This weekend Miranda and I set out to buy a new mattress.  Our current bed is old and giving us both back pain, so we decided that it was time to get a new one.  My family warned us that buying a mattress is one of the worst possible consumer experiences, and with that in mind, we planned ahead to make sure we got a good deal.

Our plan was simple.  We were going to go to three different mattress stores, lie on some beds, and make notes and take pictures of the mattresses we liked.  After collecting notes and pictures from all three stores, we planned to go home, look at all of our data, compare prices, and find the best bed for the best price.  Setting out this weekend I thought our plan was pretty solid and I was feeling pretty confident.

After several hours of visiting stores and lying on beds, we realized that our plan was not going to work at all.

All of the mattress stores we visited carried the same brands of mattresses – actually – as far as I could tell, there are only really 3 brands and they all start with an “S”: Sealy, Serta and Simmons.  Each brand has different lines of mattresses (e.g. Simmons Beautyrest World Class, or the Sealy Sterns and Foster Estate) – but inside those lines, each store has a slightly unique variant with different features and a different silly name.

For example Macy’s carried a mattress called the Simmons Beautyrest World Class Hollingsworth Luxury Firm Pillowtop, while Sleepy’s carried a mattress called Simmons Beautyrest World Class Shakespeare Luxury Firm Super Pillowtop.  As far as I could tell these were practically the same bed – but Macy’s version touted “exclusive titanium support system” while the Sleepy’s version touted “1000 Coils.”

As far as the price, goes – it was all over the map.  Mattresses that I thought were essentially identical were priced up to 2x higher at one store vs. another.  Also, every single bed was on some sort of sale and the sale price was anywhere from 20% to 80% discounted vs. the normal price.  One store even had a permanent “sale” sign stitched into the pillows on every mattress in the store.  Note: When I saw this I was stopped cold in my tracks – the irony of permanently installed “sale” signs was almost too much for me to handle.  Also, on top of the sale price, some stores were even willing to negotiate to further reduce the price (although other stores got offended when I hinted at discussing a “special” deal).

Frankly, I found the whole experience to be maddening.

But in all of this, after studying the mattress industry for 2+ days, I learned an important lesson: at the heart of every sale – especially a business to consumer sale – there is an information imbalance between the buyer and the seller.

How much did the store pay for the merchandise?  What is the profit margin?  Is the current sale a good sale or is there an even better sale coming next week?  How low will the store negotiate to sell the item?  Will the store negotiate at all?  Is the merchandise actually good quality, or does it just appear to be good quality?  Are the product attributes listed on the fancy displays at all indicative of quality or relevant?

When buying a car, a well-informed buyer can actually enter a price negotiation with almost as much information as the seller.  Cars can be bought through several channels and across several dealerships and the consistency of car make/models gives consumers a good basis for price comparison.  For example, you could price out a Honda Civic from multiple sales channels and have a pretty good idea of what price you should pay.  On top of that, Consumer Reports and other research sources will tell you, on average, what the seller paid to acquire the car – which gives you a pretty good idea of how low the seller will negotiate.

This is not the case in the mattress industry.

Some very smart (or sleezy, depending on how you look at it) people have rigged the mattress industry to give the seller every possible information advantage over the buyer.  Since you can’t compare mattresses between stores, and there isn’t even a consistent set of evaluation criteria to consider between mattresses – buyers are pretty much at the mercy of the seller.

Well, at the end of the weekend, after agonizing for several days, we ended up buying a mattress.  The price we ended up paying for our mattress was close to 50% less than the “normal price” listed – but I honestly can’t tell you if we got a good deal or not.  I guess we’ll know in a few years when we’re either happy or waking up with sore backs.

Sales Lessons: Buying a Mattress
Tagged on:         
  • The joys of information asymmetry!

    Part of the problem, of course, is that people want to see mattresses and try them out before buying them. So you need to defray very high costs of retail space + setting up the beds to look nice (those are big and require e.g. a bed frame, other equipment). If people can identify the mattress and purchase elsewhere, you’d get extreme “showrooming.”

    Since people don’t purchase mattresses very often (average lifetime: over 10 years) familiarity doesn’t exist – that is, people don’t develop a knowledge of what they want in a mattress and can buy on criteria. They need to try it out.

    Plus, there are significant preference differences which are non-generalizable across people. So a more expensive construction may not feel better, i.e. cost of manufacturer does not correlate to value provided or willingness to pay. That makes pricing very difficult.

    So: is there a *better* way for sellers to sell, given these constraints?

  • I don’t think there is a better way for sellers – the current system gives sellers all of the information advantages. There is one big flaw though – a mattress showroom can only be so big. Of all the showrooms we went to, the average only had about 20-30 mattresses (and out of those, only about 4-5 fit our specific criteria of a firm non-pillowtop, interspring mattress). So although we went to three stores, we really only got to evaluate 12-15 different mattresses – a pretty small number compared to the number of total mattresses that could have fit our criteria. I think there is an opportunity for a seller to enter the market, provide quasi-scientific method for matching people to beds and then provide a global index of all of the beds people are likely to enjoy (decoding some of the intentionally confusing names across different stores).

  • Perhaps; to me, that sounds fiendishly difficult with a high margin of error. Consider the difficulty Netflix has had with their prediction system, and they have *far* more data available.

    The problem being that if you get it wrong your return rate will be through the roof, which will cost a fortune and very quickly make you unprofitable.

    I also don’t think there’s much incentive to reduce confusion on naming schemes – for a seller. Extra sales is dicey reasoning (or: you are very unlikely to get any) and you sacrifice big in margin. If your competitor sells a $1,200 mattress which costs them $500, and you sell the same mattress – disambiguated – for $850 – you need to outsell your competitor 2:1 to match their margin. And this their advertising dollars, their customer service, the quality of their infrastructure/site/etc. It’s a very difficult thing to do, unless selling mattresses is a side-show (e.g. Wal-Mart) and they don’t devote the same kind of floorspace other companies do (e.g. Sleepy’s) which hinders customer evaluation (and willingness to pay).

    That said, research does exist. Consider, which is perhaps my favorite sleep research site:

    Including, say, a compare tool: