I think you have to love learning in this game. When the parameters change as quickly and as frequently as they do, you have to want to keep ahead of things.
I don't know whether it's cause or effect that the people you meet in SEO particularly are inquisitive and curious and typically have a lot to teach you.
At Distilled, we are surrounded by great, smart clients and we get to hang out regularly with constantly innovating competitors. I thought I would share some of the (non-SEO-specific) things I've learned recently:
Lessons from clients
Sometimes it's something from their core business and sometimes it's that they just go ahead and do something that works. I've got an example of each:
How to manage at high-speed
The first client I want to introduce you too is Reevoo. They provide review functionality to major retail sites and aggregate reviews on their own site. They recently launched a great website called Just Buy This One that reduces your technology choices down to a single gadget in each category / price-point (even to the point of having the "best toaster".
It's all based off their reviews and they are so confident that you will love the one you buy that they offer an independent satisfaction guarantee where they will take it off your hands if you end up not liking it.
Everyone I have told about JBTO has loved it and I even used it myself to buy their TV recommendation. As promised, I love it.
The company does everything fast. They even have a quote on the wall that says "if everything feels under control, you aren't going fast enough". And they certainly live up to it (JBTO was the output of devfort for example).
Our challenge has been to keep up with them. We have been learning from their version of an agile / scrum methodology and applying elements to our internal processes and project management. There are differences of course - one of the main ones being that they are working for a single client (themselves) whereas we have to trade off multiple clients' demands against one another. If anyone has any references / thoughts on modifying agile methodologies to multiple parallel sprints, I'd love to hear them!
The biggest manifestation so far has been introducing actual physical index cards to go with the concept of a "cardholder" (the person who wants to see a task done) - resulting in a great big whiteboard full of index cards:
Another client (who unfortunately has to remain nameless) has been doing great things recently with Amazon's Mechanical Turk (note - if you don't want to jump through the US-only hurdles for MT and want a simple interface to kick off small- to medium-sized jobs, I recommend SmartSheet).
They have rewritten large parts of duplicate content or poorly-written UGC from their inventory - at times taking up significant portions of all the so-called HITS (tasks) available on MT.
Out of this have come a few learnings - almost verbatim from the client:
- If you are doing a larger job, it's worth investing in your request form. The quality of results improves as you make the design and instructions clearer. It also makes it much easier to reject a submission that doesn't meet the guidelines.
- Give examples of good and bad submissions to show what will get approved and what will get rejected.
- Create templates for common feedback and responses to submissions and questions. Timeliness is important and this will help you respond quickly. Keep it informal and friendly and remember that it's real people doing the work!
- The community has tools for communicating and sharing information about requesters - in particular they highly value quick feedback and quick approval / payment - our client has found a nice feedback loop where the faster work is approved, the faster new high-quality work is done.
- Pay quick early bonuses for especially good work in order to get the best writers to return and spread the word.
- Go a few cents higher on your budget per HIT to get better initial quality and give you better grounds for rejecting poor submissions.
Lessons from competitors
Of course, there are loads of lessons you learn from competitors that you can't share without breaking confidences, but here are some things I can share:
We have been involved in a small forum or two composed entirely of people facing the same challenges we face. One surprising result to come out of these was a statistic that resulted in an overnight directional shift in our business.
11 out of 15 people around one of these tables was using Salesforce.com for their CRM. The evangelism was strong enough to cause us to start a Salesforce implementation literally the next day. Sometimes you can overthink these things...
It's always amazing who you can get introduced to by your 'competitors'. (This is just as true of us, by the way. Lots of things are cheap at the price of a pint!). I think it's one of the nice features of our industry that many supposed competitors are quite collegiate. The fact that any agencies or consultants who are any good are over-worked probably has something to do with it.
We have recently had great introductions to writers, designers and techies via people in our industry. Ask me about them next time you see me in the pub...
Lessons from colleagues
As many of you will know, we work closely with Rand and the guys at SEOmoz - even to the extent of our US guys sharing an office with them. Over the years there have been loads of lessons we have learnt from (and with!) SEOmoz. You can read some of the things Rand wished he knew when he started out on his new personal blog.
One small lesson that had a disproportionate impact came from reading Rand's email.
Let me rephrase that.
Rand and I were working on something and he searched in gmail and pulled out an old update email. It was a weekly round-up to all hands about what had been achieved by teams in the company that week. I got him to forward it to me.
Our guys have that exchange to thank for the fact that we now make them write a short update every Thursday / Friday for circulation to everyone. It gives us a chance to hear about things outside our direct circle of influence and contribute to conversations across the company. It's a bit like the way that the scrum part of the agile methodology mentioned above gives everyone on a project a chance to weigh in with "oh - I've worked on something like that
We are still learning what truly makes it work (in particular, I want to push it away from tactical status updates and more towards "here's what I learnt and finished this week")
I have no idea if SEOmoz even still does it, but I like it.
Just Do It
Every so often it's nice to get a reminder that you can go and learn new stuff from scratch and get things done - see for example the story Tom (my brother and colleague) wrote about how he built 7books in under 4 weeks (including learning to code).