Show HN: Online Furniture. Why are conversions low? | Hacker News Reader

Show HN: Online Furniture. Why are conversions low?

nomiddleman.co

23 pointsnomiddlemanco46 comments

As others have said, there is a major communication hurdle you need to solve.

To pick a page at random, consider http://www.nomiddleman.co/products/executive-desk-sz1509

I have some UX questions for you:

1. What are the dimensions of the desk?

2. What does the desk look like on the other side (eg the side the user will see every day)? Does it have drawers? A keyboard trey?

3. What is the desk surface like? Will I need to use coasters for my drinks? Does it show fingerprints?

I consider those basic questions when considering a table, but from the information you provide I can't answer them. You don't even provide a brand name, so I can't search for other sites which might be able to tell me.

For comparison, check out http://www.smartfurniture.com/products/BDI-Sequel-Desk.html I've chosen them because I bought my current desk from them (though the link I've sent is a random desk to be a fair comparison). From the product description and copious photos, you can easily answer the above questions. You also get a sense for what it's like living with the desk (what it will look like in different configurations, how you'll manage cabling with it, how hard it is to clean / how to clean it, etc). Although you can't exactly buy from SmartFurnature knowing you'll love the product, you can get a pretty good idea of what it would be like to live with it. They also have a clearly outlined return policy if things go wrong.

Okay, I think that hopefully communicated the core of my feedback for you: start doing UX tests of your product pages. Ask your users "What would it be like to use this desk?" and keep iterating until you get accurate answers.

(Also, if you haven't seen it, peek.usertesting.com will give you a couple free 5 minute tests per month. The detail provided by a full step-by-step recording of a user's confusion is IME often extraordinarily more helpful than critiques like mine)

Fantastic writeup.

Thanks, will work on these items.

What's more, when I look at http://www.nomiddleman.co/products/executive-desk-sz1511 I see a brown table that can accommodate about four chairs, with sloping metal legs, for $1089. When I look at http://www.ikea.com/us/en/catalog/products/S69857764/ I see a white table that can accommodate four chairs (at a push), with sloping metal legs, for $100.

Now, I know the former product looks larger and it's brown. It may well be a nicer product. But the page doesn't give me enough information to tell how much nicer it is. For all I can tell on that page, it could be an Ikea set of legs screwed onto a wood-effect particleboard surface with a huge profit margin.

Online shopping works when people aren't afraid to buy and, when they do, don't regret it.

Amazon has always attacked this problem by trying to have the lowest prices. They were also the first major retailer to provide sophisticated recommendations, which helped people make the right decision the first time.

And, more recently, they have tried to drive the cost of shipping and returns as close to zero as possible.

All of those are good strategies, but recommendations are tough for furniture and free shipping/returns are probably impossible. If you want to succeed, you'll have to figure out your own method of eliminating fear and buyer's remorse.

If your furniture is available in show rooms, an app/website that helps people see the piece in person and then buy from you would be great, e.g. "See this piece at [insert address here] between 9am and 5pm today"

If it's not available in show rooms... I don't have any other ideas.

Edit: another strong recommendation: NO fake/CGI product photos! Those are a screaming red flag to consumers. No one in their right mind would buy a $3k couch with a photoshopped product image.

Agree with this comment.

It gets difficult when selling an emotive product. By emotive product, I mean an item at connects at an emotional level. I know that sofa is solid because it is made of solid oak, but will I be comfortable sitting on it reading my Sunday newspaper. That question is difficult to answer from an online store.

Probably worth to take a look at home24, they start to gain traction here in Europe after years of slow growth. Doing Free Shipping / Free Returns. Rumors about IPO.

"NO fake/CGI product photos! Those are a screaming red flag to consumers. No one in their right mind would buy a $3k couch with a photoshopped product image."

Tell that to IKEA, their catalog is 75% CGI.

http://time.com/money/3207951/ikea-furniture-3d-images-techn...

Firstly it's good CGI and secondly, most importantly, IKEA has the advantage of being IKEA. Don't trust the image? Go to one of their huge, trustworthy stores.

>Tell that to IKEA, their catalog is 75% CGI.

Obviously that doesn't apply to IKEA.

First, because IKEA uses so good CGI (and hybrid techniques) that their photos are life-like and totally believable.

Second, because everybody has seen and touch IKEA products, and can visit an IKEA store at any time. They're not some mystery meat web store.

