Is Craig Wright really Satoshi Nakamoto?

I’ve been asked the question a lot: Is Craig Wright really the Bitcoin creator, known by the pseudonym Satoshi Nakamoto?




TL;DR: Who cares?

We should first question the relevance of knowing the real identity of Satoshi Nakamoto.

Those who care most about Satoshi’s true identity will probably be the tax authorities, since Satoshi sits on around half a billion dollars’ worth of Bitcoin.

The second group might be those involved in the heated Bitcoin block size debate. There has been a lot of “What would Satoshi do?” statements thrown around and I suppose having such a figure unmasked will add some gravitas. (Mr Wright, as I understand it, is for a gradual block size increase).

All of these things are more-or-less irrelevant, I believe.

Satoshi isn’t Bitcoin’s decider. Nobody is.

The beauty and strength of Bitcoin is the lack, not the presence, of a central bank(er). Being open-source and decentralised are features, not flaws. Because of it, there is no sole central figure who decides on the future of the system. The future of Bitcoin won’t be decided by Satoshi, whether that person is Mr Wright or not.

A point that I should raise is that I believe great claims require great proof. The correspondence and proof that I reviewed wasn’t conclusive beyond all doubt to me that he is in fact Satoshi. More can be done to prove it, and I know publications like the Economist have called on him to do so. Otherwise it might have very well been a cheap publicity stunt.

Lastly, even if Mr Wright turns out to be Satoshi, I’m surely very impressed by his contributions to Bitcoin thus far. I’m also excited by his potential contributions to Bitcoin in future. The thing that makes me most excited about Bitcoin, however, is that anyone can contribute, so I’m looking forward to everyone’s contributions.

In short: I don’t really care beyond my own simple curiosity if we’ve really found the true Satoshi. The fact that Bitcoin has grown, evolved and improved like it has over the years without Satoshi’s involvement is testament to me that it can and will continue to be one of the most important modern technologies of our era.

Multilingual site language list advice

I’m currently working in Indonesia (and I’ll be here for a few more weeks) enjoying the hustle and bustle from sleepy, beautiful Cape Town.

A lot of apps and websites I use are now (and understandably) showing me their content in Bahasa Indonesia. Despite finding the language interesting and easy to study, I still prefer to see these sites in English.

Most multilingual websites have standardized the location of the language/region toggle switch (usually in the header or footer).

Language selector lists done right

Here’s how FireFox did it (the right way)

firefox language list

How to create a multilingual list to annoy and confuse people

Here’s how LastPass password manager (which I adore, use every day, highly recommend and have since reached out to with the requested change):

lastpass language list

Noticed the difference?

FireFox shows the languages in their native writing, LastPass shows the translated language, the one that the app/browser/device is set to.

The correct way is to translate the content of the sites and things like the menu items (no small feat I can attest, we’re doing it right now at BitX), but not to translate the actual language list. That is, if you’re showing your website in any language the individual languages should be shown in the language of the native speaker.

Imagine the frustration if you’re on a Chinese website and you need to find “英文” (English) in the language selector; you’d need to know how to read the Chinese characters for “English”.

So instead of showing “Cine” (regognisable to Indonesian readers) or even “Traditional Chinese” (more recognizable, but still only to English readers), LastPass should show “繁体中文” (recognisable to Traditional Chinese readers, those looking for that specific language).

And then, instead of showing “Bahasa Inggris”, they ought to show the native “English” to me :-)

An even-better way of displaying it might be –depending on your design constraints– to show the language code followed by the native written format. So “ZH -繁体中文” for Traditional Chinese, “EN – English” for English etc.).

There is an extensive list of languages, as written in their native tongue with their language codes, called the ISO 639-1 on Wikipedia.

Oh, lastly, try to steer clear of country flags for languages. These are regional/country indicators, not language indicators. Traditional Chinese is, for instance, used in both Hong Kong and Taiwan, Portuguese is spoken not just in Portugal, but also in Mozambique, Cape Verde, Angola and Brazil, so it will just be confusing/wrong to use a flag.

Agree? Disagree? Reach out, I’d love to hear your thoughts!

Removing malware from first line of WordPress PHP pages


Kudos to Carel van Wyk, for helping out with this one. I’m releasing it into the wild here to help other people who might have the same problem.

The problem

There are bots trying to gain access to websites, email and social media accounts, all the time. Don’t take it personally; they’re simply looking for (usually known) weaknesses to exploit and then once they’re in, they use your account to send out spam ads, on which the spammers earn referral revenue.

It’s not limited to things like email and social messaging accounts, if a bot gained access to a website, it could send spam emails and comments from that domain.

Since a giant chunk of the Internet runs on WordPress it is logical that there are many bots that look specifically for WordPress vulnerabilities. The main things that increase their odds of gaining access to your site are:

  • Using an easy-to-crack password
  • Using a password which is used on another site
  • Using the standard “admin” username
  • Not updating your WordPress version
  • Not updating your plugins

Obviously, there’s a longer list, but that’s not the point of today’s post. Use something like iThemes Security or WordFence, review the bullets above and don’t use shitty hosting and you should be fine.

But let’s review a standard PHP malware issue I’ve seen on lots of sites.

Finding the injected malware script

If you find that things become unresponsive, the site breaks, you suddenly can’t update things or your hosting company calls you saying there are spam complaints coming from your domain, your site is probably infected.

If you look in your WordPress root or in any of the plugin or theme folders, you’ll see something like this in the first line of (usuall all) PHP files:

<?php if(!isset(...long line of crap’));?><?php


<?php eval(base64_decode('…long line of crap'));?><?php

What you should see on the first line is just the trailing <?php, the first bit of script, the long line of crap, got injected by the malware and is used to execute different spammy messages.

Fixing the mess has this resource:

Sadly, it requires a backup and chances are if you’re reading this that you don’t have a backup (or that the backup might also include infected files).

The best way is to completely kill the site and do a fresh install, but failing that, here’s a step-by-step in removing the injected script from the .php files:

1.    Make a local copy of the infected folders

The infection might be restricted to your /plugins or /themes folder, but it could be in all folders throughout your entire website. Use something like FileZilla and save the folders to your computer (like: ~/desktop/plugins/). Make an extra local copy after you’ve downloaded the site or folders, in case you break things even furhter (like ~/desktop/plugins-backup/)


2.    Download this script

Save this file, unzip it and place it in the root of your local files (like ~/desktop/plugins/ with cleany


3.    Run it

  • Open Terminal on your Mac (command-space, type “terminal” to launch it)
  • Navigate to the folder (“cd desktop” return, “cd plugins” return)
  • Run it (type in “python” and hit return)
  • It will ask you if you’re sure you want to replace the top line of individual or all PHP files and replace it with <?php (the only line it should contain), hit y and return.


Et voila! Your php files should be back to normal. Now, all you need to do is reupload the folder(s) to the website.

Hope this worked for you.


Alex Schultz lecture: Growth

Growth (Alex Schultz)

Why Bitcoin won’t work for emerging market remittances (anytime soon)


Preface: I “drank the cool-aid” early on when it comes to Bitcoin. I bought my first few bits in 2013, made some money and then lost it all again. I’m a firm believer in the technology and I get to work with some of the brightest minds on some of the most intellectually stimulating work, something I believe is the future of finance. I just don’t think it’s the future of remittances in emerging markets (anytime soon).

I’m incredibly excited about all the uses (both immediate and those yet to come) around Bitcoin, Blockchain and the incredible potential this piece of technology has. There are, however, lot of individuals and startups that oversimplify things, especially when it comes to remittances.

I dislike the oversimplication of anything and in some instances –politics, reporting of world events, science—it can be downright dangerous.

I’ll start with the simple view, the story often told and sold, and then with a slightly more detailed look at the reality.

Why Bitcoin is made for developing nation remittances*

* not the author’s view

Okay, while we’re simplifying, I’ll use simple numbers to make the maths easier.

For our example we’ll say that we have Alice, living in New York, and she is looking at sending $1000 to her brother, Bob, living in Bujumbura, Burundi.



The (completely fictional, made up) exchange rates are:

1 Bitcoin (BTC) = 100 US dollars (USD) = 1000 Burundian francs (BIF)

So, the story from the startup stage goes something like this:

  1. Alice takes 1000 USD and gets exactly 10 BTC in exchange for it, no fees
  2. The 10 BTC gets sent –broadcast to the bitcoin network in seconds and verified in minutes, not days– to Bob **
  3. Bob then exchanges the 10 BTC into 10,000 BIF (the exact equivalent of the 1000 USD Alice sent)
  4. The corpses of MoneyGram / Western Union are found in the rubble, still clutching their”5% fees” advert posters

Much amaze. Such remittance. Very wow!

** This one is the only item that is actually accurate.

OK, let’s break things down a bit and work with some real world examples.

Buying bitcoin

There are costs involved in exchanging between bitcoin and local, government-backed currency (known as fiat currency).

First, there is the cost of acquiring the consumer. Alice doesn’t just snap her fingers and get her US dollars converted into bitcoin. The current state of (or lack of) regulation and the risks of money laundering, with a dash of pressure from banks and international regulators (nasty things like the threat of terror financing) all make for a very cumbersome signup experience.

Issue #1: Sender’s customer acquisition & behaviour-change cost

Alice first needs to sign up for an account with a company that can help her purchase bitcoin, either a Bitcoin exchange or another startup that “hides” the bitcoin or the exchange and more technical aspects of the transaction.

No matter how slick this process, she’ll probably have to upload a copy of her identity document and a utility statement to prove her physical address.

This is very onerous, complicated and invasive, even if it only needs to happen once (initial account registration), especially compared to the ease of walking up to a teller at her bank or supermarket or, yes, an online transfer via Western Union, MoneyGram et al.

Also, yes, she can purchase bitcoin from a stranger using a platform that doesn’t require her to verify her identity, like LocalBitcoins, but this is the equivalent of buying cash on the black market and a) not a good idea (due to the security/fraud risks) and b) not something that can be scaled for the larger remittance market.

I’m working for a bitcoin company, we’re working very hard to lower these barriers and help people sign up and simultaneously keep the bad guys out of the system. We’ve got a long way to go, but we’ll get there.

Okay, Alice is a tech savvy person with access to things like a bank statement (proof of physical address) and driver’s license (proof of ID) and she got an account where she can convert between USD and Bitcoin. Great!

Now, once Alice deposited her $1000 the funds need to be “magically converted” from fiat into bitcoin. This happens on an exchange and is the same as the international foreign exchange market (the largest market in the world, by far).

It is simply a demand-driven market between different currencies. This means, there must be a willing buyer (“I’m willing to sell you my currency for this price”), a willing seller (“I’m willing to buy your currency from you at that price”) and enough demand, for you to exchange larger amounts of money than just a few hundred dollars, which means there might need to be more than one willing buyer and one willing seller.

Alice is a willing bitcoin buyer: she took her $1000 and said “Here’s my money, take it and give me 10 bitcoin”. Someone else (a person, company, hedge fund) must meet that order (or the order should actually already be on the market): “Hey, we’ll give you 10 BTC, for your $1000 USD”.

The US is leading the pack in terms of Bitcoin trade volume, with a very liquid market where thousands of bitcoin gets bought with and sold for the greenback. There are many people interested in buying bitcoin and many people interested in selling bitcoin; the market meets at this mid-point between supply and demand. I’ll come back to this point later, but this market is where Alice will get her Bitcoin.

But it aint free.

Issue #2: Most bitcoin exchanges charge a fee

Most exchanges (forex, bitcoin or otherwise) charge a fee for their services. They’re running a business, after all. It might be low with BTC-USD transactions, but Alice can reasonably expect to spend between 0.2 and 1% in a transaction fee (between $2 and $10 USD for her $1000 transaction).

Granted, some bitcoin startups might zero-rate their exchange fees to incentivise trading, but they are running businesses and need to earn their keep somehow (so they’ll cash in on add-ons or start charging an exchange fee in future).

Sending and receiving bitcoin

The same way that Alice needed to go through the laborious process of sending in proof of her ID and residence, Bob would need to go through the same steps to get his Bitcoin wallet (or service that will allow him to cash out his BTC).

Problem is, where Alice might be used to 4G on her latest smartphone, be an early adopter and technologically savvy, the vast amount of remittance recipients won’t be in the same, fortunate shoes. The cost of onboarding, changing the recipient’s behaviour and reaching them in whatever remote location they should be, shouldn’t be overlooked (the “last mile” of the remittance process).

Bob is probably used to the conveniences that come with MoneyGram or Western Union, where he gets to walk into a building out in Bujumbura (or even a small village in the countryside) with a reference number in hand and walks out with a wad of notes. Notes which he needs to pay his bills and buy products from his market.

Issue #3: The recipient needs to get a bitcoin wallet, exchange account AND know what they’re doing

There are some brilliant startups lowering the barriers on the recipient side and, as mentioned above, removing the layers of confusion and complexity, but changing consumer behaviour is incredibly onerous and expensive. Banks in the US are earmarking substantial budgets to educate their customers in how to use the soon-to-be-deployed (to the US) chip and PIN cards. All that will change is that shoppers will need to enter a four or five digit PIN on a pad when they use their cards (as opposed to only signing for it).

Even that small behavioural change in the rich, connected, developed world is going to be accompanied with a lot of friction and confusion.

Exchanging bitcoin back to fiat

Now, exactly as there was a digital market where Alice (or the US bitcoin startup) could trade between USD and BTC, Bob (or his bitcoin startup in Burundi) needs to go to a market where he can exchange his BTC for Burundian francs (BIF).

The snags should become apparent by slowly reading the sentence above, but in case it wasn’t clear:

Issue #4: There isn’t a Bitcoin exchange in [enter your developing country name here]

I work for BitX, a bitcoin company that focuses on emerging markets with operations in Nigeria, Kenya, South Africa, Malaysia and Indonesia, so I’m somewhat qualified in commenting on this. It’s no trivial task operating in some of these markets and there’s a good reason why many banks and financial institutions have had limited successes there. Laws, banking and financial legislation (if it exists) is completely different in every country and the effort and cost of doing business in the developing world (as rewarding as it is) is no cake walk.

Issue #5: It’s a complicated business, you can’t just copy-and-paste financial services across markets

There are some great, home-grown startups with wonderful people and ideas popping up all over emerging markets, but this is going to take a while.

And to those of us who don’t partner with enough locals, it is going to take even longer. I’m particularly fond of the Dominos Pizza story: the franchise opened up a store in Nairobi, Kenya and (very briefly) insisted in sticking to their “fresh hot pizza in 30 minutes or it’s free” rule.

Nairobi has some of the worst traffic on the continent, they’ve found, and quickly had to pull back the offer (one which, amazingly, is still in place in India).

Back to Alice and Bob and just to be clear: there isn’t actually a bitcoin exchange in Burundi. But, let’s assume there was (as it could probably and quickly change in future).

Now Bob places his 10 BTC on the market and says:

“Hey, I’m selling 10 BTC, are there any buyers out there willing to give me 10,000 Burundian francs?”

Remember, 10 BTC should be equal to 10,000 BIF which is equal to 1000 USD, the amount Alice sent and the equivalent Bob expects to see at the end of all this.

Issue #6: Extremely limited demand for Bitcoin in most developing nations

If you look at the exchange volume for USD-BTC transactions, you’ll see a huge amount of money that gets exchanged every day (lots of demand and supply, a very liquid market). In emerging markets, this is not the case.

