Answer by Matthew Carroll:
Completely Reworked Answer that was published by the Business of Fashion:
I ran a three part Forbes series: The Rise of:
[Part 3] Gilt's Strategy to Combat Full Frontal Assault by Competitors – http://onforb.es/xq0aD0
I have worked with Gilt several times as a vendor. Here is their business model:
They have excellent buyers – they find the best brands with great product. Once you have established a relationship with their buyer they purchase samples. The brand sends the ATS (Available to Sell Report: the available inventory that is currently not allocated to orders) to the buyer and then Gilt requires the vendor to block out the inventory for the sale
Gilt promotes the sale & implicitly the brand. After the Sale ends, Gilt places their firm PO against the orders that they have received payment for and based on the available inventory that the brand allocated or "blocked out". Credit terms are normally Net 10 or Net 15.
Gilt Groupe Metrics & Statistics:
25 – 34: 30%
35 – 44: 22%
45 – 54: 17%
Brands can expect to take a 55% to 60% discount on Wholesale (can be more depending on the deal you craft). Brands should always try to craft a deal that prices the deal at Landed & Stored Cost (i.e. FOB + Duty + Shipping + Landing + Allocated Storage Costs).
[Author's Note: I guess I didn’t realize how piss poor of a job I did defining the Margin Analysis in this answer, so in answer to the 15 messages in my Quora Inbox – here is a deeper dive into the context of Margin.]
Background Definitions to better understand:
Since there are not many people familiar with how the #s work for Retail, let’s breakdown some terms to serve as the foundation for further analysis.
FOB: Most footwear and apparel firms purchase FOB [Free On Board], commonly referred to also as First Cost, which is the price paid to the factory for the fully assembled, boxed, and cartonized shoe at the Port of Hong Kong.
Landing Costs and Tax & Duty: FOB is the cost to get it to Port of Hong Kong, which you hoepfully strategically planned to hit on a Thursday so you can catch the 11-day Maersk Fast Boats to Port of LA. Landing Costs include expenses like freight insurance, bonds, transport from the port to your warehouse, the warehouse inbound fee (eg the fee to unload the container and put your inventory on the shelves of the warehouse).
Landed Cost: This is FOB + Shipping + Landing Costs + Tax & Duty. This represents the baseline figure that you use to set your pricing as a brand.
Wholesale Price: This is the price that the brand (Me) charges to retailers like Urban Outfitters, Finish Line, Journeys, or Amazon. Generally speaking there is a golden rule called "keystone" where fashion companies establish their pricing by multiplying everything times 2. Here is an example:
Landed Cost: $25.00
Wholesale Price: $50.00
[Note: Landed Cost can get FAR more complicated when we start discussing issues like IRS 263a for capitalization of expenses. However, for purposes of this example we are going to assume that ALL Warehousing, Fulfillment (pick the order and ship it to the retailers, and Inventory Costs will be accounted for in the Operating Expenses (more specifically, Selling Expenses – commonly referred to as SG&A).]
Easy Peasy! Now that we have the price that the traditional retailers pay, we need to define the customer’s price that you see at the stores, more commonly referred to as Retail Price.
Retail Price: This is the price that the ultimate consumer of the product pays when you purchase a product from Urban Outfitters, Bloomingdales, Finish Line, etc. This price is also generally “keystone” or Wholesale Price times 2. For Example:
Landed Cost: $25.00
Wholesale Price: $50.00 (Landed Cost X 2)
Retail Price: $100.00 (Wholesale Price X 2)
[NOTE: Keystone is the general rule, but certainly Wholesale Price can be 2.1x Landed or Retail could be 2.2x Wholesale. This all depends on the retailer's pricing power, brand positioning, and market position to be able to make these tweaks.]
For purposes of moving forward, I am going to assume that you understand what Gross Margin is (Gross Profit / Revenue) or the % of each $ of Revenue that drops down to fund OpEx. Let’s run through a quick example, from the Retailer’s Perspective:
Retail Price: $100.00
Gross Margin: 50%
[Remember: This is from the Retailer’s Perspective, so the Wholesale Price is a COST to the Retailer as it represents the cost of the retailer to get the goods to sell]
Now that we have a baseline defined, let’s begin to apply it to Gilt Groupe. Gilt is a Private Sales site that sells items for a huge discount. Gilt Groupe is what we call a "cut-out shop" meaning that they look to purchase unsold inventory from brands at a discount. The logic is that production is based on the stupidly difficult art of inventory forecasting of multiple sizes per style. Therefore at the end of the season, you don’t sell all of your inventory – you may have sold all of your size 10 shoes in color x but have 100 pairs of 10.5s remaining in inventory. This is what makes footwear so hard: every color needs to be forecast to a 10-variable equation – if you screw up the sizing on 2 or 3 sizes, you can kill the profitability for the entire style.
Since I have 100 pairs of the 10.5s and numerous other sizes still in inventory, Gilt comes to the brand and says “hey, we like your stuff. We are going to host a sale.” Because they agree to retail inventory that is slow moving or in odd sizes (eg you only have 8.5s, 9.5s, 10.5s, 11.5s remaining because you screwed up the half size variable), Gilt looks for a 50% discount off of Wholesale. For example:
Wholesale Price: $50.00
Gilt-Offer Price: $25.00 (50% off Wholesale)
If Gilt acquires the product for 50% less they can mark it up 2x and sell it for the original Wholesale Price. For Example:
Gilt-Customer-Price: $50.00 (2x Markup yielding a 50% margin to Gilt).
Also Note: If the Original Retail (from above was $100) and Gilt’s Customer Price is $50.00 then the customer who purchases the product on Gilt received a net savings of 50%.
Now, you maybe saying “Matt, Gilt says on their site that this is 60% off retail.” The answer is that we “adjust” the price comps. If retail is $100.00 (a keystone markup), I might tell Gilt that retail is $125. Therefore, the discount that the customer sees is 60% ( use % change ((end-beginning)/beginning)-1).
There is a caveat to the Gilt offer is that you have to use their logistics company to send the product to them (i.e., ship inventory to Gilt, who then fulfills it to the customer). This is a standard plug of 5%-off. You can fight with Gilt, but it’s a pretty safe figure at 5%.
Therefore, when you add this layer of complexity:
Wholesale Price: $50.00
Gilt-Offer-Price: $22.50 (55%-off Wholesale – 50% for inventory, 5% for logistics)
Gilt Customer Price: $50.00 (2x Gilt Wholesale Price)
Gross Margin: 55%
It's a brilliant model, they run the sale and collect cash immediately on purchase. Based on Sales, they purchase only what was sold, nearly eliminating their inventory risk. Additionally because of their credit terms, they are holding cash at the market rate of equity for the 15-20 days (assume 5 days for shipping) – a killer benefit for them.
[Note: If you are building a model on this don't forget that they are taking an average of 1.75% of Gross Sales in credit card fees]
One interesting aspect of Gilt is that they pay via WIRE – probably one of the most interesting aspects of the cash cycle. Let's take the Groupon Model where the company mails the vendor a check. In addition, Groupon could implement an accounting control whereby ALL checks are issued on a Thursday. This means that Groupon effectively has somewhere between 3-10 days of cash that it's holding at the market rate of equity [think about how big of a number that is when you are talking about the discount rate applied to a $1.35B valuation on $400m in Rev – It's a ridiculous model].
This little policy of paying vendors via Wire is incredibly interesting because it is something that would go unnoticed by 99% of people. Per my detailed explanation below of the cyclical nature of fashion, this is critical money and having a vendor tell a brand that the check is in the mail is almost as good a saying "we haven't sent it." Although the wire reduces the return on holding cash, it is implicitly investing in your product by maintaining killer relationships with the coolest brands. We all bounce around from company to company, so keeping these relationships is critical.
Wholesale Price: $100.00
Gilt.com Discount Price: $45.00 (55% discount on $100.00)
Total Units: 1,000 pieces
Total Revenue for Gilt.com: $100,000.00
Total Cost of Goods: $45,000.00 (Vendor (me!) Revenue)
Credit Card Fee: 2.00%
Fulfillment: 7.5% from Vendor (me) to Gilt & back out to Customer (Gilt's Customer)
Remaining Inventory & Exchanges:
As I said, they only purchase what they have already sold. However, there is risk for chargebacks, fraud (can be a surprisingly high number), and product exchanges/returns. I am not sure of their return policy and cannot speak authoritatively on how returns and exchanges are managed on their end or affect the profitability of the sale.
[UPDATE:– Explains this directly.]
Additionally, there is a lot of interest in Drop Shipping. I give a FULL breakdown on the pros & cons of drop shipping:
Benefits to Brands:
For the past 2 years, it's been nasty for brands & cutout shops. Literally, we were at the will of these folks who were in many cases being completely unreasonable (sometimes asking for 90% discounts). Gilt is a great company that is honest, efficient, and nice to work with – they don't get greedy trying to shake you for an extra couple of points.
In addition, brands get rid of inventory, boost cash, and basically sell the inventory for about their cost. This is killer because by the time we are getting rid of the cutout for the season, we have a huge cash requirement for the next season's production. Although these sales hurt overall profitability, it's a trade off based on the upcoming cash requirements and holding costs associated with the inventory.
Furthermore, Gilt generally becomes a large source of inbound traffic to the brand's site (forecast a 5x increase in traffic during the sale with a decreasing 25% daily following the sale) with conversion rates of 1-1.75% vs industry average of .50% conversion rates). Meaning that brands can offset some the Gilt losses with the 75% margins that we make on direct sales for products not in the sale or items that were sold out. Although the volume will be decidedly smaller than the Gilt PO at least their are ancillary benefits associated with working with Gilt. Additionally, there is a level of marketing to exposing the brand to the historically high-end member of the Gilt community. We have also seen that customers who really love the brand will actually purchase the goods directly from the brand to support us – proactively paying more because they missed the sale or just want to support.
Alternative Business Models
The business models for other discounters like Karmaloop's PLNDR or Revolve Clothings Reverse/Rewind (whatever they are calling it now) are different. PLNDR purchases the inventory then holds the stuff (higher risk, but they want a larger discount – 60-70% off wholesale.) Revolve Clothing's Reverse actually takes delivery of the blocked out PO and then sends back to the brand what doesn't sell, so they can fulfill quicker. The interesting part of the Revolve Clothing model is that they pay 50% of the received PO PRIOR to the sale. They put their own $$ on the line and trust the brand to issue them a check if the brand doesn't perform well during the sale. In a strange twist of fate, you have a vendor who now possesses the credit risk from a brand, versus the other way round [credit is still effectively dead for fashion, so we all became credit analysts]. Different business models that have their benefits & risks.
Similar Business Model & Comp:
Considering the nature of your question, I am assuming that another example would be helpful. Another company that was very similar to Gilt was JackThreads.com.
They were acquired by Thrillist in May for $7m.
Q4 ‘07: 15,000 (http://on.wsj.com/umlnIe)
Q1 ’08: 75,000 (http://onforb.es/v5ykz3)
Q4 ’08: 500,000 (http://cnnmon.ie/vXJ9Af)
Q3 ‘09: 1,500,000 (http://onforb.es/spCXUW)
Q2 ‘10: 2,500,000 (http://on.wsj.com/vnhQAc )
Q2 ‘11: 3,500,000 (http://read.bi/vAKOXw)
Why does this model work?
I have kept the above fairly objective and a matter of fact/experience. However, there has been this constant nagging about addressing the factors that led to Gilt's success in the market, and I figured that I provided enough unique context above to dive into this here.
