Spark Networks (AMEX: LOV) is a company which for various reasons has been largely ignored by the market and whose business (which is made up of more than 20 independent websites) is easy to mistake as permanently declining rather than temporarily stumbling. I believe that at current levels, Spark presents an extremely compelling investment with upside on a conservative basis of 50%, on a completely reasonable basis of 100%, and on an optimistic basis of a whole lot more. Importantly, LOV has a very firm margin of safety in JDate, one of the company’s wholly owned online dating websites. I believe the value of that website alone is higher than the enterprise value of the entire company and I think the unique nature of the business guarantees its value for a very long time. Additionally, the company has a rock solid balance sheet, has been buying back huge amounts of its stock, has a superb new CEO, and possesses ample and lucrative growth opportunities. Lastly, LOV is a pure play in online (and some offline) dating [though incremental ad revenue could start to appear]. As such, it is just about as great of a recurring revenue, non-cyclical, and recession proof company as one can find. Although I certainly don’t make bets about what sort of companies/stocks the market is going to like, I can definitely see in this financial and economic environment people turning their attention to such enterprises.

Business Description and Commentary:

Spark Networks operates a variety of online dating sites. Spark’s business strategy is unique amongst the major players (Yahoo! Personals, Eharmony, match.com, etc.) in the online dating space. Instead of hosting one site aimed at the general population that allows people to limit their potential matches according to various screening criteria, Spark operates numerous niche sites aimed towards very specific ethnic, religious, racial, geographic, and other special interest communities. Their major sites include JDate.com, BlackSingles,com, and ChristianMingle.com, but the company has a host of others. You can go here: http://www.spark.net/sites.htm for the whole list. However, despite their current focus on developing the niche verticals, Spark still operates certain general market sites such as AmericanSingles.com and date.co.uk that are hangovers from the company’s strategy at an earlier time in its history.  I’ll discuss later the present situation of those sites from both a financial and strategic perspective. For organization/presentational purposes, Spark segments its business according to the following classification: 1) Jewish Networks (JDate American and JDate Israel); 2) General Market Networks (American Singles, etc.); 3) Affinity Networks (all the non-Jewish niche sites; 4) Offline and Other businesses (HurryDate – an offline speed-dating service; ad revenue from the sites; events hosted in association with the sites, such as JDate trips to Israel, etc.)

The sites themselves generate that vast majority (97%) of the company’s revenue. The source of that revenue is monthly subscription fees. Though membership on any of the company’s various sites is free and allows one to post a profile and view others’, only paying subscribers have certain privileges, most significant amongst them the ability to contact other subscribers. Without paying for a subscription, the sites are nothing more than galleries of available singles.

I highly recommend anyone interested in going here: http://www.spark.net/documents/CorpPresentation110707forwebsite.pdf for the company’s latest investor presentation. It provides a good overview of Spark’s business.

Currently, LOV trades at $4.00 a share. That gives it a FD market cap of $107mm (26.8mm shares on a diluted basis). It has $8.3mm in cash and CE, and no debt. Thus, an EV of $99mm. For that $99mm, we get the following figures: $10mm in LTM EBITDA; just under $15mm of LTM adjusted EBITDA (adding back stock-based compensation); $13mm in LTM FCF. For two reasons FCF has been and will remain for a few year very close to adjusted EBITDA, and a good deal higher than net income: 1) The company has about $40mm of NOLS as of September 30, 2007. I expect that for the next few years, the company will be paying taxes at about a 10% rate; 2) due to the extremely minimal CAPEX requirement of the business (about $.5mm a year (of course, the company has done, and may continue to pursue, discretionary acquisitions of certain businesses)), D&A outstrips CAPEX. For 2008, I think the numbers will swell to north of $14.5mm for EBITDA, $18.5mm for adj. EBITDA, and $17.5mm for FCF. For EPS (which I consider less important, especially in this case where massive and unabated cash flow should be the driver of valuation), I see about $.40 on a reported basis, and about $.50 on an adjusted basis (again, adding back stock-based compensation). 