For one, what would the average Burundian, earning only a few hundred dollars in an entire year exactly do with the bitcoin? Sure, one can send it around (virtually free), make a donation to Wikipedia or spend it at Dell or Microsoft online, but the reality is that remittance recipients want their local currency to buy local produce, not something digital and futuristic that requires an Internet connection.

Until Bitcoin becomes the global world currency, we can assume Bob and his fellow Burundians need francs, not bitcoin. If they want (or need) bitcoin, it will be in limited amounts, with insufficient liquidity to facilitate remittances at scale.

But, in sticking with my theme of oversimplified examples, let’s stick to the assumption that there was a Burundian Bitcoin exchange. Again, there are startups running exchanges in these markets, and I’ll use BitX as example.

Real world example:

Below is an actual screen grab off the BitX Kenya exchange’s open buy orders. Meaning that people already took their order to the market, saying “If you give me a certain amount of Bitcoin, I will give you a certain amount of Kenyan shillings”.


Let’s say, we were selling 10 BTC and want to receive Kenyan shilling for it.

Starting at the top line, the first 0.2253 BTC we are selling (out of the 10), will get us 34,428 Kenyan shillings, because someone said they will pay that amount of KES for that amount of BTC, that order is already on the market.

Then, for the next 0.025 BTC we’re selling, we will only receive 34421 KES and so on.

By the time we’ve sold 2.40652 BTC, I’ll only get 23850 KES for the next few bit’s I’m selling – a whole 10,578 KES fewer than my first few bits! Oh, that’s a difference of over 100 USD at the time of writing. Yes, we will lose hundreds of dollars if we sold as little as 10 BTC in a single transaction.

This means, for Bob, where he expected to receive the equivalent of $1000 USD in his bank account, using Bitcoin, he might lose out on more than $500 USD (in Burundian francs) for the transaction!

Issue #7: Low order volume/liquidity translate into a horrible exchange rate, much worse than fees

Now, I know there are holes in this example, since if we sold more bitcoin, more buyers might come to market and buy it from me rather than just the open orders on the market, but it loops back to the question: what the hell will they do with the bitcoin and what will incentivise them to trade it in the first place?

Bitcoin startups deal with this by

  • attracting traders with low fees (a good idea) or
  • “selling” the advantages of Bitcoin to the people of that specific country and telling consumers to invest in bitcoin (questionable financial advise; bitcoin is closer to a currency/commodity, so it’s pronounced “speculation”, not “investment”, something one should probably not be encouraging in poor countries) and
  • guaranteeing a fixed exchange rate (also a bad idea, since the only way you can guarantee an exchange rate is if you manipulate the market and use operating/investor funds to artificially prop it up).

So, what can Alice do instead?

The incumbent

I ran the numbers using MoneyGram’s Online Cost Estimator, for our example of Alice sending USD to BIF in Bob’s hands:


So if Alice sent 1000 USD, Bob would (after fees) end up with 1.5m BIF, which equals 975 USD, in his hands (or 2.5% of the sent amount). Not bad, considering he picked up the cash (very useful to him and the people he’s buying from) at a shop or secure building and Alice completed her transaction online. Incredible, compared to the alternative of onboarding Alice, onboarding Bob, charging 1% on the USD to BTC and another 1% on the BTC to BIF and losing out hundreds of dollars due to low liquidity/bad exchange rates.

A few more points

The true cost of remittance isn’t the digital transmission of funds. Sending money between the US and India or Mexico (two of the largest remittance corridors in the world) is a slick, fast process which takes minutes and can be done entirely online. This isn’t the place where Bitcoin is going to make big waves. The true cost of remittance comes in at the “first and last mile”, getting people to purchase it and to get the equivalent money (after fees) delivered on the other, often remote, side.

Digital remittances exist in a lot of places and in places where they are inefficient, they are inefficient for a reason. That reason might be low demand, low liquidity on the fiat forex market, greed and yes, banks and remitters trying to maximising revenue, but Bitcoin won’t fix these anytime soon.

