James Young was speaking at Dot York, these are my notes from his talk.
We set up our own business because we were getting fed up with the same requests - requests we didn't necessarily agree with. Even though we were the professionals the clients were overruling us, often with stupid ideas. Took time out to set up a business, we already knew how to set up a business and how to build an online store (design, build, copy). Spending real money put genuine pressure on us, we had to make a commitment and get some stock we had to sell.
2-3 week sprint to get the bulk of the site built. Plan was to sell from day 1. Content first to shape quickly. Whiteboard, paper and experience fleshed it out quickly. Not using waterfall model. Not having back and forth with crazy clients saved a lot of time. Took a few weeks for people to take us seriously and sales to pick up.
A lot of things we assumed would just work and bring orders. Helped us to understand the motivations behind some of the "crazy" client requests. Adverts were high cost and low reward, Adwords was just burning through money. Facebook ads were trying to get us to bid against ourselves at one point! Giving away stuff sounds like a good idea but people are just there for the competition, they're not actually engaging to buy. Got zero return from print advert in a magazine (and they didn't even send a comp copy of the magazine!)
Lower budget things have worked. Writing our own original content sounds obvious but the competition were just copy-pasting the manufacturer's description, writing our own blurb made us stand out. Sent the product to the editor of the magazine that ran our ad worked really well, she wrote an editorial which drove a lot of sales. Newsletters and social engagement needs to find the right channel, Facebook and twitter doesn't seem to be working but Pinterest and Instagram do. Good design and optimisation helps people to buy things easily .
Selling and optimisation
Our resources are limited so we focus our time on things that will be useful to us. Reporting allows us to fine tune the customer experience - by increasing the basket time out to over a year our sales increased. Just do free postage - account for it however you can but it saves so much hassle for customers. Offer a useful money back guarantee, we've only had a few people return things but we back our product - if you don't like it then we're not going to quibble. Find out why people don't commit to buying - provide clear information as to why people don't need to be concerned. Things like free postage, guarantee, price match, say who the carrier is (reassure them it's not Yodel). Adding this information increased our conversion rate.
Telling the real world
We don't have the traffic levels or time to do lots of A/B testing. Stick to good design principles, can be hard to do when you are under pressure to sell things (i.e. upselling). Ask customers "did we nearly lose your business?", find out what the friction points are in your process - about 70% of customers fill in a short survey on the order confirmation page. Gives you an insight into the dropped sales that you never see
It's not just design and code. We learned how to deal with companies that aren't a web agency - companies that aren't just building websites, they have very different problems to an agency.
This is still a side project but we're trying to grow it with more capital investment. We'd like to try this again with a different product.
Automate where possible, when possible - let computers do the things computers can do. Look for opportunities that are more than just colour tweaks and button typography, there are much bigger wins out there. We know how to run a business that isn't a web agency so that makes it easier to interact with business that aren't web agencies (who, after all, are web agencies entire customer base).
It's been fun! We're doing things that wouldn't have been possible if we were just a web agency. And it makes us a decent amount of money, allows us to have "20% time" and helps smooth over the agency cash flow cycle.