Lean Founder

Semi-autobiographical, in no particular order, being a #LeanFounder:

You don’t need that new laptop, a refurb will do, but your current laptop is fine until it dies.

Plain eggs on toast is cheaper than eggs benedict, and better for your waist.

Walk whenever you can.

Traveling for a meeting? Make the rest of your time there count.

Don’t travel at peak times.

Don’t go to every event, meetup, or conference.

You will be paid the least, and probably last.

Make breakfast at home, include lots of fruit.

Swim, ride a bike, and do Yoga, Pilates, or Taichi.

Sleeping on friends floors is cheaper than renting an appartment, but not as good for your sanity.

Take at least one day off a week.

Stop drinking lager, it’s more expensive than Ale or Whisky, and it’ll only give you a hangover.

You don’t really need that premium Spotify account.

Keep on top of your accounts, keep them simple.

Actually learn how to make Ramen.

Ask for help.

Answer others calls for help.

Listen to everyone who offers advice, but make your own decisions.

Stay hydrated.

Don’t buy Photoshop, Pixelmator is like a modern Photoshop 4.

Don’t buy Microsoft Office, you can get it for free from Bizspark.

It’s not about you. But your startup is a reflection of you.

Don’t expense everything, when it comes from your pocket you might reconsider that taxi ride.

You don’t need expensive business cards, order a small run of Moo cards. Iterate.

Understand your termsheet, it’ll save you in the future.

Understand your cap table.

Know when to say ‘fuck it, I *do* have time for this [important personal thing]’.

Design your own logo, you might surprise yourself.

Don’t get held up by pixel perfect design.

You can iterate your copy.

Leave your admin till the evenings, spend the time you have with your team on product.

Carry a toothbrush and a spare shirt.

Use coffee shop loyalty cards.

An employee will cost you 20% more than you planned.

You don’t need an Aeron. Yet.

Cut your own hair. Or don’t cut it at all.

Pay your bills as they arrive. Letting them stack up isn’t good for anyone.

You can buy your Apple Care anytime in the first year. Wait.

Keep that FaceBook tab closed.

Don’t be concerned with inbox zero, it’s replying that’s important.

Do your VAT (tax) returns quarterly.

Sublet office space or desks, better still join a coworking space instead.

Have an occasional luxury.

Github, Dropbox, Google Docs, and Skype are your friends.

Keep your email short, your time, and theirs, is scarce.

Get to the point.

If that meeting opportunity happens, jump.

Know your shit. Read everything you’ve written, and read it again. Then write it again. Have answers. Be considered.

Use your competitors’ products. This is free research. They are learning and experimenting for you.

Use your own product. Otherwise WTF?

Have a note for everything. Organise and reduce these. Every thought was generated for a reason. Track this.

Have a spreadsheet for everything. They are guides. Know your numbers. Iterate these. Model scenarios.

You do have the time, stop watching reality TV and soap operas. In fact, sell your TV.

Make sure people remember you. Put a real photo of your actual face on your card.

Failing is an option, but not an aspiration. Understand why you failed. Don’t do it again.

Use that free Google Adwords credit they send you. A/B test. Learn what converts. Test your branding and messaging.

Start reading financial blogs.

Stop reading Reddit.

Learn your stakeholder’s vocabulary.

What do your stakeholders drink? You’ll know if you’ve spent time getting to know them.

Blog or tweet about what you learn and you’ll learn even more.

Spend time with other founders. Only they will understand what you’re going through.

Keep your shoes clean.

Be as transparent and upfront as possible, then you can have proper conversations rather than games of guesswork.

Your lawyer and accountant have a lot of knowledge, but you can help them learn too.

Hold on tight.





Update: I stopped writing this post, it got too long, and I got bored. Also, it looks like Meebo might have just pwned RockMelt anyway, with their MiniBar. This is an ever-changing space, and will require a followup post!


It’s too early to form a proper opinion on this one, but here’s some initial thoughts.

From CrunchBase: (read more)

Rockmelt is a social web browser built off of Chromium and boasts deep integration with both Facebook and Twitter with it’s “Edges” which are filled with friends that are online and feeds that you follow complete with update badges.

Completely achievable with a browser plugin? I think so.

It also sports what some believe to be the next big search revolution. Unlike Google’s universal navigation field, RockMelt has a dedicated search field that shows complete search results as a drop down overtop web pages. This is helpful when switching between different search results from one single query.

So the search is displayed over the page you’re on. Again, this is just a plugin, right?