Did you know IKEA's catalogue was 75% CGI before reading that article? I knew some was, but I had no idea it was that high of a percentage.

I guess this could be rephrased to "No obviously fake/CGI product photos". The CGI images on the OP's site are very clearly CGI, and the lack of quality is definitely off-putting.

That's blogspam wrapping http://www.cgsociety.org/index.php/CGSFeatures/CGSFeatureSpe... which they link at the bottom.

Thanks. The return policy is designed to get over that.

I'm not sure what to do about fake images. We struggle with that everywhere, mostly because we are dependent to some extent on our vendors for images. Obviously some don't understand why we don't want fakes.

What cities are the vendors in? Hire a photographer to take real photos of the furniture in that city, if you need to.

It'd also be helpful if you only sold products your team had personally seen, used, and approved. That could give buyers at least a little bit of confidence. It might decrease your inventory, but usually less choice + higher quality = more conversions.

I would guess that this is a tough sell because people really, really want to test furniture out for comfort before buying it. Plus, furniture is expensive, and tends to be a long-term purchase. People are gonna be reluctant to fork out a few thousand bucks unless you've got a good reputation already.

The question I had immediately was "what happens if I don't like it"? Your return policy could maybe be more prominent. I understand that it would be hard/expensive to offer free return shipping, but you could certainly make it more clear that you do accept returns.

Other than that, I dunno... maybe something that lets them upload a picture of their room and see what the furniture would look like in place?

More: maybe have some inexpensive pieces shown on the front page. I'd be a lot more likely to risk a hundred buck end table than a $2,000 sofa. The higher-end items would be a lot easier to sell after an initial purchase.

Most of your bookcases, end tables, etc. are listed as "coming soon".

Browsing the site was confusing. I'm in the market for a new Entertainment Center for my living room, and when I went to find them, there was a filter for "Entertainment Center" and "Entertainment Centers" and one for "TV Stands" down at the bottom. I could only pick one. I should be ablke to pick multiple filter options, and you should group similiar ones under a single filter.

FWIW, most sites get this wrong, and that's why thumbnails and a drop down to show all items are so important. It allows you to choose a broad filter and then visually parse the page rather than relying on crummy filters to show you what you're looking for.

thanks, I appreciate that. I'm not sure design wise that is where we want to go, but the filters clearly needs a look based on comments.

For example I want to buy a sofa. I went through the process to collect nice sofas online, but then went to the corresponding physical stores to see them in real. As it turns out, actually sitting on some or viewing close details made me dismiss every single sofa I thought were nice, for one reason or the other.

I bought a sofa model that wasn't online available, and didn't look that good from a distance, but felt truly great once you were close and sat on it.

There are things that I wouldn't shop online alone, because of examples like this. I have no problem to buy a phone or a macbook or $GADGET online, though.

This might be only me, but the images all look fake. A few images do give rich appeal, but that's because they look least fake. Also, the shots of desk with chair look out of proportion, confirming my feeling that images are fake. Looks scammy.

The commodity names really puts me off. You're targeting a luxury-ish market, and you're advertising things under commodity names.

I dont go around thinking I want to buy a "BEAUTIFUL WHITE MODERN/CONTEMPORARY FAUX LEATHER SECTIONAL SOFA"

That's what it does. Not what it is. Brand name, brand style, something catchy - whatever it is, I want something I can chunk in my head. BEAUTIFUL WHITE MODERN/CONTEMPORARY FAUX LEATHER SECTIONAL SOFA I can't chunk at all. And it doesnt sound very appealing.

In the same vein - your related products are the same type of items. You're not telling me "You can have this awesome living room if you get this sofa, with this table, and this..." Frequently bought together or something like that, ideally listed as a range or theme or something to give it a feel of cohesiveness.

I personally find the "NOMIDDLEMAN" logo hard to read because the words blend together - maybe the word "middle" should be a lighter colour?

On the "How we do it" page, there is some kind of gramatical error in this phrase: "...a product is marked up many by the time it reaches the customer..."

I'm also not convinced by the explaination. In most cases the supply chain is simply Manufacturer -> Retailer -> Customer, and I don't see how this is changed. The name makes no sense to me because nomiddleman.co is itself a middleman.

Also, this: http://www.ronstauffer.com/blog/why-not-to-use-a-co-domain/

I agree completely. The very first thing I thought when I landed on the site is, "I'll bet NOMIDDLEMAN is the middleman." There is, however, a way to do this that doesn't immediately sound contradictory. You use the marketing term "factory direct." Lots of successful businesses use this model.

