UK Sports Warehouse, based in Oxfordshire, England, has been selling sporting goods though online marketplaces for fifteen years. They specialise in clearance – products which other retailers have been unable to sell through their normal channels.
Clearance (or liquidation) is a normal part of the retail business. Shelf space in stores and warehouses is a valuable resource, so items that don’t sell need to make way for those that do. And that often means selling them off in bulk for less than the cost price. Sometimes much less.
So clearance can be a great source of profitable stock for marketplace sellers, but there are downsides. One problem is that the supply changes every day – you can’t simply reorder bestselling products. Other businesses can sell the same line successfully for months or years, but clearance sellers need to constantly refresh all their SKUs.
UK Sports Warehouse (UKSW) saw the risk in only selling clearance products and decided to diversify, by adding a number of current product lines to their portfolio. But there was a problem. A “cheap and cheerful” image can work for a company selling clearance gear, but people who want the latest equipment expect to buy from a company that is serious about sports. That’s a very different brand.
So I caught up with Elizabeth Hitchins, an experienced ecommerce consultant who has been working with UKSW for a number of years. Elizabeth had the job of building a whole new sports brand for UKSW. This is the story of how she created SportsBubble, and the ups and downs of launching it as a new business on multiple online marketplaces.
Andy: How long have UK Sports Warehouse been selling on eBay?
Elizabeth: Going on for fifteen years now. They’ve been around a long time, but it’s always been clearance. With the recession there is less clearance out there, because retailers are being more sensible and there’s not a lot of overstocks. At the moment it’s still pretty good, but it isn’t as prolific as it was when I first started working with UK Sports Warehouse about seven years ago.
Back then they were getting clearance thrown at them every which way but loose. But now you have to go and hunt it out, and you have to really develop close relationships. I know that some sellers haven’t been as lucky and can’t hold onto those relationships, so they just can’t get the stock at a good enough price. And the time might come when big brands like Adidas decide to handle their own clearance, and run it through eBay themselves.
So it might just dry up. Or they might just get smarter and smarter and smarter at predicting what will sell and what won’t sell. Either way there’ll be less clearance for people to scrabble over. There is going to be a time when clearance will not be viable in the sports sector, so you need a plan B.
So when did UK Sports Warehouse start working on their plan B?
They saw the need to diversify about three years ago, but you have to wait until the conditions are right and the funding is there. This kind of project takes more resource than the normal running day-to-day. It needed to be set up and launched, which is where I came in, then the rest of the team could take it over and run it alongside the other business. It was really good to see it supplement and help the other side of their business.
It kicked off in earnest at the beginning of last year. So January 2014 we started all the preparations, made sure that everyone was singing from the same hymn sheet. We had to look at the pricing strategy, which marketplaces we were going to sell on, which SKUs were going to be first, which ranges we were going to do first. Sports is very seasonal, so we had to match the store launch to the up-and-coming seasons.
Initially the project was six months, and after that time they would have a certain number of listings and it would be handed over. Within that project, I was also training extra resource on inventory management. It’s not sensible to pay consultant fees to create items, but within the catalogue there were about 4,000 items up for grabs and the data was sketchy. It all had to be Amazon-ified, eBay-ified, and optimised. That’s quite a long job and you can’t rush it. So part of the project was to train external resource abroad to create that data quality, so they could be used in the business as an extra pair of inventory-creating hands.
What kind of products were they? Were they already selling them through another channel?
Some of them were the latest version of products they’ve had on clearance before, but they’ve never sold the regular lines. They were going from just having clearance to regular retail. We knew that there was already competition on eBay and Amazon for those products, so decided to compete on better data and better quality of service, not price. On the SportsBubble eBay shop, there’s some really cracking images of products, for example. They were taken professionally, to help enable the selling process.
It’s not really about undercutting everybody else that’s selling at retail, because it’s just a race to the bottom then. If they wanted to do that then they would just keep to clearance, because there’s more margin there.
But it’s a challenge not being the cheapest. On some items we could be, but on others there just wasn’t the margin. UKSW was buying from the wholesaler as a retailer, and getting the same kind of prices as everybody else. That was one of the reasons they stayed away from it for so long. While clearance is good, why try to compete?
So how did you get the project started?
The first thing we did was plan out exactly what we were going to do. We started with eBay because that was going to be easiest, and we knew we could have a second eBay account. Also if we started with eBay, all the data could be done for Amazon using ChannelAdvisor.
