Transaction Friction

I’m currently in the midst of improving the usability of rentoid.com. We are quite certain that it will improve things dramatically from where we are today. easy to follow steps and visuals to first time users.

But in truth it really won’t be enough to be game winning. After this website improvement, we’ll need to continuously iterate what we have.  And if I am to tell you all a secret here on startup blog it’s this… Some the new changes will need to be replaced in a couple of months, and so on in perpetuity. The reason is that I get phone calls all too often which sound like this:

Customer: ‘Hi I’m ring up about renting that double bed you have listed on rentoid.com’

Me: ‘Sure, are you new to rentoid? If so let me explain very quickly how it works…..

(I say we don’t own items, rather have a website that people use to rent items to and from each other. I tell her how to join / transact)

Customer: So it’s some random person?

Me: Usually things like beds are rental business – it will be identified as such saying’rentoid business partner’

Customer: Ah, no sorry… It sounds just too hard sorry…. thanks.’ Hangs up phone.

* If the first contact was with the owner, things may have been very different.

There’s too much friction. As a little reminder this is the definition of friction as it pertains to physics:

A force that resists the relative motion or tendency to such motion of two bodies or substances in contact.

Friction

The bodies in this case are the customers and the website. And transaction friction occurs whenever these bodies interact.

Currently, there are too many interactions before a transaction can happen. The reason the system has been designed this way has been to ‘protect revenue’, as all my readers will know I don’t believe that “Free” is a business model. But the question I am seriously asking myself is ‘How much revenue are we losing because of friction?’

It is free to join and list on rentoid.com and we take 5% revenue of each rental. Which means both the renter and the rentee need to do quite a few clicks before the transaction is complete. Both parties have to deal with significant fraction in order to transact with each other.

The owner has a lot of friction listing the item

The person renting also has a bit of friction when paying the 5 % deposit on rentoid to get the owners details.

This process protects our revenue, but slows things down and is painful.

There is no doubt it that turns many people away – how many we just can’t know. Don’t get me wrong, I’m not about to make it ‘free’ – but maybe it’s time to make people pay to list. Which we have avoided to reduce barriers to entry for listers, as listings (having the most items for rent in one place) is game winning. That said, listing already has a lot of friction in any case…. it takes time & effort, maybe asking for a few dollars up front isn’t really a problem? Maybe people wont mind if it creates more rentals for them because their phone number is on display for the renter to call them immediately and get renting? There is only 1 way to find out the answer to this question and that is to implement it. But friction has to be balanced. We need the minimal amount of friction to to move things forward, but enough friction to make so money as well.

The other benefit of changing the system is that it will bring the money forwards and reduce complexity of the site mechanics. But it does open to other competitive risks like screen scrapping, and listings being stolen.

I’ll let you know which way we go…. In the mean time think about this in relation to your business or startup:

“How can I reduce transaction friction in my business?’

twitter-follow-me

6 thoughts on “Transaction Friction

  1. Great insight.

    I’ve just finished reading Founders Stories by Jessica Livingston (highly recommend). It’s a series of interviews with the founders of different tech companies from Steve Wozniak on Apple to Paul Graham (her husband) of YCombinator. In it Craig Newmark from Craigslist talks about how they get 5 billion impressions a month and the reasons that most of their listing are free. They make money off of charging certain types of customers: they charge for real estate listings, some personals ads and a few other categories. Maybe there is a model like that there?

    Some AB testing on this stuff would be awesome. You’d be able to quickly work out the difference in success rates and take some of the guess work out.

    Big ups

    Ned

  2. Yeh, it’s a really big insight. I actually got the idea of Fiona from http://www.artshub.com.au who said that when they made a shift like that and everything was smoother. They just charged an annual subscription fee and people didn’t mind because the thing was easier to use. I’ll definitely be making changes in the next few weeks and find out. But there’s no doubt the major area for improvement in rentoid is reducing friction.

    Steve.

  3. Nice post Steve. I’ve been reading your blog for a while, thanks for sharing your insights.

    It’s an interesting decision you’ve got ahead of you regarding charging for listings. I take it you’re using a good website analytics tool with goals and funnels set up to track how users are converting on your site?

    When you’re A/B testing different features on your site what do you use? Google Website Optimizer (www.google.com/websiteoptimizer) is awesome if you haven’t tried it.

  4. Hi Steve, I’m glad you’re looking at ways of reducing the friction in your model. I’m just wondering whether you need to have the rentees phone number available or not – maybe the the potential renter should register interest in renting the item and the listee calls them to discuss? Thereby protecting your listing from having any usefulness in a scrape. Worth talking about more. Protect your ip, but keep the steps simple. It’s a fine dividing line isn’t it?

  5. Steve

    I once worked on an interesting market research project that defined the barriers to purchase as a series of concentric circles. At the centre was the actual product (in that case a theatre experience). Then there was a sequence of decreasing diameter circles, each representing a barrier the purchaser had to cross to access the product. In the case of theatre we found that it was often the most mundane factors that impeded purchasers’ progress through the circles to the centre – stuff like parking, transport, dress code, peer pressure, price, venue quality and so on. We found a couple of dozen issues I recall.

    It’s no different trying to push people down a website purchase funnel (and ditto re the previous commenters about Google Analytics and Optimizer, cool gadgets to help with this).

    Sometimes it’s the little things, for example, I once read research that said that green ‘Purchase Now’ buttons outperform red ones by a large margin.

    Sometimes it’s larger more obvious things, I remember at Arts Hub we improved signups markedly simply by combining two registration pages together, so instead of Step 1, 2, 3 we just had 1 and 2. I think it’s now all on one page for the same reason. Logically you’d think people would finish a signup if they’ve made the decision to pay and started filling out details. But the act of having to click from Step 1 to Step 2 resulted in people dropping off.

    Cheers…David

  6. Awesome insight David, thanks for sharing with us. I am absolutely convinced the price isn’t the major sticking point in the transactions which fall through on rentoid.com but more the ‘concentric circle’ barriers… and very soon I’ll be doing some live testing to find out.

    Cheers,
    Steve.

Leave a Reply