One other thing that might be hurting the site is that the value proposition -- "If you can wait a little longer than normal to receive your furniture, we can save you a lot of money" -- only gets fully explained by clicking on the tiny "How We Do It" link at the bottom of the page. I would make the text "IF YOU CAN WAIT 8 WEEKS, WE WILL SAVE YOU UP TO 50%" a hyperlink to that page.

Why are conversions low? Because these look like alibaba sample pictures with the prices 5X'd. I bought one of these couches from china for about $450 through a service like yours, which is the dictionary definition of a "middleman".

Your site says: "we source directly from the factory and sell..."

That's a middleman. You're not even designing these.

You say you go to these factories and build relationships? Show me one picture of you and anyone at these factories.

It looks like you're trying to sell cheap furniture rebranded as "luxury" to idiots via a crappy ecommerce template. I'm very glad that conversions are low.

I don't know if this is affecting sales, but you've got some data entry or classification problems.

If you mouse-over where it says "LIVING ROOM", the drop-down menu that appears has something called "3 Person Sofas" in the Sofas section. Clicking that reveals two items.

If you click "LIVING ROOM" and then try to narrow it in the "PRODUCT TYPE" list on the left, there is no such thing as "3 Person Sofas".

The "2 Person Sofas" has similar shenanigans, but the opposite way. It does not show up in the drop-down, but shows up in the left sidebar as a product refinement option.

The "2 Person Sofas" category, which has 1 item, seems to have been rolled into "Loveseats" in the areas where it is not available as a separate category, but this rolling in has not been done consistently, where it is available side-by-side with "Loveseats" the behavior is confusing.

• From the drop-down, click "Loveseats". 6 items are displayed, and the product type refinement options break these down as 1 "2 Person Sofa" and 5 "Loveseats". So everything is sensible here.

• From the drop-down, click the "Sofas" heading. The product type refinements include 1 "2 Person Sofas" and 5 "Loveseats". Fine, and consistent with the above.

• Click the "LIVING ROOM" category, rather than clicking something in the drop-down. Now the refinement options list 1 "2 Person Sofas" and 6 "Loveseats".

Good pics are paramount. A cute girl on the landing page about doubles the conversion rate. So, put people in your pics. If it's a folding bed or a transformer table, make a short video of said girl doing the transformation, so that female shoppers will know it's not too hard.

Why are you putting such a focus on gender? Why does the assistant need to be a "cute girl"? Do only "female shoppers" care about the difficulty of using furniture? Your advice comes off as pretty sexist to me.

>Your advice comes off as pretty sexist to me.

Sort of. That's why it works.

In this case it just means using "sex" to sell something, the oldest trick in the advertising book.

If you want to sell something using the possitive images of sex appeal etc it's not necessarily a bad thing (I mean any more bad than advertising with any other means of persuassion, personally I find advertising bad in itself).

Protestants (including atheists brought up in protestant society) have this weirdo idea that all things should be de-sexualized, else it's "sexist".

Nope, it can just be sexism: invoking desire and recognizing that sex is natural and something people enjoy.

(Wheareas "sexist" is implying a sex is inferior and should be serving the other, etc).

> Why are you putting such a focus on gender? Why does the assistant need to be a "cute girl"?

Because this is how it works. You're welcome to try your luck with male teens or whomever else you deem less sexist.

> Your advice comes off as pretty sexist to me.

My advice is based on A/B testing of user visits to the online furniture retailer I used to work for (a fairly large one, ~$10M/mo in sales). This is as close to objective data as it gets. If you perceive it as non-PC, so be it, it does not change the underlying reality.

After a 2 minutes check, the deal breaker for me on this website is that there is NO WAY I will spend several thousands of dollars with so little information on a specific product (lack of pictures for example - I would need a lot more visuals to appreciate the quality of the models proposed).

Plus, no customer evaluation (this is the first thing I look at right after pictures: what people are saying once they bought it).

Look at what Amazon, Rakuten does. Their websites may be crap visually, but they have a bunch of useful information making your purchase "safer" in the first place.

Furniture on Rakuten (in Japan at least) has litterally pages of pictures / dimensions, different setups to help the purchaser imagine how it actually fits in their environment.

I can see a whole business built around 'show HN' where you crowdsource input on what you should do to improve a customers website. Just create a novelty account, post a show HN, implement the top voted comments and profit.

