|Shares Out. (in M):||63||P/E||-||-|
|Market Cap (in M):||800||P/FCF||0.0x||0.0x|
|Net Debt (in M):||0||EBIT||-17||-6|
THE ACTIVE NETWORK (ACTV)- SHORT IDEA
Update Note: This idea was my application to VIC submitted on 10/27/11 and accepted on 11/29/11. Since then, ACTV is trading down 12%. As of Q3 '11, a new low synergy acquisition ($21m) was announced and amended operating metrics were reported in an 8-K issued on 11/4/11 that may have cured some aspects addressed in Bear Case #2.
Active is a 12yr old, 39 company roll-up that processes payments and registrations in multiple verticals through the web, retail point-of-sale systems and call centers. They had a faddish IPO in 2011 b/c mgmt sold the Street on a Salesforce.com-like cloud computing story with OpenTable-like network effects. In reality, the majority of operating activities and revenue are derived from very competitive and commoditized desktop point-of-sale software and call center registrations. Operations are complicated by acquisitions that are largely unrelated, hard to assimilate and demonstrate few network effects, all factors which will constrain EBITDA margin gains. Using public data not supplied by mgmt, I was able to estimate the impact of new contracts on overall registration volumes in the lower margin gov’t fishing/hunting license segment. I then deduced that organic growth may have stopped in their non-gov’t, higher margin registrations due to competition and pricing pressure. The acquisition history itself suggests that mgmt prefers to pay for customers. There is additional evidence that mgmt does not always act in the interest of shareholders. Although consensus est. are $22ps based on 16X 2013 EBITDA via margin expansion delivered by cloud computing leverage and growth due to network effects, I believe a forward multiple of 9.5X EBITDA (which is still generous) better characterizes Active’s real business model. Based on slower growth and modest EBITDA margin improvement, I estimate intrinsic value of $6.22ps which is 50% lower than the current price.
Easy to understand core business model w/ moats and little competition. Active claims a large TAM accessible through network effects on 47k customer base. Growing cloud-based SaaS providers (e.g. CRM, OPEN, RNOW, ACOM) have operating leverage, first-mover advantages and are poor short candidates irrespective of high valuation.
One-time contract adds in gov’t license biz hide slowing growth in core registrations.
Gross margin was 79.2% in 2003 & 56.5% in 2010 b/c biz model keeps changing.
Acquisitions don’t create network effects and EBITDA margins will not improve.
Competition is discounted by mgmt and sell-side reports.
Why the Opportunity Exists
The company claims a large TAM but nobody correctly characterizes the competition.
The IPO was done on a cloud computing/network effect claim which will rollover the investor base once the growth slow down is observed over the next 3 quarters.
Recent acquisitions have been vertical expansion and have not been accretive to existing business segments. They have low success growing the business organically.
Active developed niche software for endurance race directors in 1999. They merged with RecWare (parks & recreation apps) and bought eteamz.com (team and league sports apps) and grew rev. to $4.4m in 2001. A few acquisitions later, rev grew to $15.2m in 2003 and they filed an S-1 in 2004 but did not complete the IPO. Then they branched out into camps, education and golf through 2007. Then in 2008 and 2009, they made several acquisitions in government contracts for hunting/fishing licenses for 23 states and campground reservations for 18 states and federal parks. In 2008, they expanded their reach into business event registration. In 2011, they acquired 3 companies that provide software for churches. The company appears to draw no boundaries as to verticals they believe they can conquer.
Today, 48% of revenue is derived from online registrations managed by the company (non-government net registration revenue), 23% from fishing & hunting licenses and campground reservations (government net registration revenue), and 29% from online advertising and one-off technology services (technology fees and marketing services). Founder Dave Alberga is still CEO. Several sr. mgmt has long tenure with the company. Matthew Landa has been the President since 2002. Jon Belmonte served as COO from 2000-2011 and is now the Chief Media Officer. Kourosh Vossoughi has served as General Counsel since 2000. Matt Ehrlichman came to the company through an acquisition since 2007 and is now the Chief Strategy Officer. Alberga only has 185k shares. Ehrlichman is the only officer with a meaningful holding at 2mn shares.
BEAR CASE SUMMARY - ACCIDENTAL CONGLOMERATES ARE BAD BUSINESSES
Mgmt’s core function has been raising capital and spending $425m doing 39 mergers and acquisitions. Other than online advertising and some one-off software consulting, Active has developed no other revenues other than those gained through acquisition. Their mission statement is “We power the world’s activities and connect people with the things they love, want and need to do”. This mission is so ambiguous that it has no meaning. While there are lots of “activities” in the world every day, Active claims ability in a minimum 12 industries of “activities”. Historically for Active, the activity was an athletic endeavor. Now, the “activity” could be registering for summer camp, buying a fishing license, paying a tax bill, making an online donation, or reserving a camping site. At best, these services have in common an e-commerce function that provides software to industries that are too small to afford enterprise applications (e.g. NetSuite, PeopleSoft, SAP).
BEAR CASE #1 - FEW NETWORK EFFECTS
Mgmt cites network effects to drive future growth. Despite 70m annual transactions, organic growth is remains slow.
7% of endurance registrations originating through Active.com is their best sign of a network effect. OpenTable.com drives 63% of reservations. That is a network effect!
Taken together, the 12 industries they target exhibit low overlap and preclude the formation of network effects. Stated another way, deer hunters don’t do triathlons.
Management also cites the 60% penetration rate in triathlon as evidence of network effects. However, this penetration rate is due to a first-mover advantage 13 years ago in a tiny niche.
BEAR CASE #2 - ORGANIC GROWTH IN NON-GOV’T REGISTRATIONS IS SLOWING
Management defines the TAM as 800,000 organizations and 1.7b annual registrations. The organization counts have slowed from 8% growth in 2009 to 5% in 2010. As of Q2 2011, mgmt announced they will report organization counts only annually. The company employs 478 people in sales and marketing (100% effort is directed to new clients, not registrations) and reducing visibility on their effectiveness is highly suspicious.
Instead, registrations must be relied upon to measure the organic growth rate. However, over half of registration growth in the last 3 years has come from acquisitions of “Registrations - Campgrounds” and “Registrations - Fishing & Hunting”, whose data are publicly available.
I compiled fishing and hunting license sales from public records and then included the registrations added through 3 new state contracts to show that non-government registrations have slowed from 12.1% in 2009, to 9.1% in 2010 and are estimated at 4.1% in 2011.
Government Registration Growth
Implied Non-Government Registration Growth
|Total Registrations (+)||25,074||65,461||70,183||77,350|
|Gov't Registrations (-)||0||37,358||39,535||45,450|
|Non-Gov't Registrations (=)||25,074||28,103||30,648||31,900|
This decline is likely attributable to a reduced growth rate in the number of client organizations which is partially offset by higher utilization in the organizations that still use their software. Assuming that there is a natural migration of about 3% of registrations online (from paper or telephone methods), the rate of organization growth at Active could be truly anemic at 1% since I am forecasting a 4.1% increase in non-government registrations in 2011.
To investigate further, I spoke with mgmt and suggested that it was difficult to sort out the gov’t vs. non-gov’t registrations. Mgmt volunteered that California was going to add about 5m transactions a year for hunting and fishing. My estimate was 4.1m. Curiously, the addition of California (in April 2011) and Ohio (in March 2011) was never announced in any press release. I had to find internal state publications that discussed the roll out of the systems. Since those 2 states alone increase registrations by about 8.5% in 2011 compared to 2010, it was certainly worthy of a meaningful press release by Active! Instead management has elected to put out press releases that have negligible impacts on registration growth in order to provide the market with the impression that the core business is growing registrations. Mgmt can't really draw attention to the source of this registration growth because if they did then the lack of non-gov't registration growth would be apparent. In fact, the California contract is only briefly mentioned in 2 of 4 sell side reports and no company filings. Ohio is mentioned in zero sell side reports and no company filings.
Moreover, when I drill down on Q2 2011 registrations and cross reference to public data I located on actual license sales in 1H 2011 in California, it actually looks like Q2 2011 non-government registrations had ZERO growth relative to Q2 2010. The addition of California and Ohio can explain 100% of the increase in registrations. The only fact that would negate this finding would be if a hunting/fishing contract with a state was lost and not disclosed it in which case non-government registrations would have grown to fill that void. This is unlikely given the 5 year or longer terms for the state contracts. The results from the next few quarters can confirm these findings.
BEAR CASE #3 - ELUSIVE SYNERGIES WITH POOR MARGIN EXPANSION
Even excluding the historical M&A investments, current R&D expenditures (including capitalized software) are about 2-3X higher than peers (CRM, CTCT, OPEN, ACOM, BLKB, MCRS, RNOW, EPOC, NUAN, SFSF, CNQR, DRIV). R&D as a % of revenue was 24.7%, 30.2%, and 27.4% in 2008, 2009, and 2010, respectively, where the peer avg is 13.2%. R&D as % of gross profit showed similar divergence to peers.
Mgmt has been vague on the progress and start date (2007 or 2009?) of the Active Works platform as a consolidation attempt to reduce duplicative R&D and unify the code base. Of note, most triathlons still use legacy software written in antiquated Cold Fusion language.
Software development is still geographically decentralized relative to peer companies. I also used the career search tool to more generally examine the locations of all available positions with the word “engineer” in the job title. This returned 52 open positions available in the following locations and numbers of positions in parentheses: Ballston Spa, NY (1), Bothell, WA (3), Burnaby, BC, Canada (6), Sichuan, China (2), Dallas TX (4), Draper, UT (2), Mississauga, ON, Canada (3), Nashville, TN (2), New York, NY (2), Sacramento, CA (2), San Diego, CA (17), and Singapore (4). In contrast 36 out of 44 jobs in “Information Technology” at Salesforce.com were located in San Francisco. For OpenTable, 4 out of 4 jobs were in San Francisco, CA. For business events competitor Cvent, all 11 technology jobs were in McLean, VA. Nearly all of the 60 positions at WebMD were located in Portland, Atlanta, and New York City. While competitor companies tend to be geographically concentrated for the purposes of software development, Active has yet to even concentrate its efforts in one or two locations. The complexity of migration to a new platform is likely being hindered by this poorly assimilated structure.
