Download Snap Story
Wesley Dupler
You can make a private story on Snapchat from the Snap tab. Snap a photo or record a video and tap +New Story > Private Story (Only I can post). Choose the contacts you want to share it with and tap the checkmark to post your Private Story.
Typically, S/U is an acronym that stands for "Swipe Up" for different frames of a Snapchat story. However, people might also use it in place of "Shut Up," although this is less often the case. Considering the context will help determine the intention of the acronym.
Problem with professors is - they focus too much on excel sheet calculations. First and foremost is business model - Is it a sustainable one?
Then, all these excel sheet numbers. That's why these professors and economists rarely become great investors.
Stock Market,
Do you read posts before you comment on them? The entire post is about Snap's business model and story. I don't think I mention the spreadsheet more than once and only in passing. And if you restricted stock market commenting to just great investors, we would be looking at not much commenting, right?
Thanks for the post Aswath! Always a pleasure reading your take on IPO's.
This question relates less to the value of Snap and more toward pricing. I'm curious as to how the dozens of technology etf's will influence share price and activity (volume). Considering most funds are rule based, will not thousands of both passive and active funds be required to 'snap up' the company by virtue of its size, industry, NYSE listing etc.? Will this buying pressure outweigh all selling power from a potential misprice? Additionally, when will the S&P 500 selection committee will include the company in the index?
Thanks!
Great analysis as usual. Would have loved some more details on a few things:
1) whether the user segment (15 -24 predominantly) is attractive to advertisers. They don't have much spending power. Tumbler was similar and had hard time monetizing that user base.
2) Competition: Instagram has done a great job using the photo/video story of Snap and marketing to a wider user base.
3) Expansion: Snap is almost non-existent outside US.I was surprised how popular Instagram is with teenagers as well as all ages in India
Great post. I'd like to ask where you got the 10% chance that Snap will not make it? And whether you think it would make sense to compute a probability distribution varying that probability?
Thanks
This range forecast is a brilliant idea. I've never seen it before. It makes the IPO pricing sensible (from the perspective of the company and the bankers). I think they will get filled at that level because there will be true believers who think it's worth more. I'll take a swing at it after it lists if it tanks into the $8-10 billion range. Thanks Professor.
Professor, I am sorry if I had hurt you with my opinion/comments. I went though whole post and focus is more on numbers and estimates - that kind of approach may be right for normal companies but not for this kind of silly companies which are just riding the wave.
I am sorry.
Share count questions.
I used the prospectus and it does include shares that will be outstanding after the offering. That said, there is some fuzziness in the numbers especially in the shares set aside to cover options and RSUs for 2017. The shares set aside for options are dealt with separately in the option section of the valuation. I counted all of the other shares that are lined up, except for one offering that is set aside to replace shares that may be used to cover existing shares that will be issued on RSUs.
On the young and fickle customer base
It is true that Snap's target audience of young over-sharers has good points (they account for a disproportionate amount of online time) and bad ones (they often don't have incomes and are subject to fads). That will be one of the trickiest aspects of Snap's evolution and something worth watching.
Steven,
The 10% is just my estimate and here is what I use. For start-ups with potential and no product, I use 50-60%. For young, private companies that are dependent on VC capital, I use 20-30%. As a company goes public, I start lowering that number because you have more access to capital. Once they start making money, I will move it to 0%.
Tobias,
Thank you. Simulations and probability distributions were out of our reach (in terms of access and cost) thirty years ago. I don't know why we do not use these tools more in analysis.
Stock Market,
I don't have a thin skin. So, I did not take your comment personally. My only reason for responding was that the comment would have been better directed at a spreadsheet-based valuation, which this one is not. And I will cheerfully admit that while my family is happy with me as a portfolio manager (they cannot fire me easily), I have not made billions!
I do not think any of Twitters issues are due to management, and doubt that even if one thinks that Snap management is more capable, that it will be able to support for any period of time a "pricing" at the level of that desired by the underwriters. Here's why: Advertisers have not seen sufficient reason to advertise on Twitter, and while digital spend for advertisers has increased, advertisers have become more circumspect in where they place ad dollars. FB is an example: Advertisers are demanding better metrics of measurement to see the effectiveness of the dollars going to FB. So, with advertisers being more reluctant to just spread the ad dollars, and instead desiring meaningful targeting, it is hard to see material ad spend heading to Snap given its demographics.
Before seeing this post I pegged the IPO "pricing" as no more than Twitters current market price and would not be surprised if it trades down if the underwriters achieve their lofty IPO price.
Thank you for making these available to the large public. I love your forward thinking approach.
To add to snap's story, I believe that the psychological understanding of the user base is key to understanding the value and potential of snap.
The use of the snap filters such as the "butterfly crown" to post, for example, selfies on instagram is not a gimmick for these users. I personally have a hard time understanding the appeal but there is definitely a large loyal user base that use these as a lifestyle. So from that perspective snap being labeled a camera company makes sense; these filters are not a gimmick for the users and more of a way to represent their reality (or the reality they wish to have).
Professor,
Snapchat is the same thing as its older cousin in China WeChat. WeChat has a profound impact on China, virtually a new platform for doing almost everything. If snap could be nearly as successful in US, then the potential is huge.
"After all, if you welcome me to invest me in your company and I do, you should want my input as well, right?"
Very utopian. Really the purpose of these IPOs is for company founders, employees, and early investors to cash out, not to want public shareholder input. The purpose of public markets is to raise capital but these internet companies need very little capital, they could stay private forever.
Dear Professor,
Thank you very much for your post. My question is more general than directly related to this post. It is about the way you calculate the sales-to-capital ratio. I have noticed that sometimes is computed as sales to book value of invested capital AND sometimes as a change in sales over reinvestment (with the latter item being net capex AND net WCInv).
My first question, is which one of these is better to use and when? My personal guess is that the one dealing with BV, shows the sales/cap ratio of the existing assets in place, whereas the latter of the marginal one, if my wording is correct.
The second question is about the non-cash WC investment. There are two ways one may calculate it using either a balance sheet OR a cash flow statement. However, I notice it very often that there are differences (sometimes staggering) between the two. As such, which one approach would you personally recommend and why?
Kind regards,
Tim
Professor,
I have a question on how you computed a sales-to-capital ratio of 2 for Facebook, which you apply in your valuation of Snap. It just happened so that a couple of days ago I computed the same ratio for Facebook and my number was about 1.0. My educated guess is that you may have deducted goodwill from the book value of invested capital. I would appreciate if you could clarify this point.
Thanks,
Ankit
great stuff as usual!!
Just one minor comment: A fair assumption is that per-user the capital expenses (in terms of server cost) will continue to fall over time primarily because computer hardware costs keep falling (a GB today is far cheaper than a GB a decade back, same for bandwidth). So I think that snap is doing the right thing by outsourcing the hardware/servers and focusing on growth instead. If it becomes profitable it always has the opportunity to build out it's own data center, and that capex will also help reduce it's taxes. If user growth slows down, it just needs to pay less to the server providers today.
To cut a long story short, I think snap has an asset light model, which is very smart and flexible. It also has the option to reinvest (eventual) profits to save on hosting costs eventually. This fact may boost the valuation to some extent at least.
Dear Professor,
Thank you for the informative post. I am intrigued by your way to use probability distribution to estimate the key parameters. Can you write more on the topic on how to create probability distribution in excel please?
I tend to visualize many outcomes in the market as being bimodal or even binary. For example, I find it quite unlikely that Snap achieves an average outcome - either it successfully converts users to revenue, or it doesn't.
At a theoretical level, normal(ish) distributions arise from the Central Limit Theorem as the consequence of the interaction of many independent random samples.
With this in mind, a company that participates in many diversified and independent sources of revenue will have well-behaved outcomes. A company like Snap whose success depends almost entirely on whether the business model is viable has nothing to pull it towards an average outcome.
At the extreme, certain situations (FDA approvals, election results) consist of only a single event and the distribution of outcomes should just be two different prices, weighted by their probability of occurring.
In this model your valuation is controlled almost entirely by the probability of success/failure, which seems nearly impossible to estimate without considerable domain knowledge.