Description

INVESTMENT THESIS      

Carparts.com (“PRTS” or the “Company”) is a subscale online seller of generic aftermarket auto parts that has been in a perpetual state of turnaround for the last several years. Revenue growth had been averaging -13% over the prior 12 quarters and the Company had never found a way to reach profitability. Then the pandemic arrived and provided a number of large one-time boosts which increased sales 61% in Q2. This perennial sub-$2 penny stock ran up over 700% as the market is extrapolating one quarter of peak revenue. The business is operating at full capacity and the temporary benefits and favorable competitive dynamics that drove this performance are now abating.

Further, there are a number of underappreciated structural flaws with the business model itself. The most significant of which is the fact that PRTS derives 70% of revenue from collision related parts, products which are inherently only purchased every few years and thus require constant reinvestment to acquire customers. DIY collision repair is also niche business with a limited customer base. The second major structural issue is the fact that the Company generates 35% of its revenue from selling its generic products on the the Amazon and eBay marketplaces, a channel with very low entry barriers and negligible margins. We believe this short presents a +50% return potential as the Company’s revenue growth decelerates and proves unable to scale over the coming quarters. PRTS’s market cap now stands at $745M, trades at over 57x EV/ EBITDA and is presently GC. The recently expanded float following an equity offering and insider sales provide technical support. 

BACKGROUND & SETUP

There is not much to this business. PRTS generates approximately 60% of its revenue through direct sales on its website, 35% through marketplace listings and the balance via dropship. Its products are almost exclusively sold to retail DIY customers. The Company previously sold parts through a handful of domains and then earlier this year consolidated all its online business into a single site, carparts.com. It subsequently changed its corporate name to reflect this transition (fka US Auto Parts Network Inc). This step enabled PRTS to channel the marketing spend within its direct segment to a single entity. Starting in Q1, the Company began increasing its online ad spend. It then redoubled its AdWords buying in May and June. SimilarWeb and Alexa data indicates nearly all the increase in PRTS site traffic during the May-June period came from paid search with organic results showing only a modest increase. As we will discuss, this provided a transitory boost. While no doubt this spend was well timed and consolidating its sites was a logical step, as you drill down into the business further, several major structural issues and non-recurring benefits become apparent. 

Low customer value: Collision parts account for 70% of sales for PRTS. Those that perform their own collision repair represent a small subset of the market. These are almost exclusively lower income individuals with older cars with high insure insurance deductibles. Car owners obviously only need collision parts when they have been in an accident or damaged their car. The average American gets in a car accident every 18 years so this is clearly a sporadic and non-recurring need. With excess free time at home during the lockdown period of Covid, many people took up DIY projects they had deferred or previously ignored (ie. no excuse for your spouse when asked to fix the brake light that had been out for months). This period was a massive boon and likely will continue to be so for many consumer products. Collision auto parts, however, had a one-time benefit with limited tail. Those that used a portion of their stimulus checks to fix their car will not need to do so for years (and it does not appear will be getting another stimulus check for the foreseeable future). Meanwhile in August miles driven is trending down 17% per OPIS with credit card data showing collisions down close to 20% due to less commuting and limited vacation travel. The customers that discovered carparts.com through a Google search in the spring will have no need to return in the near-term. When they do next seek auto parts online again, they will likely have forgotten about the site and just go with whichever option is the top paid search result, has the lowest advertised price or simply default to Amazon. The poor customer satisfaction and site reviews would further suggest less likelihood of customer returning. The biggest complaint is slow delivery which brings us to Amazon who provides one-day shipping on almost all the auto parts it sells. 

Amazon: During the onset of the pandemic through July, Amazon prioritized its fulfillment of essential items and de-emphasized the shipping and selection of items like its auto parts. All FBA marketplace vendors were impacted as well. This temporarily reduced online competition for PRTS but has now normalized. Over the last two years Amazon has built an extensive catalog of easily searchable car parts available to retail consumers with one-day shipping. Amazon will almost certainly regain the market share it conceded during that four-month stretch now that it is fully back online. PRTS simply cannot match Amazon’s turnaround time and pricing.           