Of course HN's advice is about 10% good, and the up-votes may not be accurate in coaxing out the valid signal. Lot's of people will have opinions, but often it's not really coming from a source of experience. Even on this, most of the comments are about the website itself, which probably could use some work.. but only 2 i think i've seen mention what the customer might be thinking.

I think if you really want to know why your customers aren't buying, you're better off asking your customers.

> I think if you really want to know why your customers aren't buying, you're better off asking your customers.

That's more effective for sure.

I was part of a renewable energy community for a while and we found out that one member had built a whole business around picking the brains of the community members to re-sell the solutions as his own for a pretty penny.

When we finally found out about this we pulled a pretty elaborate prank on him causing him to never trust us for advice again ;)

What does low conversions even mean? Are you benchmarking your conversion rate against another business that does something similar to you?

To one business, a 3% conversion rate may mean impending death. To another, it may mean millions in profits.

I agree with the other points in here, the website looks cobbled together and untrustworthy. A similar idea to this is made.com - great design, actual images and has experienced great success.

That is where we are trying to get to. Its all about incremental improvement.

To build a successful site like this, you need to either be the kind of person it's serving, or talk to that kind of person in depth: Someone who buys $1,000+ desks.

Maybe it's the wrong area to attack in that market , because people really like to test couches. Maybe the key it either find a good way to test couches virtually(how ??) ,or some social model that lets people test couches at homes of other people(maybe something like airbnb for showrooms), or create a technology or business model to help showrooms somehow, etc.

It could be just me, but the furniture in all pictures very uncomfortable. I can't imagine sitting down and being comfortable on the couches in the very first picture (top of the home page, under "Live your dream" heading).

Thanks everyone! I really appreciate the feedback. I'll hammer away at finding and fixing the grammatical errors and classification. I truly appreciate the help and feedback.

People do test-driven furnishing and need to see that it works for them before buying. A photo is a bad test of this.

For those claiming the problem is what's being sold, made.com seem to be doing well...

Also, I killed the clickable link. For reference, how does this item look to you?

http://www.nomiddleman.co/products/doney-3-piece-white-sofa-...

Right, I'm going to give my somewhat brutal answer then show my journey.

> Why are conversions low?

Because the site is utterly broken for finding products, almost every page has some error on, and the products themselves appear to be extremely expensive while not presenting me with enough information to see why I should buy them.

Have you actually used your site? Honestly, are you loading your own site and trying to find things?

The more I look around, the more problems I find. Your site is broken on a technical level but even if it wasn't there's nowhere near enough information for me to make a decision to buy. Loads of things don't even say how big they are!

---

Here's me using the site, this is exactly what I find, not leaving out any working pages:

---

I opened the page, and there's a huge clickable image in the middle ("live your dreams at an affordable cost").

I clicked it.

> PRODUCTS > > SORRY, WE COULD NOT FIND ANY RESULTS. TRY ANOTHER SEARCH.

You have no products?

---

Kitchen and bath (should be bathroom) has these options:

> POT RACKS BAKERS RACKS KITCHEN ISLANDS

None are for the bathroom. All take me to a "coming soon"

---

The next five things I choose also have no products.

I have been on 10 pages on your site and have seen no products actually for sale.

---

http://www.nomiddleman.co/products/ameyali-makeup-table

Looks like a computer render, and is a nearly $600 piece of MDF.

---

Searches full of "COMING SOON!" I do not care about products I can't buy.

---

http://www.nomiddleman.co/products/fabric-ottoman-10025-xh-1...

> Dimensions W X D X H:

> 31.5in x 15.6in

Encoding errors too

> ContemporaryåÊottoman

---

http://www.nomiddleman.co/collections/office-officedesks-exe...

No desks under executive desks, yet there's one in the trending?

---

http://www.nomiddleman.co/products/executive-desk-sz1509

You want to take over a thousand dollars from me but can't be bothered to write more than

> Executive Desk Aluminum Alloy Replicated Wood Brown

Ironically, http://www.nomiddleman.co/products/ameyali-makeup-table, was not a computer rendered item.

You want to take over a thousand dollars from me but can't be bothered to write more than > Executive Desk Aluminum Alloy Replicated Wood Brown

I'll get to longer descriptions here quickly.

Ah, yes, my machine generated error... I am eternally chasing that gremlin it seems.

Reply on news.ycombinator.com