Description

Short: Twitter, Inc. ($27.50)

“Twitter is such a mess, it’s as if they drove a clown car into a gold mine and fell in”
—Mark Zuckerberg

Twitter is a niche product whose growth has stalled, that longs mistakenly believe belong in a peer group of products like Facebook, Instagram, and Snapchat, leading to an overvaluation of its stock. The lumping together of social media companies misses the intricacies of these products and their commercial value. This mistake is compounded by expensive late stage VC valuations spilling into the public markets, which are arguably inflated by the limitations on ownership of private equities, and may be deflated when a greater supply of large-cap technology stocks hit the IPO market. TWTR is a stalled-growth, turnaround stock which hasn’t turned yet, but trades at ~8x revenue. I believe that in the future we will look back on this valuation as being highly irrational.

Twitter is a niche product

I will start with the basics in analyzing the Twitter product. There are two main axes on which to categorize social media networks: Distribution and the media primitive.

Twitter and Instagram are public or broadcast distribution. Snapchat and WhatsApp are private messaging. Facebook is semi-public (or semi-private) distribution, based on real life social networks.

Instagram and Snapchat use an image (and to a lesser extent, video) primitive. Facebook has mixed primitives. Twitter uses a short-form text primitive. (While it is possible to post images and videos on twitter, this turns out to not be the dominant usage, for various reasons, mainly because other networks are better suited to them and social media is a network effects business)

If one is not careful, it would be easy to draw a 2x2 matrix on which each of these products can be placed in their respective quadrants, giving them the appearance of having equal stakes in the modern communication landscape. Indeed, I believe many investors and analysts have done this, either explicitly or implicitly. However, there is no logical reason to believe that these forms of communication will be equal. Indeed, Twitter is now a mature platform which has achieved a small fraction of its competitors’ scale.

Using the above analysis, I’ve concluded that Twitter has two primary use cases:

1. Broadcast of information

Celebrities tweet the details of their daily lives. Widely followed institutions utilize Twitter to disseminate information, as a lighter-weight alternative to email distribution. Journalists use Twitter to augment their primary publications. Joke and novelty accounts post one-liners.

While these are fine use cases, there is a limited population that can use Twitter for broadcast, and a limited demand for consuming this content. Furthermore, Twitter lacks a unique value proposition in broadcast. Other networks also work very well for broadcast. The main differentiating factor would be the short form, which may work better for specific types of broadcast.

2. Exchange of ideas

Networks of like-minded users have formed spontaneously to discuss topics of niche interest. A great example of this is Twitter users who discuss finance.

This is an even smaller niche than broadcast.

The reason Twitter has failed to achieve Facebook scale can be traced back to its distribution and primitive. Text primitives by their nature attract a much smaller market than image based primitives, and this has been true in every media generation. Books and newspapers hardly have the same mass appeal as movies and television. The difference is more stark in the age of social media, where users have to create compelling text content, not just consume it.

As VIC’s resident cynic, I’ll point out the obvious. The mass market has no interest in discussing ideas. As Eleanor Roosevelt famously said, “Great minds discuss ideas; average minds discuss events; small minds discuss people”. The mass market is not filled with great minds.

Now, you might ask, why can’t Twitter users simply discuss other people? They could, theoretically, but Facebook turns out to be a far better medium for this because of its semi-private distribution and real-life network / real-name foundation. In the early days of social media, before the Facebook phenomenon, there were many social network ideas that failed to gain VC buy-in. Entrepreneurs would pitch their product demos with dummy data, and VCs would say, who cares if Joe Bloggs went to the baseball game today? But if Joe Bloggs is replaced by one of your best friends, that information suddenly becomes interesting and compelling. Facebook succeeded because it is semi-private, allowing the sharing of more intimate data, and because of its real name policy, creating a compelling context for otherwise mundane information. Twitter lacks both of these features.

Twitter also has a major problem with abusive and “negative” content. As a public medium, Twitter users have no way of ex ante filtering out random strangers making undesirable replies to their tweets. While some users brush off such comments, the vast majority of the population does not want to deal with such things and will simply opt out of the product. One could argue, Twitter should be able to solve this to some degree at the platform level, but they haven’t yet, which indicates that either Twitter is incompetently run, and/or that Twitter fears stifling debate which is crucial to a medium that serves an exchange of ideas.

Twitter growth has stalled

MAUs excluding SMS Fast Followers in the US (in millions) for the last 6 quarters are 60, 63, 63, 65, 65, 66.

MAUs excluding SMS Fast Followers in ROW (in millions) for the last 6 quarters are 211, 221, 225, 236, 239, 241.

Note that Twitter does not report DAUs or other engagement metrics. The market seems to have shrugged off the user plateau, but this seems extremely serious to me. The market still views TWTR as a growth stock because Revenue is growing 60% year over year. But realistically, how long can that continue without user growth?

A key question is how much room is there to improve monetization. There is reason to believe that advertising is close to capacity. Twitter recently added auto-play videos, which are the most intrusive form of advertising and particularly intrusive for a text medium. Anecdotally, users have reported significant increases in ad impressions in the weeks leading up to and post the end of the quarter.

Twitter likes to tout its non-logged in users, but they have not shared much information on that usage and I am highly skeptical that these users have significant value. I developed an iOS app for another social network and non-logged in users engage in much smaller proportion to the overall network proportion of non-logged in users to logged-in users. The reason for this is fairly simple, most non-logged in users are simply browsing the web and have clicked a link that leads them to a tweet on the twitter.com website. These should really be presented as unique visitors (UVs) rather implied to be a lesser form of MAU.

Twitter doesn’t report engagement metrics (DAUs, timeline views, tweets per user). In fact, Twitter stopped reporting timeline views in Q1, and they did not replace the metric with any other appropriate metric. The logical reason for this is that the timeline views metric paints an unflattering view of overall engagement. Consequently, there is no reason to believe that engagement is increasing significantly or will do so in the future. If TWTR Revenue growth begins to slow, growth investors will sell and value investors will not step in until the multiple is a fraction of where it is now.

The turnaround will not succeed

Many investors including the high-profile VC and early twitter investor, Chris Sacca, hail Jack Dorsey as Twitter’s savior. Steve Jobs seems to have ignited the myth of founders returning to save their companies, and Jack running Twitter and Square simultaneously has reminded many of Jobs running Apple and Pixar. But this isn’t really the type of situation where you simply apply pattern matching to reach a conclusion.

Apple, from the very beginning, was very much a company driven by Jobs and the culture he created. Apple had no natural advantages and succeeded in a highly competitive market by making good products. Twitter, on the other hand, seems to be an accident of history; in the right place at the right time; a terribly-run company that has survived on being early to market, network effects, and good marketing (all 3 of these factors, btw, are gone now). Some might call that a “clown car that fell into a gold mine”. It’s not even clear how much of a role Dorsey played in the initial success of the company; this has been the contentious subject of some press articles and books, such as Hatching Twitter. A NYT article adapted from excerpts of the book can be found here:

http://www.nytimes.com/2013/10/13/magazine/all-is-fair-in-love-and-twitter.html?_r=0

Dorsey often tried to act as if he were in control, posturing that his actions were all part of a bigger plan, but employees saw him frequently pacing in frustration around South Park. He also habitually left around 6 p.m. for drawing classes, hot yoga sessions and a course at a local fashion school. (He wanted to learn to make an A-line skirt and, eventually, jeans.)

With Dorsey’s return, Twitter’s PR team went on an offensive to explain that Dorsey had matured in his time away from Twitter (and at Square), that he was a changed man. It may well be possible that Dorsey is a better leader than he was previously, but this does not magically solve the problem of Twitter’s being fundamentally a niche product.

Dorsey’s return of 1/3 of his shares (approximately 1% of the company) to the equity pool was lauded as a selfless move to save the company. While no doubt it was smart, it seems likely that if a turnaround succeeds, the company will likely grant Dorsey compensation well in excess of 1% of the equity. If it fails, his remaining shares aren’t worth much anyway. Either way, the vast majority of his wealth was already outside his TWTR shares. To me, the bigger issue is not whether Jack is a nice guy, the issue is that Twitter has been losing its best people because the product has stalled, and the product has stalled because Twitter can’t get better people. There have been a number of high profile departures, and morale is low beneath the headlines at all levels of the company.

New products do not appear to be gaining enough traction. Periscope, while interesting, is likely to be a niche product, with wider applications being better suited for other networks like YouTube. Moments is unlikely to succeed. Twitter seems to have realized that its core product has a niche appeal, and that it has to release a fundamentally different product to gain mass market appeal. And that product (at least for now) is Moments. Moments is a curation of tweets from “Broadcast” users. Moments is retaining many of the problems outlined above in my analysis of broadcast usage. Furthermore, the curation itself seems like it adds little value; and with many of Twitter’s users being low-engagement users, it’s not clear how much value could be added through curation even theoretically.

To be clear, there will be additional changes coming, the most prominent of which appears be a gradual phaseout of the chronological timeline. This is a risky move and while it’s unclear exactly how this will play out, I do not view this as a solution to the niche problem, which is created by the distribution scope and primitive data type (text) of Twitter, not the order in which the data is presented.

Twitter does not have a track record of introducing successful products. Most of Twitter’s defining features, such as hashtags, retweets, @-replies, and link shortening were introduced organically by Twitter’s users. Twitter’s user base has proven to be far more innovative than Twitter itself. The problem the company faces today is that its user base has no interest in Twitter’s commercial success or innovating toward that end. Furthermore, Twitter product has evolved organically into its current form through input by its power users. Those power users have fundamentally different needs and preferences than the mass market, and those users are highly vocal and create all the content on the network. As a result, Twitter has an unbridgeable gap between the present and its desired future. I view this as being similar to the problems faced by reddit today.

We know that many more users (likely a 2-3x of current MAU) have tried the product and are inactive. In an age of a thousand apps competing for attention, how likely is it that users will try Twitter a second time, having already given up on it? Twitter has a massively uphill battle to fight.

Twitter is too expensive to acquire

At a market cap of almost $20 billion, TWTR would need a bid of at least $25 billion to be acquired. That’s a lot of money, even in today’s environment. Realistically, there are only two companies that could acquire Twitter: Google and Facebook. No one besides Larry Page and Mark Zuckerberg knows for sure if such an acquisition would happen, but I find it unlikely.

I believe that at some point in the past, Google had a genuine interesting acquiring Twitter. A few years ago, Google management had become obsessed with social and the idea that social was a gigantic onslaught that could destroy the company. I think they’ve realized over time that their core search business is relatively safe from social, and that it’s better to innovate and win new markets than try to compete in already-lost markets. Google historically acquires companies that will create value when integrated into existing product platforms. Twitter would do no such thing. While Twitter real-time data is useful as a search signal, Google already has a partnership with Twitter to provide such data.

As for Facebook, there is no need to acquire Twitter when Facebook is destroying them in the marketplace. Furthermore, if the quote attributed to Zuckerberg is accurate, it’s likely that he wants nothing to do with them. Facebook has proven to be a smart acquirer, but it has always bought very fast-growing companies for high prices. Twitter isn’t growing its users at all!

Valuation

At $27.50, Twitter trades at 8x run-rate revenue and 30x Adjusted EBITDA. However, Adjusted EBITDA is not good to use because stock-based compensation exceeds Adjusted EBITDA. In an acquisition scenario, perhaps SBC could be cut significantly with layoffs, but as an independent company the business is unprofitable with its current structure.

Catalyst

• Growth stalls and growth investors exit
• Additional IPO supply causes rotation out of TWTR