Marketplaces: eBay and Amazon account for 35% of PRTS sales. These are both obviously highly competitive and challenging venues to do business with nearly zero barriers to entry. This is made particularly tough for PRTS as it sells un-branded, generic parts so can only compete on price in a venue where price is easily compared across vendors. Most items almost certainly sell at zero or negative margins after fulfillment costs. The Company is selling one-off SKUs with a single item per shipment so this reality will persist regardless of volume level.

Big Four: O’Reilly (ORLY), Advanced Auto Parts (AAP), Genuine Parts (GPC) and AutoZone (AZO), all experienced major boosts in DIY sales in Q2 as well. These four omni-channel competitors dedicated resources and efforts toward keeping their brick-and-mortar locations safely open and pulled back on digital marketing during the spring into the summer. With the larger peers out of the market, CPMs were likely temporarily lower in May and June. However, in response to the environment, each of these four have since redoubled marketing efforts, particular online and BOPIS. So PRTS not only benefitted from less competition but lower marketing costs in May and June, cost that have now normalized. No surprise, the Company’s Google search ranking has dropped off as well. Below is a breakdown of the top 10 auto parts websites ranked by monthly views (data for Amazon’s auto parts sub-site is not available) showing the surge in site traffic for PRTS and relative pullback for the larger peers in Q2.

FINANCIAL PROFILE & VALUATION 

In order to optically inflate gross margins and obscure its lack of scalability, the Company excludes distribution expenses from COGS. This is a practice we have seen with other companies, including our previous write up, the RealReal. Like REAL, the economics of PRTS are not likely profitable at the unit level and thus the busines will struggle to ever profitably scale. Starting in Q1’20, the new management team stopped disclosing operating expense details and any KPIs (previously provided number of orders, average order value and conversion metrics). This curiously coincided increased AdWords spend to pump sales. The below lays out the prior three years and the actual margin profile. Bottomline: This is a business that is not profitable on a variable cost basis so inherently cannot scale. It operates at a large competitive disadvantage with very low margins (even before fixed costs) and will perpetually lose money.

The weakness of the business model can further be evidenced by the fact that it just experienced the equivalent to its World Series and Super Bowl in Q2 yet only generated pennies of profit. That Q2 revenue level will almost certainly prove to be the absolute peak. Even if demand persists, the Company will struggle to grow volume off these levels given existing capacity constraints (management has noted the business is presently at full capacity and are experiencing high of out-of-stock rates). The marginal profits PRTS did record in Q2 are going to be eroded in the near-term by additional fixed costs it is assuming with the opening of a new DC and an ERP system rollout. This and the normalization of digital marketing rates will expose the model’s lack of scalability.

This is a classic example of the market capitalizing one good quarter of financial performance PRTS is currently trading at over 60x LTM EV/ EBITDA. Sell-side and paid research promotion, Russell 2000 inclusion in June and Robinhood popularity contributed to the stock’s run up and irrational valuation. On a sell-side hosted investor call earlier this year management commented, “anyone with $10M could do this.” An exaggeration but not far from reality. We would estimate its replacement value at $80-90M ($70M in inventory plus a domain and $5M of additional working capital) versus its current enterprise value of $742M. Regardless, the stock should not bear greater than a 20x EV/ EBITDA multiple. At that level, PRTS would trade at $5 per share, 60-70% below its current price. Note this would still be twice the stock price it had previously ever traded pre-Covid. Activist, Kanen Wealth Management, stepped down from the board in June and sold all its stock and a co-founder just sold 2M shares last month. Inevitable earnings misses will spur a reappraisal of the stock. This will likely be compounded by the less sophisticated, weaker hands that have chased momentum.  

OTHER RED FLAGS & CONSIDERATIONS   

• None of the top management team members have experience in the auto space. Current CEO was the former marketing director at a small photography equipment retailer. 

Catalyst

- Revenue deceleration/ reversion 

- Further margin deterioration

- Appreciation for that fact that a business that is not profitable on a variable cost basis cannot scale