They were actually exclusive:
65% less than 35 years old
67% greater than $150k in Annual Income
87% bachelor degrees
In 2007/2008, you really had to be someone to get an invite. It took me having to step up and run a pretty damn cool footwear brand before I got my login. It was not like the bandwagon camps that say "members only" and there is an automatic approval system. Gilt had a highly sophisticated (aesthetically and culturally) audience that had the ability, the means, and the inclination to purchase your product at full retail (more often than not, Gilt purchases are from people purchasing additional items for brands that they had paid full price for at the beginning of the season and were really stoked on the purchase – the opportunity and the exclusivity compelling them to purchase again).
This is a business that has grown predominantly through word-of-mouth marketing. Seventy-five percent of our membership has come from the suggestion of a friend, using our onsite 'Invite Friends' feature [in exchange for a $25 credit.] That's how we launched. We sent invites to a list of about 15,000 people—friends, former colleagues and classmates, dating back to grade school! I'm surprised you didn't get one.
Susan Lyne, Former CEO Gilt Groupe,
http://on.wsj.com/r5DQiN – Wall Street Journal Interview on Oct, 29th 2010
A Sense of Urgency:
Not really in the sense of time of the sale. But in the sense that the customer FEELS like s/he will never be able to find this brand producing this model ever again at this price. You HAVE to get it before everyone else snakes it.
The most exciting challenge we've faced is how 50% of all revenue from Gilt sales come in less than an hour [after the sales start online]
Alexis Maybank, Co-Founder Gilt Groupe, http://tech.fortune.cnn.com/2010/06/25/gilt-groupe-brings-you-the-online-sample-sale/
In addition, these are incredible deals that simply are not available during an end of the season "half-yearly" sale. Due to the inventory risk of carrying product, retailers could not take in the inventory, ESPECIALLY in the last two years. CIT died so boutiques didn't have credit (READ "limited size & product variations available as lost sales were better than sitting on product in their eyes") and the transportation costs of moving these units between stores would KILL margins.
Gilt enables their target demographic to have the ability to access the aggregated excess inventory from across the country in one central location with guaranteed availability. This is a value proposition that represents a critical risk to their members as the chance of finding it (the featured product) at a major's half-yearly sale poses significant transportation (they are a lot of bloody people, it takes a lot of time) and high opportunity costs (there is no guarantee that this style in this color in this size is at the accessible location to, its a lot of bloody effort to get there, and suffer through the lines without finite expectations of tangible results). These push factors instill a sense of urgency for executing the transaction with a "screw it, I'll get them both" mindset.
Optimized Product & Pricing:
Gilt maintains a tight locus of control over their product selection with highly informed statistics about their target audience. Humans are inherently very poor when it comes to wide product selection as the more options there are the more that they experience significant cognitive dissonance after the execution of their purchase. When you evaluate the product selection, notice that each sale/opportunity segments their target demographic into purchase categories that increase vertical associations (i.e. shirts & pants) and reduces horizontal choice (i.e. do I want the blue shirt or the green shirt). These segments focus on the guy who would rock a grey-plaid shirt, is a blue t-shirt guy, and these pants go with both. I have an insane amount of respect for their analytics/data mining team, but also for their buyers who manifest these data points into purchasing decisions that they are held accountable for.
“If we have 50 sales a day, we can't fit all 50 in one email,” she explains. “With SAS (http://www.sas.com/technologies/analytics/), we can figure out the top sales to put in front of you based on your preferences.”
In addition to cross-shopping, the company witnessed a 10-20% lift for customers browsing in new merchandise categories who had not purchased in those categories (percentages vary depending on the category). It also saw a 20% increase in new member conversion rates.
Tamara Gruzbarg, senior director of customer analytics and research at private sale site (http://www.dmnews.com/gilt-groupe-turns-to-sas-to-enhance-customer-analytics/article/203136/)
Since a large portion of you folks are web app folks, think of this as moving from a 5-step sign-up process to a 2-step (I log in, the product is arranged so my "style", draws me to grab x, y, and z and bam I am checking out – not stuck sorting between color A or color B. The potential to miss my size being available is too great (i.e. opportunity cost) – I have to just do it.