Regarding valuation: I’ll present two valuations 1) basic multiples; 2) basic DCF.  In both cases, I’m aiming to make conservative valuations that show easily attainable upside, and demonstrate very well protected downside. I think better upside can be expected through multiple expansion, better than low-ball estimates, or a combination of the two. I think really great upside can come from some really superb numbers that show certain segments are growing strongly.

Basic multiples: Let’s assume 10.0x multiples for all metrics. Cash on balance sheet included in per share value. 

 

Basic DCF: Let’s assume the company can generation $15mm-$17mm in cash for the foreseeable future and discount it at various rates. Cash on balance sheet included in per share.

 

I think the combination of those two valuations support the 50% upside ($6.00 stock) with conservative assumptions.

Regarding FY08: My forward estimates for FY08 are not based on any heroic growth assumptions. I expect total revenue to drop in FY08. From about $64.5mm this year (it depends how Q4 ends up), to about $62mm next year. Of course, things could be better – considerably so. I’m just not banking on it. My estimates are based on three factors: 1) revenue expectations for the individual segments; 2) contribution (revenue less direct marketing expenses) for the individual segments; 3) unallocated operating expenses. I’m going to go through the individual segments later, so there I will address revenue and contribution estimates for FY08. Here, I’ll address operating expenses.

For operating expenses, I foresee modest improvements of a couple million dollars or so. If one looks at operating expenses for LTM, its actually a bit misleading. There’s two reasons for that. One, the company took a $1.8mm write-down in the value of certain goodwill. Of, course, that gets added back to EBITDA, etc. Secondly, during this year, the company underwent what it calls a ‘scheme-of-arrangement.’ In short, what this means, is that the company re-incorporated itself as an American entity. Formerly, it was a European incorporated entity and traded on the Frankfurt stock exchange. Only its extremely illiquid ADS’s were traded in America. Of course, those two factors only greatly exacerbated a relatively small company’s chances of being noticed by investors. So during the course of this year, the company spent more $2.4mm in re-incorporating itself, delisting from the Frankfurt exchange, listing its common shares on the AMEX, and taking the requisite initial steps to meet Sarbox. [This whole process was completed in July. Long term, this is great for the company as it should eventually allow it go attract greater investor attention. Currently, there still isn’t  much liquidity in the stock, but things are improving.] All the expenses related to ‘the scheme’ were tagged under G&A. So if you look at the financials, ignore what appears to be an increase in G&A expenses. [As an aside, I think the appearance of an increase  in expenses without a corresponding increase in revenues has contributed to investor negligence/negativity on the stock and helped to bring about the current opportunity]. Here are the G&A expenses for the last 7 quarters:

 

Backing out the $2.4mm in non-recurring expenses, over the last three quarters the company has spent, on average, less than $4.9mm on G&A. However, I think the company can do a lot better than that. For one, the company was doing a lot better during the last two quarters of FY06 when they started to focus on cost containment, and I think there’s some added G&A associated with, but not directly classified under, expenses for ‘the scheme.’ Secondly, the new CEO (appointed in February), Adam Berger, is extremely focused on cost containment through the entire business. That is really something he has emphasized – trimming the business and making it more efficient. It something to keep in mind when considering operating expenses for 2008. I think my projections will likely prove conservative. So far, Mr. Berger has surprised me with his continued ability and willingness to cut costs so aggressively. Lastly, in addition to attracting greater investor attention, another reason for undergoing ‘the scheme’ was to save $600,000 a year in various expenses related to European incorporation. So to be conservative, I’m forecasting $4.7mm a quarter for G&A. For the other operating expense categories, I have no problem forecasting forwards Q307 number. In customer service, technical operations, and development, Mr. Berger has already proven his ability to trim spending. He very well might do more so, but I’ll assume he just keeps things where they are.

Jewish Networks:

The gem within Spark is JDate – in particular the American version. JDate is the premier online dating site for Jewish singles. The absurdly elevated status that JDate has within the Jewish singles culture endows it with a moat – a wide moat with alligators swimming around it. I’m serious. Among Jewish singles, JDate is synonymous with Jewish dating. The site is regularly referenced and discussed in a variety of cultural mediums – both specifically Jewish and not. Very popular TV shows such as House, Entourage, and South Park have all made reference to the site. For a certain sub-population, JDate has become absolutely ubiquitous and irreplaceable for their dating lives.

As of Q307, the Jewish Networks segment of Spark had 94,072 subscribers. That figure is aggregate for both the English and Hebrew JDate as well as certain smaller Jewish sites. Before Q106, the Jewish sites were still in considerable growth mode. However, they then transformed into steady and slow growth mode. The material decrease in subscribers in Q207 was the result of a 14% increase in the one month subscription price to the JDate. Management anticipated a short-term drop in subscribers, but felt that from a long-term revenue and contribution perspective, the move made sense. Q307 already showed the sites regaining footing. Honestly, I see no reason why the sites won’t return to Q107 numbers within 2-3 quarters, and continue their slow but steady growth from there.

 

The cultural status and importance of JDate do not only ensure a rock-solid subscriber base. It also allows JDate to spend a pittance on marketing for the site. The vast majority of marketing for the site is through word-of-mouth. Greater than 80%% of all traffic to the site is organic – that means it comes through the people actually typing in www.jdate.com into their web-browser – not through clicking a paid for Google link or a paid ad appearing on another site. As a result, the contribution margin (revenue less marketing expenses) for the Jewish Networks are astounding, and the company has only continued to improve upon them, both in absolute dollars as well as margin percentages. Additionally, the company is committed to further experimenting with both subscription fees for the Jewish sites, as well as the minimum amount of marketing dollars it needs to spend.

 

According to the latest Q3 run rate, the Jewish Networks are generating almost $31mm in contribution margin a year. I believe there is significant upside to that number from the following four factors/scenarios: 1) a return to the subscriber levels seen in Q107; 2) further price increases in subscription rates; 3) higher conversion rates of members to subscribers. The Jewish networks have well over 400,000 members, but only 94,000 subscribers. Figuring out how to convert a higher proportion of people who have registered for the site into paid users is key to driving up subscription numbers, revenues, and margins. CEO Berger has stated repeatedly that this is an attainable goal of the company, and that they have specific initiatives in place to realize their goals; 4) attracting more members/subscribers from the target population. A national Jewish Population Survey (used by Spark) estimates that there are 1.8mm Jewish singles older than 18 in America. Yet the American JDate site has only about 75,000 subscribers (about 375,000 members). Every 1% increase in subscriber penetration within the target community equals about $5mm in revenue. Obviously, numerous factors are at play that would allow or prevent JDate from making further inroads of a significant size, but I think continued modest and steady growth for the foreseeable future is highly likely. All those factors, combined with the run rate based on the Q307 number, lead me to expect, conservatively, better than $32mm in FY08 contribution from the Jewish Networks (compared to $29.4mm for LTM).

Valuation of JDate as a standalone entity: I believe the Jewish networks alone are worth considerably more than the EV of the entire company. I think the easiest way to show this is on an EDBITA/Adj. EBITDA/FCF multiple basis. The company doesn’t disclose how its unallocated operating expenses should be spread across its business segments. I thought that a fair way to approximate the amount was to do it according to subscribers. Accordingly, the Jewish Networks get about 46% of all the unallocated operating expenses thrown their way. They also get that ratio of D&A as well as stock-based compensation.  For LTM, the Jewish networks produced $14mm in EBITDA, over $16mm in adjusted EBITDA, about $15mm in FCF. I think all of those numbers should improve by about $2.5mm-$3mm in 2008 for the aforementioned reasons of increased revenues and contribution margins in the Jewish networks coupled with cost containment in unallocated operating expenses. Of course, even at its LTM numbers, the Jewish Networks business is absurdly cheap (7.0x EBITDA, 6.2x Adj. EBITDA, 6.6x FCF) for a business with an incredibly strong moat, ridiculously high margins (90% gross margins, >40% EBITDA and FCF margins, and 50% Adj. EBITDA margins) and with ample and diversified growth opportunities [see discussion later on incremental ad revenue potential]. I think to assume a 10x multiple on any of those figures for a business with such characteristics is still very conservative. Doing so for my 2008 estimates and including company cash, yields the following per share values.