I don’t know of any plugin that got almost $10m investment. But a browser, ooh, now that’s compelling. I wonder which one have the widest adoption, the easiest path to become a user, the smallest change in behaviour required. The answer these days, is a webapp, Not an installable piece of software. (A discussion for another time maybe) But for the sake of this argument, I would bet that plugins have lower barrier to entry, technically, but emotionally, people are more likely to be compelled to try out a new browser.

So it’s a browser with FaceBook, Twitter, and some kind of clever search thing going on, right? Well yeah, basically.

UX and bugs aside (those are solved with time, iterate and deploy, small and often), there are a bunch of interesting things to note. I’m not even going to touch on the so-called social features. What I’m going to talk about are the decisions they’ve made so far, and where I think that means they’re heading.


The integration with FaceBook goes too far in my opinion. The first time you open the app, you’re asked to connect it to FaceBook. Every time you open it it connects to FaceBook.


Not only does this slow you down, (not to mention being offline?), but let me put this simply, to use your browser, to access the web, you have to first login to FaceBook. As some have mentioned, this may just be how things work now, we give up some data for a perceivably more valuable return, we think we’re the ones winning.

That is basically Google’s business model. It’s also FaceBook’s business model, but RockMelt’s? Aren’t they giving it all to FaceBook? Anyway, that’s just half of the story. What if FaceBook is down? You can’t browse the web? As I tweeted:

when #facebookdown, taking Facebook Connect & fb social plugins with it, how do you log in to all those other websites then? now do you see?

This potential single point of failure is just unacceptable. Anyone building a browser in 2010 should know better. No doubt this will change. This probably won’t matter to most ‘normal’ users until it’s a problem. When you’re at a gig, you only notice the sound engineer when he fucks up. But will a ‘normal’ person ever use RockMelt? No. Unless FaceBook buy them. Which may well be the plan here.

Your data

Where is it?

Well some is in the browser: browsing history, bookmarks. Is it being passed to FaceBook? I doubt it. Why would they? Access to the data is a reason to acquire (and a reason FaceBook would build a browser, like Google with Chrome, extra reach).

Some of your data is still in FaceBook, or Twitter. Which is fine I guess. As fine as it would be anyway, at least.

Is RockMelt going to do anything clever with these two sources.. now that’s a question. With the permission models that we all accept now, apps are able to do things that would have been considered evil if it was based on cookies. We ticked the terms box, right? Yes, there are technical, and legal, things to consider here, but essentially there is an opportunity to do something cool, and now your users are ok (technically) with you doing it.

Certainly what you’re not getting at this point is any sort of backup or sync of your data to a service in the cloud. Why not make all your browser bookmarks FaceBook Likes? (ok, a few reasons, it’s just an example of integration)

Why a browser?

So you want to connect to the places you hang out most (for some reason). And you want to let something watch you the whole time while you’re online, for some sort of benefit (that might not be clear yet). And you probably spend 99% of your time in only a few places.

I guess it makes sense to use a browser that allows you to do those things. But this comes with baggage.

Firstly, you have to install it. Who does that anymore? Installed apps have to be updated. Your users may be on several different versions of your stuff. Rolling out fixes takes quick action on your user’s part. And bandwidth to deploy. And then there’s the various OSs that it might be used on. Aren’t we favouring webapps to solve these problems?

How about a compromise, a browser plugin. Sure, you could do all the same stuff, while not having all the same issues. But plugins? Until they’re called apps, people won’t care. There’s a thought..

Meebo have just done this, with the Meebo MiniBar. What took so long!

Additional: (just sayin’.. look how close we’re getting to how all this should be!)

Notice the Meebo bar ON FACEBOOK OMG! Surely I’ll be able to drag, drop, share, blah.. pretty soon. If people do that anyway?


RockMelt could simply be a webapp. We have a few things already that do most of what it does.. Cliqset, Brizzly, to name a couple. What does it take to do basic browsing inside it? Or for a plugin (app) to work closely with it’s web-based parts to bridge this gap.

The reason I would make it a new browser is easy. You’re always using it, without even thinking about it. Assuming you’ve installed it, set it as your default, and stop clicking that Firefox icon in your dock. When you occasionally need the thing this new browser does, the button is there. But most of what it does can be done on the FaceBook or Twitter sites themselves, so as long as you don’t forget that they exist, what do you need a special browser for?

We should be abstracting the browser layer away entirely in my opinion. As we’ve basically done with the OS. RockMelt is highlighting that most of what we do is on the web itself.