Network Effect & Switching Costs for Intuit’s QuickBooks Online

Intuit’s QuickBooks Online is its Software As A Service (SAAS) business. It is replacing Intuit’s legacy QuickBooks Desktop software. QuickBooks Online is part of Intuit’s effort to transform its revenue from highly repeatable transactional sales to highly recurring subscriptions.

QuickBooks Online is experiencing strong adoption and growth. I wouldn’t say it has built an enduring business moat yet. But I see two potential drivers for Intuit to build a moat around its QuickBooks Online business: Network Effect and Switching Costs.

Network Effect

Accountants & Small Businesses

One Intuit’s strategies for growing QuickBooks Online is to target the accountants for small businesses and the self-employed. These are the nodes from which QuickBooks Online can branch off of.

If a small business hires an outside accountant and that accountant uses QuickBooks Online then you as the small business owner would most likely use QuickBooks Online too.

The most significant one right now is really working with your accountant and we know that if an accountant switches their practice to QBO, they’ll bring their clients along with them and they’ll help the client in the transition. So that’s the big idea we’ve been working on for the last year or so, but I think it’s going to be pretty steady for a while. -Neil Williams from INTU Bank of America Global Tech Conference June 7, 2017

It also works in reverse.

The more small businesses that use QuickBooks Online the more pressure it is for accountants to use QuickBooks Online too. It can help the accountants win more business.

Two-Sided Network Effect

I see this as a Two-Sided Network Effect

Network effects can also be two-sided: increases in usage by one set of users increases the value of a complementary product to another distinct set of users, and vice versa. In many cases, one may think of indirect network effects as a one-directional version of two-sided network effects.

3rd Party Integrations

QuickBooks Online is an open platform. Other companies can build and integrate their software with it. For example, Intuit has its own Payroll and Payment Processing products but Intuit allows ADP, Paypal, Square, and others to integrate with QuickBooks Online.

The more 3rd party integrations QuickBooks online has the more valuable it is to their customers. And the more customers QuickBooks Online has the more valuable it is to 3rd parties to integrate their software with QuickBooks Online.

Indirect Network Effect

This is the Indirect Network Effect.

Network effects may also be indirect, where increased in usage of the product spawns the production of increasingly valuable complementary goods, and this results in an increase in the value of the original product. For instance, while there are some direct network effects associated with Windows, the indirect network effects that arise from the increased quality and availability of complementary applications software are probably much more important.

Switching Costs

If you’re a small business that uses QuickBooks Online, your accountant uses QuickBooks online, and all your other business software integrates with QuickBooks Online then it becomes hard to switch over to a new platform.

  • Can all your information and transaction history transfer over seamlessly?
  • How much work do you have to do the transfers?
  • How much will it cost in time and man hours?
  • How long will it take to learn the new system?
  • How many people need to be trained on the new platform?
  • Will you have to find a new outside accountant?

Unless something drastic happens with QuickBooks Online and the decision to switch becomes urgent and vital, you as the business owner are unlikely to switch.

I put QuickBooks Online’s switching costs in the following categories as laid out by Strategyzer.

Data Trap

The ‘Data trap’ consists encourages customers to create or purchase content and apps that are exclusively hosted on a platform. These platforms can be websites, software or devices. But, leaving one platform for another forces customers to let go of data or activity that can’t be migrated to another app.

For QuickBooks Online it is your data. It must be transferable. I see less of a risk of losing your data.

But I see the Data Trap as the worry that your data won’t transition smoothly between software platforms. You risk spending a lot of time trying to fix it. Or you’ll have to spend a lot of money and time working with your new software’s data transition team. Or you’ll have to do it through Excl exports increasing the time it will take and increasing the risk for entering the data wrong.

The time, money, and effort to transfer your data must be worth it.

Learning Curve Trap

Customers can be discouraged when they have to start over and learn how to use a new product. The ‘learning curve’ trap is centered around offering a great value proposition that’s only accessible to those willing to train to know how to use it. Salesforce and Adobe use the ‘learning curve trap’ to get customers hooked to their products–some users get so good at using their software that they become certified experts. They don’t feel like switching to something else unless they experience a very strong pain with their existing product.

If QuickBooks meets your needs and the user interface isn’t utter garbage then why learn a new system? A small business owner wants to focus on running and growing their business not learning new accounting software.

Plus, Intuit wants to retain its users and Intuit knows it needs to continually update and improve QuickBooks’ user interface to do so. Any current user interface issues can be resolved quickly.

Industry Standard

And then to a smaller extent the Industry Standard trap. This could be a bigger switching cost if QuickBooks Online becomes the industry standard for small business accountants. Right now QuickBooks online is the top choice, not the necessary choice.

QuickBooks Online Opportunity

Intuit has a real opportunity with QuickBooks Online to build a product with a strong network effect and high switching costs. It is why they are aggressively trying to gain more users. QuickBooks Online doesn’t have an enduring business moat yet and it’s still vulnerable to outside competition at this stage. But Intuit has a big head start with its entrenched users of its legacy QuickBooks Desktop software. Converting all of them to QuickBooks Online will make it extremely hard for competition to overtake Intuit.

What I Wish I Wrote ~ June 30, 2017

Charlie Munger’s talk on the Psychology of Human Misjudgement via Marketfolly

Checking in with ESPN, the current state of its business and its future. (Sports Business Daily)

You probably hear a lot that Tesla is disrupting this and that business. But is Tesla really disrupting any business in the true Clayton Christensen definition? (Vox)

How would your portfolio have done if you invested in the Dividend Champions in 2007? (The Dividend Growth Investor)

Brown v Board of Education was a landmark Supreme Court case but did we mess up who and how we integrated the schools? (Revisionist History)

Has the meteoric rise of passive investing generated the “greatest bubble ever”? (13D va Medium)

Tiger Cubs battle short sellers over Transdigm. (Financial Times)

Faceless Publishers. The changing infrastructure of online publishing and monetization. (Stratechery)

Best Habits of Highly Successful Investors (Steve Romick in Barron’s)

70 years ago Russ Gremel bought $1,000 worth of Walgreens now the stake is worth $2 million and he is donating it to set-up a wildlife refuge. (Chicago Tribune)

Compound interest works with learning too. (Kottke)

This is Your Nightmare Scenario (The Irrelevant Investor)

Warre Buffet on the PBS News Hour (Video 1, Video 2)

JP Morgan, Morgan Stanley, & Wells Fargo Capital Return Plans

One of our themes for investing in JP Morgan Chase (JPM), Wells Fargo (WFC), and Morgan Stanley (MS) is increased returns of capital to shareholders.

We believed that all three had more than enough capital to pass the Federal Reserve’s stress tests and return a large portion of their excess capital back to shareholders. Yesterday all banks passed their stress tests. Afterward JP Morgan, Morgan Stanley, and Wells Fargo, all announced bigger capital returns.

JP Morgan

From JP Morgan’s press release.

JPMorgan Chase’s Board of Directors intends to increase the quarterly common stock dividend to $0.56 per share (up from the current $0.50 per share), effective the third quarter of 2017, and has authorized gross common equity repurchases of up to $19.4 billion between July 1, 2017 and June 30, 2018 under a new common equity repurchase program.

That is a 12% increase in its quarterly dividend and at today’s trading price a potential 6% reduction in shares outstanding.

Morgan Stanley (MS)

From Morgan Stanley’s press release.

The capital plan includes the repurchase of up to $5 billion of outstanding common stock for the four quarters beginning in the third quarter of 2017 through the end of the second quarter of 2018, an increase from $3.5 billion in the 2016 Capital Plan, as well as an increase in the Firm’s quarterly common stock dividend to $0.25 per share from the current $0.20 per share, beginning with the common dividend expected to be declared for the third quarter of 2017.

This is a 25% increase in Morgan Stanley’s quarterly dividend. Morgan Stanley raised its buyback prorgam from $3.5 billion in 2016 to $5 billion, $1.5 billion more. At today’s trading price this a further 1.8% reduction in shares outstanding on top of the 2016 plan.

Wells Fargo (WFC)

From Wells Fargo’s press release.

Wells Fargo’s 2017 Capital Plan covers the four-quarter period from the third quarter of 2017 through the second quarter of 2018. As part of this plan, the Company expects to increase the third quarter 2017 common stock dividend to $0.39 per share from $0.38 per share, subject to approval by the Company’s Board of Directors. The plan also includes up to $11.5 billion of common stock repurchases1 for the same four-quarter period. For the previous four quarters ended first quarter 2017, the Company repurchased $8.3 billion of common stock.

Wells Fargo is only growing its dividend by 2.6% but the $11.5 billion earmarked for buybacks could reduce shares outstanding by 4% at today’s trading price.

How Likely is 9% Growth for Starbucks?

A key component in the economic profit models I ran for Starbucks is the assumption that Starbucks will maintain a growth rate of 9% over the next 5-10 years. A common estimate I’ve seen in other research pieces.

The question then is how likely is Starbucks, an already large company, to grow its sales at this rate?

Like I did when evaluating Ron Baron’s sales growth estimates for Tesla, I’ll turn to Michael Mauboussin and Credit Suisse’s Base Rate Book.

Since 1950 for the full universe of publicly traded companies 24.2% of all companies have been able to grow their sales at a compound annual growth rate of 5-10% over 5-years and 28.3% of all companies were able to grow at this rate for 10-years.

Sales Base Rate
Table from The Base Rate Book

We can narrow it down even further.

Starbucks Sales Growth Base Rate

Starbucks has trailing twelve-month revenue of $21.98 billion. This puts Starbucks in the cohort of companies wth sales between $12-25 billion.

base rate book
From The Base Rate Book. Click image to enlarge.

Only 20.8% of companies with sales totals similar to Starbucks grew their sales at a compound annual rate of 5-10% over 10-years.

41.7% of similar companies were able to grow their sales at a compound annual rate of 0-5% over 10-years.

Based on this data it is more likely for Starbucks to grow its sales at a compound annual rate of 0-5%.

It makes sense.

Starbucks and other companies with similar sales totals are already large. And large companies tend to follow the overall growth of the economy. They are well past their rapid growth stages.

This is not to say a CAGR of 5-10% is not possible.

Path to 10% Growth

Does Starbucks have a pathway to 5-10% growth over the next 10 years? Yes.


Starbucks is just getting started in China. Tea has always been the caffeinated drink of choice in China. If Starbucks can get Chinese consumers hooked on its specialty coffee drinks then this will be a large driver of growth for many years.

Mobile Ordering & Sales Per Sq. Foot

Starbucks is improving its mobile ordering and pickup to reduce the long lines at physical stores and the mosh pit of people waiting for their orders. When this is cleaned up, a Starbucks store can handle more volume and generate more sales per square foot.


Mobile ordering with delivery. The next step is figuring out delivery for mobile orders in densely populated cities.

Reserve Stores

Howard Schultz stepped down as CEO but he did not leave the company. He is focused on growing out Starbucks’ ultra-premium coffee stores, Reserve Stores. These are lower volume stores but with higher average ticket sales. Reserve stores are an opportunity to increase sales in mature markets like the U.S.

Wine, Beer, & Dinner

The weakest part of the day for sales is late afternoon and nighttime. Starbucks is experimenting with selling wine, beer, and a dinner menu to increase sales at night. People still want a quiet place to work at night and sometimes they want a drink when they do that too.

Base Rate Valuation

How confident am I that Starbucks will execute on these growth opportunities and achieve the expected 9% compound growth rate? Fairly confident.

But that is the problem. My confidence is skewed from my inside view. I need to take a more measured approach.

Given that only 20% of similar size companies were able to grow at a rate of 5-10% for 10-years. I need to adjust my primary valuation model for the more likely scenario, 0-5% growth.



Valuing Starbucks with Economic Profit Models

[edit]: My original economic profit model had an error. It was overestimating future invested capital. I kept the original images of each economic profit model but added updated values below them. This also changes my conclusion about Starbucks being overvalued or not.

If you only looked at the P/E ratio Starbucks (SBUX) looks expensive with a current P/E of 29. I want to look at it a different way. I want to look at Starbucks’ value through discounted economic profit models.

The discounted economic profit model allows us to move away from multiple based valuations and value companies based on the returns they generate on their invested capital.

Economic Profit = Invested Capital * (Return on Invested Capital – Weighted Average Cost of Capital)

In this exercise, I only want to play around with one variable, ROIC. Yes, I use two different discount rates but I do this to look at a range of estimated prices.

In each model, I will keep Starbucks’ expected growth rate over the next 10 years and its continuing value the same. Again, playing with one variable.

Starbucks Growth Estimates

The common expectation for Starbucks is it will grow 9-10% per year over the next 5+ years. I used this growth rate in each model.

I create models with lower growth estimates too. I want to look at a range of price estimates given different assumptions. I also do not want to make this post extremely long so I will use 9% as my growth rate assumption and just play around with different ROIC scenarios.

Starbucks WACC

Given the low-interest rate environment, I calculated Starbucks’ WACC to be around 6.62%.

When I adjusted current interest rates to a more “normal” environment, 4% on the U.S. 10 Year, then Starbuck’s WACC increased to 8.2%.

In the following models, I used the 6.62%. Maybe the U.S. 10 Year creeps back up to 4% but the issue is when will this happen?

What I wanted to do in this exercise is value Starbucks in the world we live in now.

Continuing Value

A big part of Starbucks present value comes from my estimate of its continuing value. My economic profit model projects the next 10 years for ROIC, Invested Capital, WACC, and NOPAT (Net Operating Profit After Tax).

For Starbuck’s continuing value, I had to come up with values for ROIC, NOPAT, the growth of NOPAT, WACC, Return on Newly Invested Capital for year 11 and beyond.

For continuing ROIC I used 15%. This is the average ROIC for food, beverage, and tobacco companies as calculated by McKinsey.

For NOPAT I used Starbucks’ 10-year average NOPAT margin to get these values.

For NOPAT Growth I assumed a 2% growth rate. I kept it in line with estimates for long-term U.S. GDP growth.