 

In short, this is what I meant in the introduction that JDate alone – which makes up some 75,000 out of the 95,000 total Jewish Network subscribers – offers a very solid margin of safety. JDate is easily worth $5-$6 a share or more, with the entire Jewish Networks business probably worth upwards of $7. Even assuming no growth, no other businesses, and even meaningful subscriber deterioration in the Jewish Networks (something which I think has almost no chance of happening), still leaves your $4 extremely well protected.

Obviously, LOV would not be selling at the price it is if all segments of its business were operating like and had the various characteristics of the Jewish networks. So there is a problem here – the General Market Networks. I’ll explain that problem and I’ll also try to explain why I believe that the problem has merely opened up the opportunity that presently exists, but why long term, for the company as a whole, it doesn’t pose any meaningful risk of destroying the value that exists.

The General Market Networks is made up of up sites such as AmericanSingles.com and date.co.uk. It is the segment of Spark’s business that is not directed towards specific niche verticals. Rather, the sites are aimed towards the general populations, similar to Eharmony, match.com, and the countless other generic dating sites. As the numbers below show, deterioration in Spark’s general Markets membership base has been severe and shows no signs of abating. While initially this was something Spark tried to fight, CEO Berger has decided to take a different approach. He has explained that the company realized a few things about the general market dating sites: 1) there is a tremendous amount of competition from well funded peers such as match.com and Eharmony; 2) all companies have to “buy the top-line” in the business by outspending one another on marketing to acquire subscribers; 3) the general online dating market is well saturated and leaves little, if any, room for growth; 4) due to the above factors, coupled with the lack of stickiness in the general market as compared to the niche ones, no generic site will ever be able to replicate the cultural and financial dominance of a site like JDate. Berger’s current focus is not to grow or even maintain the top line – rather, he aims to maximize contribution margin from the segment so long as a subscriber base does exist. So far, he has been successful in his goal. Even while subscribers and revenues have deteriorated unabated, contribution amounts have recovered and leveled out. In fact, Q307 featured the highest contribution margin for the General Market segment in company history. Despite having seen subscribers and revenue fall by over 50% since Q106, contribution from the segment has fallen only 31% during that time period. In short, Berger is trying to maximize the economic value of the segment, even while it dies a slow death.

 

One might raise the question of why the company doesn’t just shut the doors of the General Market Networks. For one, the company still produces $2.2 in contribution. I don’t know how much of the unallocated operating expenses they could shed if they got rid entirely of the site, so I can’t say if on an operating level it is profitable or not (or at least contributes to the profitability of the company as a whole). The CEO has explained the issue interestingly. He has saud that company currently uses the General Market Network sites as marketing tools for their niche sites. Say, for instance, that someone registers for AmericanSingles.com and classifies themselves as a Mormon. Spark would then have an e-mail sent to that person telling them about LDSSingles.com.

I see results for the General Market Networks becoming increasingly poor in 2008. According to Q3 numbers, their run-rate contribution is $8.8mm. I’m expecting about $5mm for next year.

Ultimately, I see the demise of the General Market Networks as a positive. The segment has never been especially or at all profitable (when it had a much larger subscriber base the company was spending very heavily to acquire those subscribers), the general market for online dating will forever be extremely competitive, and devoting less resources and energy to the segment means more resources (i.e. cash) for shareholders, and more energy (i.e. management focus) on the profitable and growing parts of the business – the Jewish Networks and the Affinity Networks.

Affinity Networks is comprised of Sparks’ various niche verticals excluding the Jewish sites. A wide variety of religions, ethnicities, and nationalities are represented. Some of Spark’s most successful sites (in terms of membership base) in this segment are those devoted to LDS, enthusiastic Christians, and African Americans. Many other sites have respectable numbers as well. Spark’s stated goal is for one or more of the sites within these segments to become the dominant site within the niche vertical it represents – just as JDate had become within the Jewish world. Developing such a marquis site will produce the same tremendous benefits as JDate – a highly predictable and strong revenue base coupled with extremely high margins. However, even if amongst the various sites there does not develop another JDate, I expect the segment to keep growing and improve its profitability as it does. The numbers for the Affinity Networks segment present the picture of a growing business (in contrast to the stable Jewish Networks and the declining General Market Networks) for which management is trying to identify the right amount of marketing spending.

 

 

For 2008, I see Spark continually improving upon Q307 numbers. I think $7.5 in contribution is a reasonable expectation. (The number is partly the result of a recent 20% hike in subscription fees at enthusiastic Christian site. Management has indicated it is exploring further price hikes in other Affinity Network verticals).

I think it makes good sense to divert some of the spending formerly allocated to the General Market Networks to this segment. Management has certainly been doing this. Obviously, for the moment at least, we are hardly seeing inspiring growth here. But I think this is a lot about critical mass and a lot about reaching the right people with the right amount of marketing dollars. The company largely devoting itself to realizing success in this segment, and I think that if the numbers start to shine here, we could see the large incremental upside in this stock. If management become able to identify one of the sites gaining truly marquis potential, the upside could really be tremendous. Whatever the case may be, I think that by buying LOV for significantly less than the value of the Jewish Networks alone, we’re getting better than a free call option on the future success of this segment.

This segment is made up primarily of three things. 1) HurryDate; 2) Various trips and events associated with specific sites (usually JDate); 3) Incremental ad-revenue. I’ll address briefly HurryDate and incremental ad-revenue.

Hurry Date: Spark bought HurryDate for $2.25mm in February of 2007. HurryDate organizes speed dating events. People pay money to participate in an event, which is usually held at a bar or restaurant. For those unfamiliar with the phenomenon, speed dating involves a lot of people getting together and then having a bunch of 5-7 minute ‘dates’ in a single night. Basically, it offers a chance to meet a lot of people in a short amount of time. Whatever you think of it, this thing is pretty popular.  As a standalone, HurryDate had to attract people interested in speed dating to come to its site and get them to sign up for their events. Under Spark ownership, Spark can use the HuirryDate name, website, and infrastructure as a way up-sell to their already existing membership base the service that HurryDate offers. Only very recently (within the last month or so), did Spark begin directly linking HurryDate events to their websites (i.e. on JDate now there are links to sign up for a JDate sponsored HurryDate event). I think there is potential here for meaningful revenue contribution. Over the next quarter or so, as HurryDate is more fully integrated into Spark’s dating sites, will we see if synergies can actually be realized.

Incremental ad-revenue: this is an area where meaningful revenue contribution is extremely likely to come. CEO Berger has, on numerous occasions, been very candid about the fact that previous management entirely missed the boat in appreciating the advertising potential of their sites. He spoke of the issue almost entirely in the context of the American JDate site, so I will as well.

Firstly, the demographics of the 375,000+ membership base are incredibly attractive to advertisers. 79% are college educated; 69% have incomes over $50K; 34% have incomes over $100K. The age breakdown is also extremely desirable. See this link: http://newscenter.jdate.com/newscenter/members.htm for full demographic information (some of the data is slightly different than that reported in the most recent investor’s presentation – I have given precedence to the later data, but the posted link has more complete data not presented since). Secondly, members have submitted huge amounts of personal information to the site. As the CEO recently said at an investor’s conference, an online dating site is perhaps the only place where people will willfully answer very personal questions without thinking twice about it. Income, education, religious views, political views, life plans, eating and drinking habits, hobbies, ideal vacations, etc. etc. etc. The sites ask it all and the members answer it all. This database obviously allows for much more targeted advertising than web-sites can traditionally offer advertisers [think about how excited everyone is about the advertising potential for Facebook because of the information database]. Related, the community of the site itself is, by its very nature, highly targeted/specific, and consequently very desirable to advertisers. What this all adds up to are many sites  that have tremendous advertising potential based on their various unique characteristics and aggregate 400mm page views a month. For a company of Spark’s size and current revenue stream, the contribution from taking advantage of this opportunity could offer huge upside.

