The decision is usually framed as a cost comparison, and that framing is why it’s so often made badly. Building and partnering don’t differ mainly in cost. They differ in when you have the capability, what happens if it fails, and what you’re able to learn.
A credit union that decides to enter commercial real estate lending needs underwriting judgment, servicing infrastructure, portfolio reporting, and workout capability. Those exist somewhere on day one or they don’t. Building means they don’t exist for a while, and your first loans are underwritten by people learning the asset class. Partnering means they exist immediately, and you accept a counterparty and an oversight obligation.
This article sets out what each actually requires, where the crossover sits, and how to tell which fits your credit union now. For the partnership model itself, see our guide to what a CUSO is.
Key Takeaways
- Building is a multi-year investment before the first loan closes — hiring, systems, policy, and process, then the learning curve on live credit.
- Partnering converts a fixed cost into a variable one and adds a third-party oversight obligation you must staff.
- The crossover is driven by sustained volume, not by a single good year.
- Workout capability is the hardest thing to build and the thing you need least often, which is why it is the most commonly partnered function.
- Judgment should never be outsourced, whichever model you choose. The credit decision stays with your credit union.
What Building Actually Requires
People. A commercial credit officer with genuine CRE experience, underwriting support, a servicer who understands commercial documentation, and access to workout expertise. That’s not one hire. It’s a small team, and the senior member is expensive and hard to find.
Systems. A servicing platform that handles commercial structures — draws, reserves, escrows, covenant tracking — plus reporting that satisfies your board and your examiner. Consumer platforms don’t do this well.
Policy and process. Credit policy for each loan type, risk rating methodology, approval authorities, annual review procedures, appraisal review standards, and concentration limits.
Time. Hiring, implementation, and policy development happen before the first loan. Then the learning curve happens on live credit, and CRE cycles are long enough that a team can look excellent for several years before being tested.
The unbudgeted item. The cost of loans that go wrong because the team was learning. It rarely appears in a build business case, and it is frequently the largest number in it.
What Partnering Actually Requires
Partnering isn’t free of obligation; the obligations are just different.
Fees that scale with volume — the point of the model, and also the reason it gets expensive at scale.
Oversight. Third-party due diligence at onboarding and continuing monitoring of the provider’s performance and financial condition. This is real staff time, and NCUA expects it to be proportionate to how critical the service is. Our CUSO diligence guide covers the programme.
Dependency. Your capability lives partly outside your building. Exit planning matters from day one.
Judgment you must still exercise. A partner can underwrite and recommend; your credit committee decides. Regulatory expectation on participations is explicit that the buying credit union underwrites to its own standards, and the same logic applies to work a partner performs.
Where the Crossover Sits
The honest version is that it depends on sustained volume, and the sequence matters more than the arithmetic.
At low volume, building is clearly wrong. A commercial credit officer underwriting six loans a year is expensive per loan and, worse, isn’t getting enough reps to stay sharp.
At high sustained volume, building becomes compelling. Fees scale with the portfolio while a team’s cost doesn’t, and at some point you’re paying more than a department would cost while still carrying oversight.
In the middle — where most credit unions entering CRE actually sit — the answer is usually partial. Partner for the functions that need depth and are used intermittently; build the ones you use constantly.
The trap is deciding on a peak year. Commercial volume is lumpy, and a team hired against one strong year becomes a fixed cost through the quiet one. Judge on the volume you expect to sustain.
The Cost Components Nobody Prices
Build business cases tend to count salaries and a system. The components below are the ones that decide whether the number was right.
| Component | Build | Partner |
|---|---|---|
| Senior credit officer | Salary, benefits, and a search that can take months | Included in the service |
| Underwriting support | One or more analysts | Included |
| Servicing staff and platform | Salaries plus licensing and implementation | Fee-based |
| Policy and process development | Months of senior time, often external help | Provider’s, adapted to your policy |
| Appraisal review capability | Specialist skill, often contracted anyway | Usually included |
| Workout capability | Rarely hireable at the moment of need | Retained on call |
| Learning-curve credit losses | Real, and usually omitted | Borne by an experienced team |
| Third-party oversight | Not applicable | New internal function you must staff |
| Exit or unwind cost | Redundancy and stranded systems | Transition assistance, if negotiated |
Two rows change most decisions. Learning-curve losses belong in the build column and almost never appear there, though they’re frequently the largest figure. Third-party oversight belongs in the partner column and is equally often omitted, which is how credit unions end up with a relationship nobody has time to supervise.
Run the comparison over a realistic horizon as well. A build looks expensive in year one and better in year five — but only if the volume you projected in year one actually arrived, and only if the team you hired is still there.