Acquisitions remain geographically unconsolidated suggesting that SG&A will remain high. There have been 4 acquisitions in 2011 thus far suggesting that the past will look more like the future.
BEAR CASE #4 - LOTS OF COMPETITION, POOR PRICING POWER
There are multiple direct competitors in every industry segment. I systematically searched the first 5 pages of Google organic and paid search results to identify the majority of Active’s competitors within each of the verticals that it serves. There are 317 direct competitors among the 28 market segments that Active defines in the “Industries We Serve” section shown on the homepage of activenetwork.com. Based on this competition, the assertion by the sell-side that the 60% penetration rate in triathlon could be replicated across any of the other verticals is wishful thinking.
The software Active offers is low complexity (i.e. web-based forms, customer relationship management, and payment processing), highly commoditized, and not defended by patents.
Active’s effort level in any one segment is much less than their competition. Active’s attempt to de-fragment the market for registrations is a dead-end due to intense competition and low network effects. For example, although Active claims to be the 5X larger than their next largest competitor, this is not at all true in the business conference space which they entered as the result of two acquisitions in 2008. Competitor Cvent was founded in 1999 and will soon employ 1,000 people committed to providing enterprise class business conference management software. In fact, business so so good that they completed the largest single private investment in the software industry this year by raising $136m to fund 200 additional positions, half of which will be dedicated to IT. They are currently hiring 35 people, 6 of which are in their sales division. The conference management segment is particularly mature as there are a large number of established competitors in this space. A 2011 survey included a review of 86 venue ticketing solutions, Active was not listed as the solution for any of these venues. Active has limited their marketing materials to the smallest segments such as museum, zoos, and other simple attractions with visitors who purchase a ticket.
Active charges comparatively higher transaction fees (6.5%) so they inherently attract small clients.
Current 8% increases in revenue per registration are not due to price increases but instead a mix shift into community and business events. All my models assume this 8% mix-shift will continue.
BEAR CASE #5 - MANAGEMENT IS AGGRESSIVE ON SEVERAL FRONTS
This negative working capital business may have had cash shortfalls in 2008-2010.
|Cash Flow Related Events|
|05/08||$5m letter of credit required by client|
|09/08||Raise $92m Series F Preferred|
|01/09||Reduce Sales & Marketing headcount by 8% due to “recession”.|
|02/09||Raise $50m Series F Preferred|
|04/09 – 12/10||Series of missed interest payments. Interest increased 12.75% plus issuance of 80k stock warrants with $0.001 exercise price|
|04/09||$4m convertible notes issued|
|05/11||All debt paid off post-IPO prior to maturity|
In the S-1, Active indicated that the 2008 recession caused a decline in the organic growth rate of the company. When comparing registration revenue from 2009 to 2010, they cite that “Overall growth was negatively impacted by the oil spill in the Gulf of Mexico in April 2010 and the continued economic downturn.” However, when asked about the impact of flooding and stormy weather in the spring of 2011 on the 2Q 2011 conference call, management said that their customers were distributed over a large enough geography that such effects were minimal and were easily compensated by other areas. Given the TAM available to the company, it is unusual that sales and marketing staff would be reduced by 8% due to a recession and/or extrinsic factors like the oil spill. At a recent ThinkEquity conference, management reiterated that it was important to to maintain their investment in sales and marketing staff in order to achieve their TAM. In my opinion, there was a substantial cash shortfall that was triggered by a slight slowdown in growth and required layoffs in 01/09. Three months later, the CEO and other existing investors made a $4m loan to the company.
While cash balances appear sufficient, their payables cycle is very demanding. Credit card sales for event registrations are collected in 2-3 days and then paid out to event organizers typically within 14 days. Essentially, all their clients have lent Active their cash (see Registration Fees Payable on balance sheet) in exchange for the right to use their software for free. Event participants pay a technology fee of $2.00-$4.00 during the checkout process which generates the “net registration revenue” which is 71% of total revenue.
From company filings:
“Registration fees payable represent the portion of the registration fees payable to event organizers, park and recreation department administrators, league administrators and other customers. Cash collected on behalf of customers is included in cash and cash equivalents prior to remitting the amounts owed to the Company’s customers.”
The company does not report gross registration sales. However, I have backed into the number by looking at changes in credit card processing fees for 2008-2009 which increased by $13.1m during this period. Online credit card merchant fees run in the range of 2.5% for an increase of $524m in gross registrations due to an increase of 45.1m registrations. This is the increase in gross registrations not actual gross registrations. The most recent 2 quarters (1H 2011) are illustrative where credit card fees increased by $3.9m or $156m increase in gross registrations over the period, or $0.87m/day. On an 14 day float, Active increased real cash on the balance sheet by $12.1m.