For WACC I used the average WACC for the Food and Beverage Industry. I estimated this average to be around 8.84%.

Return on Newly Invested Capital (RONIC). I brought this down to equal WACC. This is what would happen in a highly competitive industry. It is easy to argue that Starbucks’ brand and leading market position will allow it to earn higher returns on newly invested capital. I kept my expectations low.

Average ROIC

Over the last 10-years, Starbucks averaged a ROIC of 21.66%. In this model, Starbucks continues to earn 21.66% over the next 10-years. Then I discounted the economic profit created at Starbucks’ WACC, 6.67%, and then by 10%.

WACC Discount Rate

Economic Profit Model
WACC Discount Rate & 10 yr average ROIC. Click image to enlarge.

This model produced a per share equity value of $46.50.

[Edit] The updated model produces a per share value of $32.46.

10% Discount Rate

Economic Profit Model
10% Discount Rate & 10yr Average ROIC. Click image to enlarge.

This model produced a per share equity value of $43.69.

[Edit] The updated model produces a per share value of $30.78.


What if Starbucks maintains its current ROIC of 31.91% over the next 10 years?

It’s possible.

Starbucks is a big well-known brand with loyal customers. It’s app, rewards program, branded food sales, prime locations, and international growth can all certainly help Starbucks maintain a high future ROIC.

WACC Discount Rate

economic profit model
WACC discount rate & current ROIC. Click image to enlarge.

This economic profit model produced a per share equity value for Starbucks of $58.45.

[Edit] The updated model produces a per share value of $39.66.

10% Discount Rate

economic profit model
10% discount rate & high ROIC. Click image to enlarge.

In this economic profit model, Starbuck’s per share equity value is $53.72.

[Edit] The updated model produce a per share value of $36.84.

Fading ROIC

It is easy to argue that Starbucks has a large business moat that will allow it to continue to earn high returns on invested capital for many years into the future.

But Starbucks can lose its high ROIC. It operates in a highly competitive industry. Starbucks’ ROIC could fade towards the industry average as competitors take business away from Starbucks.

This economic profit model starts with Starbucks’ current ROIC and faded it to the Food & Beverage industry average of 15%.

WACC Discount Rate

economic profit model
WACC discount rate & fading ROIC.

This produced an equity per share value of $47.15 for Starbucks.

[Edit] The updated model produces a per share value of $32.92.

10% Discount Rate

economic profit model
10% discount rate & fading ROIC. Click image to enlarge.

This economic profit model produces a per share equity value of $44.66.

[Eit] The updated model produces a per share value of $31.42.

Industry Average ROIC

In this model let’s assume Starbucks loses its dominant market position and its ROIC immediately falls to the food and beverage industry average of 15%.

Why would we keep the growth rate at 9% if Starbucks experienced an immediate reversal like this? For simplicity sake. I just want to play with one variable, ROIC, in this exercise.

WACC Discount Rate

economic profit model
WACC discount rate & industry average ROIC. Click image to enlarge.

Starbucks has a per share equity value of $38.74 in this economic profit model.

[Edit] The updated model produces a per share value of $27.79.

10% Discount Rate

economic profit moddel
10% discount rate & industry average ROIC. Click image to enlarge.

Starbucks has a price per share of $37.17 in this model.

[Edit] The updated model produces a per share value of $26.85.

Which is Correct?

The industry average ROIC model is worth looking at. Is it highly likely to happen over the next 5-10 years? I don’t think so.

The fading ROIC is helpful if I made a mistake about Starbucks’ business moat. How much am I overpaying if Starbucks is not as dominant as I thought and Starbucks mean reverts towards being an average business? Based on these models, I wouldn’t be overpaying by too much.

The two models that should garner the most weight are the 10-year average model and the continuing high ROIC model.

I think Starbucks does have a large business moat that will allow it to generate high returns on capital going forward.

Starbucks currently trades around $60 per share. This looks like a fair price for Starbucks based solely on the economic profit models used in this post. Maybe a few dollars over the exact estimates of fair value but models are not an exact science.

Contrary to Starbucks P/E ratio and based solely on these economic profit models Starbucks does not appear overvalued.

[Edit]: The 10-year average ROIC and the High ROIC model are still the most applicable valuation models. But now instead of being reasonably priced the models are saying that SBUX is overvalued. Using ROIC, WACC, and Growth rate, and the Gordon Growth Model, I come up with a levered P/E ratio of 22. Applying this multiple to Starbucks’ earnings produces a per share price of $44.

Using ROIC, WACC, expected growth rate, and the Gordon Growth Model, I come up with a levered P/E ratio of 22. Applying this multiple to Starbucks’ earnings produces a per share price of $44. In line with what the updated economic profit models are saying.