The Capability Ladder
Most credit unions that end up with strong commercial operations didn’t choose once. They moved in stages.
Stage one — partner for everything. Enter the asset class with sourcing, underwriting, servicing, and reporting provided. You learn on real deals without your members bearing the cost of your learning.
Stage two — insource what you use constantly. Usually credit analysis first, because it’s the function where in-house judgment compounds and where you want your own view.
Stage three — insource servicing if volume justifies it, keeping specialist functions like workout with a partner.
Stage four — full in-house, with a partner retained for participation flow or overflow.
The advantage of the ladder is that each step is reversible and each is justified by volume you’ve actually seen rather than volume you projected.
What You Should Never Outsource
The credit decision. A partner underwrites and recommends. You approve or decline, against your policy and appetite.
Risk appetite. Concentration limits, loan types, geography, and exposure ceilings are board decisions.
The member relationship. Structure the arrangement so your member deals with your credit union.
Oversight. Delegating work never delegates accountability, and the examination lands on you.
If a provider wants the credit decision, that isn’t a partnership — it’s a referral arrangement wearing a partnership’s clothes.
Signals It's Time to Insource
Fees are approaching the cost of a team, on sustained volume rather than a spike.
You have the reps. Your staff have participated in enough credits to hold their own view, and they’re disagreeing with the partner occasionally and being right.
The work is predictable. Volume and loan types have settled enough to staff against.
Strategic weight. Commercial has grown from an experiment into a core line, and you want the capability permanently.
Signals you should wait: none of the above, plus any period when the portfolio is stressed. Insourcing during a workout cycle means learning the hardest part of the discipline at the worst time.
Common Mistakes in Both Directions
Building too early, on a business case that omits the learning-curve losses.
Partnering without oversight capacity, which converts one problem into two — you still don’t have the capability, and now you have a relationship you can’t supervise.
Assuming the decision is reversible at no cost. Moving in either direction has a price: redundancy and stranded systems coming out of a build, transition and data migration coming out of a partnership. Neither is prohibitive, but both belong in the comparison.
Choosing on fee alone. A cheaper partner with no workout capability is not cheaper. The saving is real until the first loan deteriorates, at which point the difference between a servicer that has run a foreclosure and one that hasn’t is worth more than several years of fee differential.
Never revisiting. A decision correct at $20 million of commercial exposure may be wrong at $200 million. Put the review on a schedule, and give someone ownership of bringing it back to the board.
Building the easy parts and assuming the hard parts. Credit unions often hire origination and discover that servicing, monitoring, and workout were the actual constraints.
Treating the decision as permanent. Neither model is a commitment for life. The credit unions that handle this best revisit the question on a schedule and move deliberately along the ladder, rather than defending a choice made years earlier under different conditions.
Frequently Asked Questions
Is it cheaper to outsource commercial lending or build it in-house? At low and moderate volume, partnering is usually cheaper because fees scale with the portfolio while a team’s cost does not. At high sustained volume the comparison reverses. The larger factor is time to capability and the cost of loans underwritten while a new team is learning.
What’s the hardest capability to build? Workout. It requires experience you can’t hire cheaply, is needed infrequently, and matters enormously when it’s needed, which is why it’s the function most commonly retained with a partner even by credit unions that have insourced everything else.
Can we partner now and bring it in-house later? Yes, and that’s the most common successful path. Partner to enter the asset class, then insource functions as sustained volume justifies permanent headcount.
Does partnering mean giving up control of credit decisions? No. In a properly structured arrangement the partner underwrites and recommends while your credit committee approves or declines against your policy. A partner that wants the decision is offering the wrong arrangement.
What do we need internally even if we partner for everything? Someone who owns the credit relationship and can challenge the partner’s analysis, and someone who owns third-party oversight. Without those two, you have a vendor rather than a capability.
Making the Decision
Frame it as a capability question with a time dimension, not a line item. What do you need to be able to do, by when, and what happens if that capability fails on a Tuesday?
Three questions settle most of it. What volume will you sustain — not peak, sustain. What would it cost to be wrong while a new team learns on live credit? And who owns oversight if you partner, because that role has to exist and be staffed.
AVANA CUSO partners with credit unions on commercial real estate lending — origination, credit administration, servicing, participation, and portfolio management, à la carte so you can take only what you haven’t built. AVANA Companies, our parent organization, has closed $6.5 billion+ in loans across all 50 states.
POWERFUL PARTNERSHIPS. DIVERSE OPPORTUNITIES. INCOMPARABLE SERVICE.
This article is general information, not legal, compliance, or investment advice.