Stated another way, registration fees payable increase from $40.7m on Dec 31 2010 to $99.6m by Jun 30 2011 at an accumulation of $0.33m/day or an increase of $3.6m in real cash. The baseline contribution of gross registration fees to cash in unknown. However, the working capital of the business is furnished by this payable. Also, 88% of cash from operations in 1H 2011 were due to an increase in registration fees payable. Similarly, in 2009 and 2010, about 25% of fiscal yr. CFO was due to increases in registration fees payable. Nonetheless, these payables should be closely monitored in the event of a slight decline in revenue.
Pre-IPO stock valuation appears to have been manipulated to benefit management and related parties.
Incentive compensation plans include top line revenue targets which are inappropriate for a company of acquisitions.
Related party transactions with ESPN and USTA bring into question the sales strength of the organization.
The above 3 bullet points are detailed below.
1.) On August 22nd, 2008, the company repurchased $1.1m in stock and $5.3m in options from senior management and two directors at a purchase price of $14.43 per share which had an exercise price of near $2.00 per share as did the 1.3m of new options issued in 2008 with an exercise price of $1.96. Prior to date, this was the highest reported internal valuation, and coincided with the ESPN-led private placement of $80m in Series F preferred stock also priced at $14.43 on August 22nd, 2008. However, only 2 months later a much lower valuation was established. From the S-1:
“As of October 2008, our common stock valuation, utilizing the methodology as previously described, was $4.78. The valuation used a risk-adjusted discount of 19.6% and a non-marketability discount of 23.9%. This valuation reflected a 50% probability of IPO, a 25% probability of sale/merger and a 25% probability we would remain private. It was assumed that a liquidity event timing of 13 months. The valuation was conducted in preparation for the 801,306 options granted by our board of directors during this time period.”
In January of 2009, Active acquired ReserveAmerica from Elicia (who's CEO was on Active's Board at the time), for 3,461,018 shares of preferred stock valued at 14.43 per share.
It was not until April 2011, just before the IPO pricing, that I could identify a common stock fair value estimate of $13.78 which was still below the repurchase price of $14.43 made in 2008. Thus, it seems that management was willing to select a valuation appropriate to their purposes. Even in Feb 2011 with the acquisition of Fellowship Technologies for 1.125m shares or $8.9m at $7.91 per share potentially understates the acquisition cost substantially.
2) In 2009, one year after the option repurchase, Dave Alberga, CEO, lent back $1m as a convertible note paying 10%. Other prior secured loans made to the company only had an interest rate of 6.75%, although I am not familiar with more specific information regarding the loan terms.
3) The bar is set too low and there are too many ways to manipulate the executive incentive plans. The plan is based on topline revenue and adjusted EBITDA targets. Obviously, given the acquisitive nature of the firm, there is no merit is tying the bonus plan to topline revenue. Adjusted EBITDA, adds back stock compensation and excludes the accretion of redeemable convertible preferred stock. In 2010, the lowest adjusted EBITDA target was $25m and the lowest topline revenue target was $275m. Actual 2010 adjusted EBITDA was $25,120m and actual revenue was $279m. Since the bonus was narrowly achieved, I wanted to be sure no levers were pulled. The main item which stands out is that although income tax provisions were positive prior to Q4 2010, this item was reported as negative expense of $0.843m to decrease the net loss and increase adjusted EBITDA to be >$25m. Software capitalization was increasing but there did not appear to be any inconsistencies with amounts capitalized vs. those expensed. Thus, it appears that the tax provisions were manipulated to achieve the adjusted EBITDA targets. In fact, the SEC requested that the company disclose these targets in its S-1 filing and Active objected to this by responding that doing so would result in competitive harm. The SEC responded that they did not concur with that reasoning and requested again that the adjusted EBTIDA and revenue targets be disclosed at which point the $25m adjusted EBITDA and $275m revenue target was finally disclosed.
4) In 2009, the Bonus Plan objectives were not met, but 4 names officers were granted an aggregate retention bonus anyway of $770k payable the earlier of May 25, 2012 or 60 days post-IPO. Each of the officers already owned substantial amounts of vested stock options worth several million dollars post-IPO. Each had already gain liquidity from the 2008 stock and option repurchase as well. In my opinion, the use of a retention bonus here was gratuitous and not in keeping with the spirit of the Bonus Plan.
5) The 2nd largest acquisition that Active Network completed was ReserveAmerica Holdings for $47.7m in Series F preferred in Active. ReserveAmerica was owned by Elicia Acquisition Corp (already a 5% holder of Active), a subsidiary of Mindspark Interactive Network whose CEO, Jospeh Levin, is a member of the Board of Active. Levin’s previous position at IAC, parent of Mindspark, was Sr. V.P., Mergers & Acquisitions and Finance. I suspect he had acquired ReserveAmerica on behalf of IAC but could not confirm this.
6) It would appear that gaining access to USTA tennis tournament registrations had proven difficult given the traditional sales tools at their disposal. Active sealed the deal with a contract and the gift of warrants for 239,027 Active shares to the USTA in 2006. During the period 2008 to Q2 2011, the USTA provided $16m in net revenue to Active. The warrant was exercised at the IPO and exchanges for 91,148 shares or $1.4m. Good work by the USTA in selling access to its members and passing along the cost of software to tournament registrants. This arrangement potentially provides Active with tremendous pricing power as long as registrants do not mind paying the technology fees. Colorado Tennis explained a recent fee change as follows: “The registration processing fee for senior, adult and junior leagues has been increased from the USTA charge of $2 to Active Network’s charge of $3 per registration. For the tournament player, the processing fee is the greater of $3 or 7.5% plus $.50 per person per event entered”. Within this segment, I was not able to locate any material complaints regarding these fees. Being a club sport, tennis may consist of a demographic that does not mind the add-on fees. Scott Schultz, managing director for recreational tennis at the USTA, was elected to the Active Board of Directors in 2009. Although the cost of sales via warrants appears to have been a good investment, it sets a poor precedent when seeking access to the membership of similar organizations.
7) Prior to the IPO, ESPN was the largest shareholder of Active with a 21.5% stake following about $96m in investments in 2006-2008. ESPN also purchased online advertising through Active of $4.4m, $6.4m, and $6.5m during 2008, 2009, and 2010, respectively. ESPN contributed 16%, 20%, and 16% of all of Active’s marketing services revenues for the same period. Should ESPN substantially reduce their holdings in Active, it will be interesting to observe if ESPN will continue to purchase media from Active.
8) Active produced a press released for their signing up International Telecommunication Union ITU Telecom World 2011 event. The press release is dated 14 Sept 2011 and the conference is October 24-27 2011. In 2009, the conference consisted of 456 exhibitors and 20,000 visitors. Is it really possible that Active was selected as the technology provider only 5 weeks before the event? Of course not. This contract was likely signed at least 1 year ago and this announcement is pure fluff. The url for registration is as follows: https://www.ituworld2011.com/portal/startNewRegistration.ww
The “ww” suffix most certainly stands for WingateWeb, which is Active’s technology. I don’t mind fluffy press releases but the following statement is patently false.
SAN DIEGO, Sep 14, 2011 (BUSINESS WIRE) -- The Active Network, Inc. (NYSE: ACTV), the leading provider of organization-based cloud computing applications, today announced it has been appointed the official event registration and engagement technology provider for the International Telecommunication UnionITU Telecom World 2011 event, which celebrates its 40thanniversary this year.
According to RegOnline (Active’s event management software), they charge $3.95 per registrant and about 20,000 have attended in the recent past for a potential gross revenue of $0.08m. Given a forecast of $226m in total net registration revenue, winning this business hardly moves the needle and the timing of the release shows that they are more than willing to manipulate the news cycle in any way that they choose.
EV/EBITDA Multiple - 50% of EBITDA is point of sales sys. and call centers to handle 32m offline fishing/hunting licenses. This mature BPO business deserves a 7X EV/EBITDA, low end of peer multiple (see Call Center Composite). The balance of EBITDA earns a 12X multiple on my base case 3 yr EBITDA CAGR of 18.6%. The blended multiple is 9.5X vs.16X consensus.
Excluding registration fees payable from CFO, FCF likely to remain negative through 2013 based on mgmt committment to achieving about 20% of all growth via acquisition.
Base Case 4% organic growth in non-gov’t registrations is generous but will be difficult to discern based on acquisitions. My base case scenario assumes FY13 EBITDA of $41.9m and a more mature EBITDA multiple of 9.5X for price of $6.22 per share.
For all cases I assumed:
1) An 8% annual mix-shift improvement in non-gov’t revenue per transaction as they look to expand into community, government, and events business.
2) COGS% stagnates at 42.5% which is consistent with 46.1%, 44.1%, 43.4%, 43.2%, from 2008, 2009, 2010, and 1H 2011, respectively. Offline to online shift is only very slowly accruing to COGS.
3) Sales & Marketing remains at 21.1% which is high end of company-reported long term goals.
4) G&A is 13.2% (ex stock comp), which is 2% above long term goals due to organizational complexity (see new CTO pay package).
5) R&D remains at 21.9% due to organizational complexity (Bear Case #3 and #4)
Expected value is $8.77 for a 26.5% annualized IRR over 2 years. Base Case is intrinsic value of $6.22.