For 2008, I’ve assumed $1mm in contribution from this segment. I think that may prove extremely conservative, but I have no guarantees that HurryDate will start to deliver or that the CEO will fulfill his promises and start to monetize the advertising potential of the site. But if he does, I could see that as being a near term catalyst.

The management strikes me as truly first rate. This is a critical factor in my confidence in the company. I highly recommend anyone who is interested in this idea to listen to conference calls, as well as the presentation from the RBC Technology Conference (all available at http://www.spark.net/investor.htm). The CEO, Adam Berger, came over from WeddingChannel.com in February of 2007. He ran that company for seven years before it was bought out by theknot.com in 2006. In listening to Mr. Berger, I am struck by his willingness to admit the company’s problems and accept that the burden lies upon him to fix them. He readily recognizes that the previous management rested on their laurels by letting JDate do its magic and letting the other sites flounder. He admits that previous management entirely overlooked the advertising potential of the sites. He is candid about the need to make the company operate more efficiently, trim fat in operating expenses, and maximize the impact of every dollar of marketing expenses. His commitment to improve the company’s operations as well as to grow it strongly is manifest in both his words and actions. As mentioned, there have been very meaningful cost cuts in both direct marketing as well as operating expenses. The company has re-oriented itself entirely around the niche verticals, and is letting the story of the General Market Networks play out in the most economically beneficial manner possible. Tuck in acquisitions such as HurryDate are enabling the company to take advantage of its infrastructure and membership base. Ads should soon start to appear on the dating sites.  

Additionally, Mr. Berger is very well incentivized to perform. Along with a performance bonus that varies widely depending on performance, Mr. Berger holds 1,300,000 options granted to him as part of his employment agreement. Currently, all of these options are under water. None of them begin vesting until February 2008, and they will not be completely vested until 4 years later. The average strike price of the options is $5.70. Clearly, therefore, Mr. Berger should be plenty focused on maximizing shareholder value. (Indeed, there are 4.3mm options currently outstanding – the average strike price for these options is $6.19. So all insiders have very good reason to align their interests with ours.)

The company has been extremely prudent and shareholder friendly in allocating their capital. In the last two years, the company has made only two meaningful acquisitions. They bought LDSSingles for $2.2 mm in May 2006 and HurryDate for $2.25mm in February 2007. LDSSingles was a pure tuck in acquisition. They were simply adding more niche-oriented sites under their umbrella. HurryDate was a bolder move, but one I think was wise.

Mr. Berger has made it clear that he is interested in doing further tuck in acquisitions of already existing niche online dating sites – but only, he stresses, if the price is right. I’m extremely encouraged by the fact that the CEO recognizes that his job is not empire building, but rather maximization of economic/shareholders value. Over and over again – through various cost cutting measures, in his willingness to let AmericanSingles fade, in his apprehension about acquisitions, etc. – Mr. Berger makes it clear that he realizes his job lies in realizing the latter of the two goals.

Since the company has not been spending its cash on acquisitions, they have been deploying it towards stock buybacks. Since the buyback plan was launched last November, the company has bought a full 15% of its shares outstanding. Last quarter alone, the company bought back 4.1mm shares at an average price of $4.20. I expect the company to continue buying back shares as it languishes at these levels – especially as the company continues to generate great amounts of cash and is unable to find attractive (and attractively priced) acquisitions.

There are a few risks here, but only one significant risk relates to the actual core performance/intrinsic value of the company:

1)      Poor liquidity of stock

2)      Market focuses on further deterioration in General Market Networks

3)      Ownership concentration – Great Hill Partners owns 30% of the stock

4)      Fortunes change radically in Affinity or Jewish Networks (extremely unlikely I believe)

There are really no specific, near term catalysts for this idea. It’s just a good business that I think is really cheap. Looking out a bit (at some point in time), there are foreseeable catalysts:

1)      Pick-up in liquidity of stock and investor attention to the company

2)      Continued strong performance in Jewish and Affinity Networks – the valuations will start to look absurdly low and people will notice

3)      Material success with HurryDate and or rollout of ads on sites

4)      Strategic or financial buyout