The ABC of the P&L...

If you are the GM of a for-profit lending business, you wear many hats - from marketing to product to risk management to operations to engineering to legal and compliance and so on. But by far your number one responsibility is to generate profit for the business, the company and your shareholders... or at least build a viable path towards profitability initially. In the case of most businesses, the path to profitability is fairly well laid out (that does not mean it is easy to tread :-) ) 

1. You build a product with great product-market fit
2. You launch the product and then scale using the right marketing, sales, distribution channels
3. You continue to innovate and improve
4. You optimize your operations to reduce unit cost 

However in lending, as discussed in one of my earlier posts, the path to profitability is not always linear - because the losses are asymmetrical and if you are not careful enough, attempts to grow revenue can also drive up losses at an asymmetric rate thereby potentially blowing up the business. 

To recap, the simple dynamics of a lending P&L is that you borrow money short to lend long. You deduct expenses and losses and whatever is left is your profit. Let's quickly recap the key elements of a Lending P&L:

1. Revenue - This is the top line and particularly important if you are a new/growth company. After-all, your investors want to see a healthy top line growing at a rapid clip to feel confident in continuing to invest in your business. This is typically a direct function of the number of loans you book, the size of the loans and the price - i.e. the interest rate you charge.
2. Cost of Funds - To lend money, you need to ‘borrow’ money - be it from the capital markets, depositors (if you are a bank), or other banks. Another way to fund loans of course is to use equity from the investors. For non-banks (that do not have access to cheap customer deposits), this is one of the biggest costs.
3. Losses - Anytime you lend, you will have losses - people who will not pay you pack - either because they cannot (credit loss) or because they won’t (fraud losses). In lending, credit and fraud losses have an outsize impact to the P&L - so this should be a high priority item for the business owner. Higher losses also drive up the cost of funds.
4. Marketing - Obviously, if you want your business to grow and revenue to expand, you need to invest in the right marketing, sales, distribution engine. All that costs money.
5. Operations - Running a lending business requires investments in operations. This includes the various tools and infrastructure to extend a loan, to running the customer service centers, to paying the third party vendors who provide you with various critical solutions etc. 
6. Fixed costs - These are staffing and technology costs and any other one time set up costs that you have incurred to launch and run the business.

The curse of the J curve: What you have in front of you then is an optimization problem -  “How do you grow the revenue while managing credit and fraud losses and reducing/optimizing operations and marketing cost?" What makes lending so interesting (read challenging) and different from the other businesses is the complex interplay between growth and losses and the timing mismatch between the losses and the revenue. Assuming you carry the loan balances in your books (I will cover the P&L dynamics of the held-for-sale model in a separate post), a disproportionate part of the expenses in a lending business happen upfront - you typically take the largest chunk of your credit and fraud losses within the first year, your marketing expenses hit your P&L right away (although some marketing expenses can be amortized over a longer time horizon), but your revenue takes years to accumulate - as the borrower pays you back over time. The recent migration to the CECL accounting rules means that provisions for future credit losses have to be taken right away - i.e. as soon as you make the loan, even before you earn a single cent in revenue! So when you look at the typical lending account, what you will see that marketing expenses, fraud losses and provisions for credit losses are all very front-loaded. Whereas on an average it could take upto 60 months or longer in some cases for the revenue to accrue. This leads to the dreaded J curve in lending. What makes matters even more challenging is that the more you step on the gas to grow the loans (both in volume and size of loans) thinking it will help your P&L, the worse the J curve gets. Two things happen almost right away - your loss provisions go up because your losses are also a direct function of the loan balances and your marketing cost goes up which means that your P&L for that initial period looks quite shitty - and the faster you grow, the worse the P&L looks. And if your investors and leaders do not understand this dynamics, to them it seems like you are digging a perennial hole for the business. 

The horizontal and vertical views of the P&L: To address the curse of the J curve on the P&L, what GMs of lending businesses do is to focus on creating a ‘horizontal view’ of the P&L along with a ‘vertical view’. A horizontal view is basically looking at the economics of an account or a vintage of accounts over a longer time horizon - aka the lifetime valuation of the account or the vintage. Now building a valuation model is no easy feat and I will dedicate a separate post to it, but as the GM of the business, you should absolutely invest the time and effort to get a 'V.0' account/vintage valuation model as quickly as possible and then continue to iterate on it. Based on what you see from your model, you can make some critical business decisions. If your valuation model shows that your account/vintage is not net positive within a reasonable time frame (say 5 years), i.e. your strategy is booking the wrong set of customers, then you know it is time to change course. If and once you see a positive valuation (with your current strategy or with some adjustments), you then build a multi-year P&L . Obviously, in the short term, things are going to look very rocky, as your portfolio consists of only new accounts which initially have very shitty economics. However, year after year, as you stack up accounts with good positive economics in the outer years, something interesting happens - you end up with a book where the dynamics starts to flip - you will have a bigger share of more seasoned accounts (>18-24 months) compared to new accounts. That’s when your business will start generating a positive vertical P&L. 

As a GM, when you launch a business, you cannot entirely avoid looking at the vertical P&L but I would be wary of making strategic decisions based only on the vertical P&L. Early on in the business, the vertical P&L does not give you much useful information at best and at worst may lead you to make the wrong bets. To illustrate this point, let’s go through a hypothetical (and very simplistic) example. Say you launch a lending business and you have two strategies that you are using to book accounts - very creatively, we will call them strategy 1 and 2!

You have done some early valuation work and have come up with the following 10 year account-level valuation for the two strategies. We will assume that beyond year 4, the economics of the two strategies stabilize - i.e. strategy 1 generates $120 per year per account and strategy 2 generates $175.

It’s clear that all things being equal, strategy 2 has better longer-term unit economics although it has worse year 1 economics. It is very common to have higher 'costs' initially to ensure better economics down the road - think of all the investments you will need to make upfront on product, infrastructure, data, marketing, hiring etc. But, how does the vertical P&L look for the two strategies? To keep things simple, we will assume that we are booking the same number of accounts from both the strategies:

Now, if we did not have the benefit of the long-term valuation of the two strategies and only looked at the vertical P&L, things would look pretty bad for strategy 2 all the way to year 6 because of the inferior year 1 economics...

…and I bet by year 3 or 4, you will be under tremendous pressure to shut down strategy 2 altogether. But because you now have the benefit of the longer term valuation, you know that if you persist with strategy 2, things will eventually turn around and they indeed do starting year 7.

This is a very simplistic example but I hope it drives home the importance of investing in building a valuation framework for the accounts/strategy you book as quickly as possible - otherwise you run the risk of not making the right business decision. 

So what? Okay, so till this point, some of the concepts that I've shared may be familiar or even obvious to many of you. The burning question though is “so what levers do I pull to improve my P&L so that my account-level valuation can go from this:

to this...

Unfortunately, there is no silver bullet. Improving the economics of an account and building a viable lending business is a long, painful and arduous process. The exact strategy depends on a lot of things - the product, the state of competition, the economic environment you are in, the segment you are targeting, your brand (or lack thereof :-)) and so on. It’s important to start this journey by acknowledging that each one of your competitors is looking to do the same things and many of them likely even have a leg-up over you - better data, better products, better/more talent and resources and better brand (assuming you are a new player in this space). Your mindset when you launch a lending product should  always be to ask “how can I build an advantage and/or differentiate my business from what is already out there” and unless you have a unique and differentiated offer or are catering to an underserved segment or meeting an unmet need, your chances of generating a profitable long term account value (and consequently a business) are slim. As a GM of a lending business unit, it is good to start with the following basic framework and probe them each to understand where and how you can create that 'wedge'.

1. The underwriting and risk management  - Can you build an underwriting and risk management framework that gives you a distinct advantage over your competition. For example it could be that your business has an asymmetric data or technology advantage - if you are building a lending business to cross-sell to customers of an existing product, you likely have proprietary information on your target customers that others will not have or if you can  build an in-house decision engine that can in real-time leverage data from multiple data sources that you know your competition is not using than you have an advantage. The exact type of the advantage will vary but goal is to leverage some kind of ‘advantage moat’ that will set you apart from what is already available. 
2. The segment - An extension of the above point is can you find a segment that is underserved in the market. Back in the day, my alma mater Capital One was the first company to break into sub-prime credit card market when no other bank was willing/able to offer credit cards to that segment. They did that using their famous and now often imitated IBS (Information Based Strategy) approach. Here's a link to a famous HBS review on this topic.
3. The product - You will need to extend the same concept on the product side. Having a generic product that likely everyone else is also offering in the market will likely not cut it for a new lender - you will have to differentiate yourself on some product dimension - either the product has to have very attractive terms and pricing or have some unique set of features or the customer experience has to be something new and unique. As you can imagine, all of these things will require initial investment and will make the P&L look worse. However, if you have a horizontal economics view, you can hopefully show that they pay-off handsomely in the long term.
4. The Marketing and Distribution - Notwithstanding the early losses incurred by Marcus's  #applecard, the product is a great example of leveraging a unique distribution channel. If you can create a unique distribution channel or a marketing strategy that allows you to target more effectively and efficiently, you will likely have an advantage. The key to building a successful lending business is to find the 'Goldilocks' borrower who has some meaningful credit need but can also handle credit responsibly. To do that you will have to make meaningful investments in the data and infrastructure.
5. The servicing and portfolio management - Finally for products like Credit card, servicing cost is a huge drag on the P&L. Leverage data as much as you can to create a very segmented servicing strategies - i.e. you should be able and willing to spend more on high value customers - the returns will be there. Also not to state the obvious, as a new lender, investing in a digital-first servicing strategy, especially if your target demographics skwes on the younger side, will help the P&L.

So a lot of these things will take time and effort to iterate and perfect over time. Creating a set of 'differentiated' features also takes initial investment making the J curve look worse. But your hope is that that eventually these investments will improve the longer term economics. Make sure you adopt a robust test and learn approach and each time you make changes, make sure to test it out with a small group and then within the first 3-6 months and feed the early results from the test to the valuation model to assess the impact on the valuation. Hopefully, as you iterate your strategy over time, you will start generating accounts that will have the desirable valuation and lead to a healthy and robust lending business.

So there it is folks! Curious to hear from other business leaders how they navigated "the curse of the J curve" in their respective lending businesses. Please offer your thoughts in the comments below or feel free to DM me.