We had a SKU and a title, attributes like colour and size, then a link to a catalogue image and pricing. That’s the only data we had. The titles weren’t optimised, so it would just say something like, “rugby short maroon”. The data was very minimal. If you think about a wholesale catalogue, they don’t need anything more than that because you’re supposed to know what you’re buying.
So we had to expand the titles, we had to expand the descriptions, to appeal to real consumers. Then we had to create the parent and child relationships for all the variations where needed. For eBay we had to go through and add the item specifics data. In the sports categories every category has a different set of data, so we had to categorise things up and make sure we had the standard set for that.
We had to expand the titles, expand the descriptions, to appeal to real consumers
Then we went onto Amazon, and had to make sure we had all the Amazon specifics as well as things like colour mapping, department, and the nice new images of course. For eBay we had a logo on the images, but for Amazon you can’t have that. Also the browse nodes and keywords for Amazon are very essential.
So we had to build up all the data and format it correctly for both marketplaces. But that was just the eBay and Amazon project. In the background, the company was also setting up on Tesco, which has a different set of data again. So we were working with the Amazon format of data, the eBay format of data and now a completely different set of data needed for Tesco.
As you can probably imagine, the data requirements got really big. The grunt work would include things like, “Okay. For eBay we use the colour red, for Amazon we want to call the colour rose, but the colour map is red. For Tesco we need to use pink.” So it was three lots of data and they all had to be really good quality. Also Tesco was a very new marketplace, and there was a new best practice, a new shipping set up etc. I was setting it up and also training two UKSW teams, one in Oxfordshire and one in the Far East, on those data requirements so they could carry on with the work.
Which country are the external team in, and what were they like to work with?
They’re based in Bangladesh, and they’re a very good team. But when English is not your first language, it can be very hard. I think a lot of people fail when they use outsourced teams because they don’t train them properly. We actually trained them these guys into the sports industry, so they can do optimised titles and descriptions in English. They’re doing really well now, but the first file we got from them was disastrous.
It was like, “Oh God, we’re never going to get anybody to do it.” But then we realised, actually they’re not rubbish. We need to train them better. Part of my task was making them into really good listers, so they could generate inventory data the way the client wanted it, as well as for the three marketplaces.
How did you go about training them?
When we got their first file back I went through it to see where they had fallen down, and what they didn’t understand. They’d taken descriptions from a Polish website and put it in an English listing, and mixed some German words in there too. They couldn’t recognise the English and the German and the Czech and Polish as different. They were looking at items on somebody else’s website to see the description but weren’t delivering it properly.
You have to have patience, because you are dealing with people who aren’t experts
So I went through and dissected the file in detail, and told them how it should have been. They corrected the file, and when we got it back again it was a lot better. We kept doing that process, so it’s just a little bit better each time. You have to have patience, because you are dealing with people who aren’t experts in sports either. How do they know the difference between a hand ball and a football and a match ball and a training ball? You can’t assume they will pick it up just like that.
So there’s been a lot of training, mentoring, monitoring. And now the team is turning around very good inventory data. It’s a very, very cost effective way to list. But because they have gone through a development process, they’re being paid more for this than they get for their normal work. They’re happy to do it because they’re getting double what they normally charge, but that is so they keep the quality high.
How did you give your feedback to them?
It was all written. I would do a feedback report and say, “This is what you’re doing wrong here. This is what it needs to be.” It started as a couple of pages, then it started to go down and down and down. It’s a constant QC [quality control]. When files come through now, the team at UK Sports Warehouse scan over all the items, then spot check a few in detail.
Things fall down for a lot of companies that use resource abroad, because they don’t take the time to train
You can’t say, “Well, we’re just going to leave them to it, we’ve done enough training now.” It’s an ongoing process, because you are dealing with people whose first language is not English. If they’re working on a different sport, or a different set of items, the mistakes will be there. But we’re finding that they are minimal now. Normally the in-house team just corrects them, uploads to ChannelAdvisor and goes live.
We’re finding that they don’t have to put in as much resource, but it took a couple of months to train the team. It’s like anybody you take on within a business, you can’t assume that they’re going to know everything, you have to train. That’s where things fall down for a lot of companies that use resource abroad, they don’t take the time to train, especially if what they get back on their first pass is rubbish.
Were there any other problems or surprises with the data?
It was very hard to predict how long the inventory was going to take and the cost behind that. We also had the cost for the photography. And I don’t think we realised how much checking we would have to do of the data. We knew we had a catalogue and a certain amount of data from the supplier, but we didn’t know the accuracy of that data and how much we’d have to change. We thought, “We should be able to get 100 items a day done, that should be fine.” But actually we had to physically cross-reference the products lot, and that meant we were only able to do 50 items a day.
We would see a photo of a red product, and say, “Okay, these goggles are red”, but when we ordered a box there were two red, five pink and so on – it was an assorted box. That wasn’t mentioned anywhere. So then we needed a photo of each colour. We found we couldn’t rely on their stock number, because they don’t split it out in colours. So we had to make sure for those items that we only sold them if we had the stock on site. We had to recreate SKUs and split up the colours.
Then the supplier would say, “This is the new version”, but we only had a photo and details for the old version. That slowed us down a lot, because when we received the item it would turn out they hadn’t got any more of the pink ones we’d been selling, only black ones. We were working on the inventory data as we were ordering it in, but would be surprised when it arrived. But it was a good learning curve on the difference between how a wholesaler uses data, and how they structure their inventory, as opposed to a retailer.
What else did you need to do to get it live on eBay?
We started by getting the inventory data ready for about 100 items in February 2014. We had to do the professional images and inventory data, and I needed to set up the channel of either account to work with this project because it needed to be set up a little differently to a clearance account.
So we opened a new posting account and controlled things a little better than before. It was quite a different set up to the main UKSW account because the main account had grown organically over the years. This account was in hindsight, so we could set up a perfect account with all those lessons learned.
We didn’t start out with a professional design template on the first batch of listings. I just did a quick responsive eBay design template with a logo and a little bit of information to wrap around the data to start. We just wanted to go live with those first 100 items and start to work out the kinks.
We went live around March/April and that was like our testing period. Off the back of that we started the design project, with Widshop and createyourtemplate, so we could have a professional design. We were getting a brand design together as well.
How did the SportsBubble name come about?
We decided on the name pretty early on. We’re basically putting a ring around what we sell: sports equipment for men and women and kids. So cricket helmets, training cones, training gear for teams. Not tops to wear out shopping, tops for training – performance wear. There are a few fun things and outdoor games, but it’s all kept within our bubble. We’re not going to step outside and say, “We do sports fashion now.”
We sell some really good value brands, and also higher end brands like Li-Ning which is a Chinese celebrity brand for badminton and other sports. The quality of the photos is very high so you can really see the detail and quality of the item [here’s an example].
And how did the rest of the design come about?
I did the first version of the logo, and then we got a professional branding company to come in.
They created a logo and a style guide, with the colours and fonts to use – the look and feel we were going for. We weren’t sure whether to go with bubbles or not, because that might be too kitsch or obvious. But in the end we thought, what the heck. We kind of made it subtle.
Then the eBay template was designed around that, and went live around August 2014. It was all going on while we were building up the inventory data.
What kind of sales volume were you doing?
We started to really ramp things up in the summer. We were doing about 15 orders a day, and then it suddenly jumped to 30/40 orders a day and just kept rising. It was like someone switched on the tap in July/August.
Then it was like, “We’ve got 100 orders a day, oh, that was quick.” We’re stable at around 200 orders a day now. On eBay your account is limited for a certain amount of time, but it suddenly takes off when the limits get lifted.
We found that it had really good traction over Christmas, and we sold a lot of the fitness items. One of our biggest sellers was exercise bikes, especially at the tail end of Christmas and in January. Training balls were very popular as well. It pointed out a few items that were going to be better sellers, so we could then do the data a lot better.
Is that just on eBay? When did you start selling on Amazon?
We sold only on eBay until around September, because we were trying to open up a second Amazon account. We had a problem though, because normally you can’t have a second Amazon account if you’re selling the same product category. If we were going to sell beauty items, that would have been fine, but not sports and sports.
So we applied for a second account, and Amazon gave us the run-around, as they do. In the end they said no. We played with a few ideas on how to get a separate Amazon account, but decided in the end to run the new lines through the existing UK Sports Warehouse account. The account was in very good standing so we decided to run with it. That’s when the orders really started to rise.