A Symbiotic Ecosystem Compelling Vendors
Gilt has some of the best buyers in the industry who are bloody cool. Not 'uber-trendy' or 'holier-than-thou fashionista' cool, but down to earth people who fundamentally understand, are passionate about what they do, and have a DAMN good eye. More importantly, selling your excess inventory to Gilt was like receiving a nod of approval (not so much now as they are facing insane competition and expanded their reach). Generally speaking, a brand studies very closely who & where they close their inventory out to in a manner that optimizes cash flow and ensures that the target audience DOESN'T see it. For fashion (emphasis on small, higher-fashion brands – the Theory's/Bloomies/Saks staple brands of the world do not necessary apply), once your customer knows you are at Nordstrom Rack/Barney's CoOp, it means you are mass market and your brand is trending down. As I alluded to above, Gilt could actually carry as much, and in some cases even more, weight than retailing at the best boutiques in the country (i.e. Fred Segal, Atrium, American Rag, CNCPTS, or Bodega).
I am referencing the above to illustrate the critical link in the symbiotic ecosystem required for a model like this to function effectively. The confluence of about 30 different factors all boil down to a systemic value proposition for ALL parties (Gilt, its members, and the brand). This is the clear differentiator as the core nature of Gilt compels great brands to work with them and provide them with the best styles, in a manner that serves to benefit Gilt & its members.
As a brand, you could sell your excess inventory to your target demographic through a venue that actually serves to augment your core intangible asset (i.e. the brand). OH YA… Keep in mind that these sales occur at a CRITICAL juncture in your seasonal cash flow cycle as the brand has plowed every dollar back/drawn down on ALL of their revolving credit facilities to finance next season's production hitting their ex-factory dates. International distributors have just paid and this partnership with Gilt bridges your cash flow through landing the goods and collecting on your CODs and net 30 term accounts – well, at least until you can coerce your Factor into accepting the accounts that achieve your minimum receivables factor rate.
[Definition: Minimum Receivables Factor Rate = the % of your credit accounts that a company NEEDS to sell to be able to meet obligations for the next 30-45 days or uphold the required cash reserves of your financing covenants.]
A Unique Approach to Pricing Power:
This update is in response to an answer that I found toOne of the answers there stated that the special-sauce was their pricing power that gives them some sort of advantage. If we think about pricing power in the classical sense of the word, we think of the Walmart model. The more powerful you are the more you can drive your cost down, a concept called "Economies of Scale" & "Opportunity Cost."
After digging into this deeper, Gilt in fact does leverage its brand and pricing ability in a form of pricing power, however in a very different manner than would be classically defined. The Gilt model is NOT, as Tom Friedman would say, in a race to the bottom. A lower price for Gilt does not create long term value for their business. As you are drive down prices, your brands/vendors get pissed off that we are getting screwed, and we stop selling to you. If they are trying to force down prices for discounts in the near term they are destroying their long term shareholder wealth by alienating the vendors that enable them to have a product-focused value proposition.
In addition, considering the level of the SES (Socioeconomic Status -http://en.wikipedia.org/wiki/Socioeconomic_status) of the members of Gilt, the product will not sell any better at 65% versus 55% off wholesale. As I explained above in the Urgency section, these are emotional purchases with very limited quantity available. Meaning that if there is a 10% difference in discount, the product's scarcity supports the higher price. Furthermore, since Gilt maintains no inventory risk of slow moving product there is no incentive for them to drive down the prices to the vendors.
In stark contrast to the classical sense of Pricing Power, Gilt actually employs its brand position to secure a fair market rate for cut-out-inventory. They have effectively established a market-rate for cut-out at 55% of wholesale, for an industry that was in a bloody free-fall.
Gilt's Business Units:
Gilt Competitor Analysis: