rights reserved, this course of dealing implies nothing" — which is standard: the lender preserving its rights while granting the ask. Read it knowing that's what it is.

Your amendment price is set by your track record, months in advance. The borrower who reported on time, flagged every problem first, and beat a conservative plan gets the cheap amendment. The borrower who went quiet for two quarters and arrived needing an extension gets the expensive one. By the time you need the amendment, the price is already set — you set it with every update you sent or didn't. (This is also the compounding case for Part 7's undersized facility: the beating-plan borrower simply needs fewer amendments.)

Tactically: go early — a covenant you might trip in two quarters is a conversation; a covenant you tripped last week is a confession. Bring the analysis pre-done: what happened, why, the fix, the projections under the fix, the specific relief requested. Make it easy to say yes to — their deal lead has to take your ask to the same committee that approved the facility, and (exactly like Part 7's negotiation asks) a specific, reasoned, well-documented request is a memo paragraph, while a vague plea is a workout referral.

And know your walk-away. Even mid-facility, alternatives discipline pricing. A performing book always has other homes, and your lender knows it — it's why the exclusivity and ROFR terms of Part 5 were worth negotiating carefully back when you had the pen. If you kept your escape hatches, you still have a market. If you didn't, this conversation is where you find out what that cost.

Run all of this well — clean reporting, candid updates, documented ops, plan-beating performance — and something pleasant happens: the facility stops being a constraint and becomes a flywheel. Which brings us to the endgame.