So with the branding finished and sales coming through, was it just a case of increasing the number of products?
Yeah. We decided to just feed the machine, ramping up the products, so by the end of the year there were about 1,300 lines. Those were 1,300 optimised SKUs that had gone onto the sportsbubble eBay account and also onto Amazon through the UK Sports Warehouse account.
On top of that we added another 2,000 quick-and-dirty SKUs onto Amazon. We were in the process of taking the catalogue data for nearly 4,000 items and making it a lot better, creating our own listings for eBay and Amazon, but that takes a lot of time. So we decided to list all the other items in the catalogue via their EAN numbers.
We were competing with ourselves, but the UKSW account was a top-rated seller with all the bells and whistles
So the account ended up having half of its products with really good optimised data and half using pre-existing catalogue data. It wasn’t long before Christmas and we didn’t want to miss out, even if the data wasn’t great. We needed to do a certain amount of volume to make it all worth doing, and it gave us a boost.
Because it was just before Christmas, and we were still building up a new eBay account, we decided to run all the items through the UK Sports Warehouse eBay account as well. We were kind of competing with ourselves, but the UKSW account was a top-rated seller with all the bells and whistles that provides.
In an ideal world all the data would have been perfect, and everything run just through the SportsBubble account. But it was taking a lot of time to train the team up, and we wanted a quicker return to help fund the project. It really did pay off.
How did you keep inventory levels accurate across the different channels? How did you handle reordering for so many product lines?
The wholesaler actually has a live stock feed. We needed to run that through ChannelAdvisor so we could update the stock, and reconcile inventory on site with what had already been ordered, and what was available for us to buy. We kept a low level of stock on site so we could do one-day despatch, and then kept backfilling it. The live feed kept us out of the water.
Ordering wasn’t automated through ChannelAdvisor, but we did an order every day for the stuff we were running out of. We were dealing with a large range of items, up to 4,000, and we kept about 3,000 items in stock. But we couldn’t have the depth of inventory because that’s a massive investment.
Over Christmas the velocity would be a lot quicker than we could order it in. It was kind of a balancing act.
We would have one to three of each line in stock, but for lines that were getting really popular we’d buy in all the stock available, to make sure we had all of them. We also needed to be able to split out regular lines from clearance items, even though we were running both through the UK Sports Warehouse accounts.
We needed to know our SKU velocity, and make sure we always had three in stock. The warehouse manager had to do a lot of work maintaining that, and we had to make sure ChannelAdvisor was giving us accurate data. Fortunately we could get stock within a day or two, but sometimes things would be missed off orders, and especially over Christmas the velocity would be a lot quicker than we could order it in. It was kind of a balancing act.
As the business matures you’re able to order in more and more. And you can order less frequently because you can predict a little better: what are the popular items? Which are going to sit there? Those are the kind of lessons you learn in your first year.
I noticed the SportsBubble account on eBay was registered in 2005. So is that an existing one that was renamed?
It was, yes. It was a secondary account for a while, and also used as a personal account. It was sensible to repurpose it because it already had seller feedback. We could use variations from the off, because normally you have to have (I think) about 15 feedback before you can do that. And you also have a higher limit than if we had to start a completely fresh account. So we could put 100 items up early on, rather than having to wait.
It’s pretty rare to be a third-party seller on Tesco Direct. All the sellers there fit on one page!
UK Sports Warehouse were already selling there, but we decided that regular lines would be much better for Tesco. There’s a lot of data input that needs to be done, and you wouldn’t do all of that work for clearance items you’ve only got one or two of.
If there’s some great clearance stock with very long depth, like 3,000 of them, they work great with Tesco. But it’s about weighing up the time you need to get them listed and perfect for Tesco, versus how quickly they’re going to sell out on other marketplaces. We had it where, especially during Christmas, we’d prepare all of the items for Tesco but within a couple of days they’d be wiped off the map through Amazon and eBay. So we decided to focus on regular lines.
UK Sports Warehouse only went live on Tesco right before Christmas. It’s a longer process for integrating with the Tesco marketplace, and Tesco themselves had some delays. We couldn’t do click-and-collect, which is run by Yodel, until after Christmas because Yodel couldn’t fulfil any more orders. We didn’t feel fully live until January 2015.
There were also a lot of validation steps with Tesco. It’s roughly a seven-month implementation period from start to finish, depending on the amount of SKUs and the quality of the data. We would be going through the data and then Tesco would be testing it, but there were delays. We would be thinking, “We’ve done those items but they’re definitely going to sell out before we go live.” We decided that the regular lines were more sensible because they’d still be there in January.
Do the marketplaces vary a lot in terms of order volume, and which products sell best?
We found in our early days that anything under £10 sold on eBay, and everything over £10 didn’t. We still hadn’t proven our worth I suppose. We weren’t a top ranking seller, so only small items were selling. So we changed the strategy to listing everything that was cheap, because we knew the smaller items were selling.
Before that we were going to list all the inventory one brand at a time, but we found that even if we were the cheapest and had the best listings the more expensive items weren’t selling. People were happy to spend up to £20 with a new seller, but no more than that. Only around October time did we start seeing people putting orders in for £50, £80, and we thought, “Ah, now we’ve gained peoples’ trust.”
Anything under £10 sold on eBay, and everything over £10 didn’t. We weren’t a top ranking seller.
Fitness and training equipment did really well, things like training cones. I think they’re quite hard to get hold of. So people were buying a lot of the training cones and other training items early on. We were also finding that the house brands were doing better than the big brands. I think that’s do with the account perception too. Accessories like strings for tennis rackets, training balls, were also going quicker to start with.
A little later on in the year we started to sell hockey sticks, cricket bats, baseball bats. Baseball items did really well early on, I think because there wasn’t that much competition. It was interesting working with a really wide range of sports and wide range of items. We wanted to sell footwear and larger equipment as well, because there is more of a profit margin in that, and that came with trust. When we pushed the items through UK Sports Warehouse, the big items sold because the trust was already there.
On Amazon the higher priced stuff was selling from the start. I don’t think Amazon outperformed eBay consistently on those items, but it did match the sales in the end, especially over the Christmas period.
Tesco had a strange beginning because we had around 700 lines, but were doing all the sales on about three lines. We were selling 40 to 50 a day of one line, and it was great because on eBay and Amazon we sold three to four of that item a day. It’s like one product gets all the love, and the other items are not shifting. So it was a very different experience selling on Tesco. I can only assume that it’s to do with the algorithm, or maybe shoppers are very seasonal on Tesco or go for very specific things. But in other sectors it might be completely different.
And was it profitable? You said that the profit margins were better for clearance lines.
We structured the pricing so there was a clear profit margin within every sale. I don’t expect it will have made a profit in the first year overall, because of the additional set up costs. But within each price-build there was a clear 23% profit included. That’s after all the direct costs.
With indirect costs like the staff and warehouse, which were pre-existing, 3% or more is probably going to be lost. If we ended up with 15% profit at the end I think that would be quite good for regular lines. In clearance the profit margins are much higher, but the stock is less regular.
Do you get more returns on certain marketplaces?
Not really, no. It’s all kind of equal. The return rate and demand is more category and product-based rather than marketplace-based. You find that within sports there’s not that many returns especially if you’re selling sports equipment. You don’t have sizing issues. People pretty much know what they want. Of course when it comes to clothing and shoes, you get more returns because garments and sizing will always be an issue.
Has your personal involvement ended now? How did the handover work?
My involvement ended around January 2015, and I finished handing it over to the in-house team. They’ve been running it exclusively since then.
The handover is a gradual process. The team already run a large Amazon and eBay business so you’re not going to teach an old dog any new tricks about the marketplaces. But it’s just about making sure everybody knows the slight differences in the process, making sure the stock is checked and the quantities are managed, and handling the different order format. Because it was a gradual process, it was quite an easy handover. If a client didn’t know eBay and Amazon, that would be different.
If I tried to hand it over to a large brand who didn’t already have a team working with an eBay and Amazon business, it would be difficult to bring that team up to speed. But in this case I’d been working on site and they’d all been there while I was setting it up. It would transfer almost by osmosis because they’d been through all the trials and tribulations with me.
So what’s next for SportsBubble?
The big job is finishing all the product data, getting all the SKUs listed with good quality descriptions and photographs. Then maybe new marketplaces, a web store, international expansion. But they need to see how things pan out. SportsBubble was a very new project, a kind of experiment. It worked in the first year, and basically needs to survive its second year. The business is growing and it’s down to them to take it further.
Thank you for sharing so much from your experiences with UKSW and SportsBubble.