|ACTV Decision Tree|
|Case||Probability||Drivers||CAGR / Margin||FY13 EBITDA (m)||EBITDA CAGR||FY13 EBITDA Multiple||Comments||Price||Return||Cont.|
|Bear||20%||Gov't Rev Growth||2.0%||Precludes accretive M&A|
|Non-gov't Rev. Growth||0.0%||41.4||18.2%||7.0||$4.53||68.2%||$0.91|
|Base||60%||Gov't Rev Growth||3.0%||Best reflects current M&A trends & growth. Discounts cash balance.|
|Non-gov't Rev. Growth||4.0%||41.9||18.6%||9.5||$6.22||56.4%||$3.73|
|Consensus||15%||Gov't Rev Growth||7.0%||Currently trades at 12.5X consensus|
|Non-gov't Rev. Growth||7.0%||72.8||40.3%||17.2||$19.51||-36.9%||$2.93|
|Bull||5%||Gov't Rev Growth||10.0%||Assumes execution on growth and margins not seen yet|
|Non-gov't Rev. Growth||10.0%||77.3||42.0%||20.0||$24.16||-69.5%||$1.21|
Now to the bull cases:
BULL CASE #1 - NON-GOV’T CORE GROWTH WILL IMPROVE
My take: The sales organization is weak. In the Q2 2011 analyst call, mgmt referred to verticals such as education and golf as having strong growth but then said it wouldn’t “move the needle”. They have been heavily involved in golf for 5 years and I’m not aware of innovation in their golf product. Anecdotally, several Active account executives complain on glassdoor.com that they don’t receive training. They claim that mgmt does not do adequate market analysis. Sell side analyst suffer from the same problem. Inside sales phone numbers for Golf and Endurance go directly to voicemail during business hours, a sign of a poorly functioning sales team. Regardless, investors won’t know the effectiveness of the sales teams until end of year since organization counts are not being reported quarterly. Additionally, I question now how easily registration counts could be manipulated. For a 12 yr old company, FCF might be the best marker of progress.
BULL CASE #2 - ACTIVE WORKS IS EITHER SUCCESSFUL OR CANCELLED AND R&D EXPENSE WILL DROP
My take: The new CTO from Monster is capable and incentivized ($6.7m options to vest over 4 yrs). While the Gordian knot of code and business plans is solvable it will require continuous investment in R&D and leverage will be elusive. The constant entrance into new verticals (i.e. Churches in 2011) will continue to add complication. The CEO acknowledges large failed software projects (he referred to one as the “Death Star”) so they might quietly wave the white flag on a failed Active Works 1.0. Then, I would wonder about their long term competitiveness if they have a defunct process for innovation in view of the competitors who have laser focus on one or two verticals at the most, not the 12 that Active is trying to manage. The S-1 mentions that part of the R&D effort is actually systems maintenance so reducing headcount adds operations risk. The geographic dispersion of IT staff suggest that a reorganization will be difficult. Regardless, the slow roll-out of Active Works suggest that the project cannot be abandoned w/o exposing an important execution issue. Headcount trends in R&D could reveal the new CTO impact.
BULL CASE #3 - OFFLINE TO ONLINE TRANSITION WILL INCREASE AND IMPROVE GROSS MARGIN
My take: Of the 32m offline transactions in the outdoor segment, mgmt reports that about 4% of them are migrating online each year. The contracts already pay Active an additional $0.50 for phone transactions so the costs are likely covered or could be a profit center. Active still offers call center services in many of its verticals so the idea that they will reducing telephonic transactions in favor of Internet transactions is not clear. I asked mgmt if there was any ability to convince the state govt’s to save money and eliminate thousands of point of sale systems in favor of having just a few locations in each community (i.e. courthouses, post offices) where licenses are available. They were of the opinion that this would not happen. Mgmt commented that retailers were not enamored with training people to use their systems but they probably tolerate selling licenses as a service to gain additional customer foot traffic. With respect to their other verticals, some customers do not register for events online to avoid the technology fees. Active has traditionally positioned itself as a no-cost solution to the organization by making their fee a surcharge visible to the registrant. This is an impediment to growth. Most consumers resent this fee since they know that online transactions are more efficient and should therefore not impose an additional cost to them.
BULL CASE #4 - MIX SHIFT WILL EXCEED THE CURRENT 8% INCREASE IN NON-GOV’T REV/TRANSACTION
My take: The most leverage in the model would be a mix shift to higher dollar value transactions such as in government (ex-outdoor division), education or business conferences. Assuming a 10% y/y mix shift (i.e. 20% higher than the base case 8% increase) in non-outdoor rev/transaction, EBITDA would increase by 25%. Although Active would appear to be at scale to penetrate these markets, the Cvent discussion indicates that these higher value verticals are fully recognized by the competition given that Cvent raised more private capital this summer than did Active in its IPO.
BULL CASE #5 - FUTURE ACQUISITIONS WILL BE ACCRETIVE TUCK-INS
My take: Presumably this is what management has been doing all along and yet the model still has negative FCF. Regardless, since Active is publicly traded they do have access to certain targets who will accept their shares (e.g. Sept. 2011 ServiceU acquisition). I don’t believe that mgmt will slow down to properly assimilate these acquisitions and make them work before moving onto the next idea. The golf course customer count has only grown about 5% since those businesses were acquired in 2006. They have added 3 states in hunting/fishing license business in 3 years. If anything, Active’s acquisition’s have languished.
BULL CASE #6 - 85% OF REVENUE IS UNDER CONTRACT, GREAT LOOK-THROUGH ON REVENUE
My take: Mgmt makes a big deal of their average contract term of 3 years for their non-gov’t business. Competitors anecdotally highlight that they have no contract terms like Active and report stories of non-profits being threatened with lawsuits for violating the terms of Active’s 3 yr contract. The software that Active provides are largely self-service applications where Active has few or no upfront investments when adding a new client. The contract length originates with a sales process that offers “free” software in exchange for a multi-year contract. The downside is that event organizers are effectively committing their registrants to pay the contracted technology fee, typically 6.5% plus $1.00, every time they transact online. I think that competitors fully recognize this tactic and therefore innovate their software products to win clients from Active and provide the flexibility most smaller organizations prefer.
Status quo with 4-6 acquisitions/year (some in new industries) paid with equity.
Management restructures the entire organization, takes charges, and reboots the org. chart.
R&D headcount and costs do not fall
>80% cash from operations remain derived from deferred accounts payable.
Private markets weaken and accretive acquisitions materialize at good prices.
|Subject||RE: valuation metrics|
|Entry||12/02/2011 06:22 PM|
I agree. For 2011, 2012, and 2013, I estimate NI of $-20m, $-8m, and $-1m, respectively. The run rate on CapEx has been $30m. D&A should be about $40m-ish going forward depending on acquisition allocations. Non-cash working capital (NCWC) is difficult to use since this is negative working cap biz, and it got more negative by $30m YTD, so the NCWC is potentially transitory so I'd rather exclude it. 2011 FCFE is about $-31m due to one-time debt repayment of $27m. I forecast 2012 and 2013 FCFE at $0.9m and $6.2m, or currently trading at 127X 2013 FCFE. I would like to esimate ROIC better but there are acquisitions not yet disclosed such as this one: http://www.serviceu.com/news/active-network.
|Subject||RE: RE: valuation metrics|
|Entry||12/12/2011 07:11 PM|
Thanks for the idea and the write-up. A couple of questions:
Can you identify any more specific catalysts? Other than the lockup I didn't understand your list. Is there any point at which you think St. estimates are not achievable without more M&A?
Is there seasonality to the business that shorts need to be careful of? It looks like things are seasonally stronger in the summer quarters.
|Subject||RE: RE: RE: valuation metrics|
|Entry||12/13/2011 11:35 AM|
I believe that M&A has already been necessary in late 2011 to get to street EBITDA estimates and that pattern should continue in 2012 and beyond. The changes in operating metrics have been used very strategically by management to mask the lack of organic growth. The Q2 / Q3 seasonality should not be a concern to shorts. A secondary offering in 2012 might alert the market to the poor fundamentals but management has been clever about defining the TAM so it could be sold as "higher market opportunity" to the current base.
|Subject||Q4 2011 Author Update|
|Entry||02/27/2012 09:00 AM|
ACTV reported Q4 on 2/23/12. The original short thesis is intact.
Registration Growth Not Sustainable Ex-Acquisitions in 2012 – Registration counts increased by 2m from Q4 2011, of which 1.7m is estimated to be due to the addition of CA and OH hunting and fishing revenue – few large states left to add. Registrations were reserved as insurance anyway per the 8-K amendment filed 11/4/11 and prove management willingness to manipulate operating metrics. Thus, only 300k in registration counts (1.7% growth) could be attributed to “other” organic growth or inorganic growth. The 4 acquisitions made in 2011 at a cost $43.3m 2011 are likely more oriented to subscriptions in line with the 92.4% y/y growth in software revenue and a very unimpressive 2.4% decline in OpEx given the 22% revenue growth. I still cannot preclude the possibility of a decline in core registrations being masked by acquisitions. Regardless, registration and organization counts are too easily manipulated due to the different business models operated by ACTV and are increasingly less useful.
Gross Margin Changes Not Indicative of Cloud Service Model – It is reasonable that a cloud services businesses with any organic growth should demonstrate declining COGS%. However, ACTV does not show this pattern as COGS% increased from 45.6% in Q4 '10 to 48.2% in Q4 '11, similar to the pattern from Q3. I attribute this primarily to a lack of growth from legacy divisions and business model changes due to acquisitions in church management and ski-resort software. Given that 30% of COGS is credit card processing fees which is 100% variable cost, it appears that labor and call center costs are poorly controlled. Moreover, the offline-to-online transition for the hunting and fishing business may be decelerating as management has previously stated that 4% was migrating online each year and in the Q4 call they cited a 2% transition rate. IPO initiation reports from sell-side analysts had modeled a mean 42.2% COGS for FY 2011, but actual results were a 1.3% y/y de-leverage (42.4% to 44.7%) - a meaningful decline in gross margins. Investors should seriously consider if ACTV is a real cloud business.
Management Comp and Selling – Guided for $21m in stock comp for 2012 vs. $8m in 2011, a healthy 162% increase. The CEO has unloaded essentially all his stock - over $4.3m in first 2 months of 2012 and owns almost no shares currently. Management is not interested in improving net income which the Street had forecasted at $4-8m in 2012 but now management is guiding a loss of $35-39m for 2012, thanks to $98m in acquisitions post-IPO, which like ACTV, do not yet generate an operating profit, but do meet EBITDA and revenue targets which are used for incentive pay. Management guides negative FCF for this 13yo firm in the neighborhood of -$18m (-$37NI + $60 D&A - $41 CapEx).
Active Works Does Not Really Exist – Because R&D costs are persisting and analysts keep asking about Active Works, Alberga now explains that software development via Active Works is an “evolution” and “customers don't often even know that they have been moved to the new platform”. The ability of the engineers to centralize these different data silos into “Active Works” remains to be seen. The Alberga audio from min. 49:00-53:00 in the Q4 call does a good job of illustrating this point.
Acquisition Commentary – I anticipate that ACTV will continue to pursue any number of unprofitable SaaS firms that can be acquired at EBITDA multiples below the 33X EBITDA ($28m ex 2011 acquisitions) at which ACTV trades. Starcite, at a cost of $57m, was acquired at 9.7X 2012 EBITDA forecast. I would anticipate that execution on gross and operating margin will worsen as the overall enterprise becomes more and more bloated from acquisitions that make little sense.
|Subject||RE: RE: Q4 2011 Author Update|
|Entry||02/27/2012 02:17 PM|
I did not add back the deferred revenue to the -18m FCF estimate primarily b/c I don't understand where the deferred revenue is coming from. Management stated that Startcite will allocate rev. 50/50 between technology and registration revenue in ACTV income statement. Assuming ACTV has modeled Starcite 2012 growth in line with ACTV (ex-acquisitions; 15%ish), then Starcite was doing about $40m in total rev in 2011, with only $20m being non-registration. So, I don't see how they would carry $10-12m of that as deferred revenue (50-60%) which seems sort of high since Active historically carried about 15% for year-end 2010 ($40.6 deferred rev / $279.6 rev) and the business models are not wildly different. Given that the Starcite deal transacted Dec 30 2011, the timing is also suspicious. Had Starcite recorded that much deferred revenue very recently, then why would they sell for only 9.7X '12 EBITDA?
I'm confident that $50-100m will be spent in dubious acquisitions in 2012 to replace stagnant or declining revenue in existing divisions such that add-backs for deferred revenues will be of little consequence in measuring the ability of the firm to create any real free cash.
The 1.7m hunt/fish estimate for CA and OH is an update to the original Bear Case #2 discussion and is also supported by the following sources:
1. 11/4/11 8-K registration count restatement.
|Subject||RE: RE: RE: RE: Q4 2011 Author Update|
|Entry||03/16/2012 09:26 PM|
Thank you for your good questions. Sorry it has taken me a while to answer. I was waiting on the K, but it is not out yet.
The business model has moved away from registration and is becoming more tilted to subscriptions vs. registrations so I don't agree that the working capital generation is germane to FCF going forward, particularly for a growing firm. In fact, I think that this neg working capital will deteriorate and be a drain on cash over the next 24-36 mo. Competitors are increasingly offering clients to utilize their own merchant services (via Paypal, etc) and those competitors that are offering payment processing, are offering faster payment terms that ACTV.
The deferred revenue issue is still not clear to me.
We will see soon as to the comp plan metrics but I can only assume that EBITDA and rev will remain the targets, so I can't in good faith add back the $21m comp to get an "adjusted FCF" which is positive. Management is still running the firm for their own benefit and I can only infer that the elevated comp plan is also due to the fact that the business would have major execution issues if any of the founders/managers of the multiple acquired firms jumped ship at this stage with the ACTV network having achieved so little operating centralization.
|Subject||RE: RE: RE: Q4 2011 Author Update|
|Entry||04/19/2012 09:02 AM|
This is a well done writeup, and I generally agree with the thesis, but I'm having a hard time backing into the hunting and fishing registrations based upon the CA/OH license data. Would you mind walking through your assumptions, step-by-step? Thanks.
|Subject||RE: RE: RE: RE: Q4 2011 Author Update|
|Entry||05/14/2012 02:00 PM|
Sorry to have taken so long to reply here...
In order to determine the source of these government registration revenues I accessed fishing license purchases on a state by state basis from data provided by the American Sportfishing Association. By manually visiting all of the states' websites, I was able to identify 19 of the 23 states that used Active's software in 2009. The 4 states unaccounted for were estimated to issue an average number of fishing licenses. I was not able to locate state by state hunting license data for all states so fishing data was used as a proxy on a relative basis (i.e. Active was estimated to handle 51% of all 15m hunting licenses in 2009 to match the 51% of fishing licenses as reported by the states. Active also receives payment for transactions regarding harvest data such as reported deer harvests which I indexed to hunting permits (i.e. the more hunters, the more harvest collections). .
|Subject||RE: RE: ACTV|
|Entry||11/04/2012 10:49 PM|
Registration fees payable is likely to be a drain on cash in 2012 whereas is was a source of cash of $31.7m in 2011. The poor cost controls at all levels make it difficult for the firm to manage its cash with this declining payable. There was $239k in interest expense in Q3 which implied an average daily balance of $12.85m on the revolver, which is higher than the $0 balance reported at end of period. I'm doubtful that registration fees payable will be a source of cash in the near future with the growth trajectory laid out last week and in Q4 and it is anyone's guess as to management's near term intentions given the remix of positions.
Probably not a zero in the near term unless the SEC has an opinion on the revenue manipulations that causes their creditors to restrict the line of credit.