Lastly, Bitcoin is technical and complicated, but it isn’t that technical and complicated. If it was that much cheaper to remit money using bitcoin between two countries using bitcoin (at scale) MoneyGram and Western Union (or the banks) would have started doing so to stamp the competitors out a long time ago.

Bitcoin and the blockchain can and will drastically change things in emerging markets, I’m just not convinced that it will be through emerging market remittances (anytime soon).

Let me know if you disagree or if you think I missed anything ;-)


How to make an international transfer with TransferWise

Having made my fair share of international transfers, I was mildly blown away by the speed and simplicity of TransferWise. For this very type of transaction, it blows banks, Paypal and yes (with the limited liquidity in markets) Bitcoin out of the water and is part of the reason I’m convinced Bitcoin isn’t a viable solution for remittances yet.

I decided to take step-by-step screen grabs of the process. In this example I sent 100 GBP from the UK to the US.

  1. Sign up with TransferWise (2 minutes)
  2. Enter the currency you’re sending it from (GBP) and the currency/recipient bank you’re sending it to (USD)
  3. Review the details of the transaction:
    Transferwise review
  4. Make deposit to TransferWise’s UK account via online banking or debit card.
    (note that online banking would have taken a day or two to reflect in their account, so I went with instant debit card)
    Transferwise online banking
    Transferwise debit card
  5. Ta-daa! The funds appeared in the recipient’s bank account in two days.
    Transferwise success

My alternatives would have been:

  • Online banking from UK bank account (too costly, bad exchange rate)
  • PayPal (bad exchange rate, high fee)
  • Bitcoin (inconvenient for recipient without Bitcoin account with linked a US bank account)

Obviously, there are times when Bitcoin would be a brilliant mechanism (like if it got sent to a place where bitcoin is trading at a premium or for micro-transactions), but I’m convinced I’ll use TransferWise again.

Give it a go and let me know if you got stuck!

A farewell to PayFast

This is a combination of my thoughts on quitting and the letter I sent to my colleagues announcing my resignation. I don’t want this site to really be about my personal life, but so much of what has become my personal life is intertwined with what is my professional life. 

It is with a heavy heart –and an industry-appropriate sad emoticon– that I’m announcing my resignation from PayFast :-(


I returned to South Africa (and more to the point: ended in Cape Town) in September 2011, after a seven year stint abroad. I wasn’t quite sure which exact direction to head into and thought a year in Cape Town would help me clear my head. I was seriously considering the MBA programme at the University of Cape Town and even started studying for the dreaded GMAT, when I rather fortuitously, met Jonathan Smit (PayFast’s founder) on Twitter.

After I suggested we go for a coffee or beer he said that an “after-hours beer is a great idea since people look at me weird when I drink beer in the morning”. Long story short, a few beers in, he suggested that I apply for a position at PayFast. Some call it luck, some call it networking, some call it fortune (some call it three pints of lager!), but a week later I started my incredibly fun and meaningful journey with PayFast.

Now, after nearly four years in this amazing city and office, things have changed, improved and grown up considerably. What started as a penny-pinching startup has evolved into a brilliant high-growth scale-up. When I started I always had to explain to industry peers what exactly it is that we do, now we’ve become a known and trusted South African industry name. I was the fourth employee on the company’s payroll, now it’s not uncommon to that amount of new faces sign up in a two-month period (we’ve added six in the past two months, actually). We’ve built some amazing things; I’m particularly proud of the time when we became one of the first payment gateways (in the world, in fact) to facilitate Bitcoin payments (nod from the Founding Director of the Bitcoin Foundation below).

Scrap that: I’m proud of all of the things that went into growing the PayFast brand into what it is today. Hundreds of fun hackathons, meetups and workshops were held, heaps of meaningful partnerships and friendships fostered. It has been absolutely amazing.

It is then for very personal reasons —all external to my work– that I decided to draw this chapter to a close (and one where, flatteringly enough, Jonathan and the company fought hard to keep me around). It unfortunately boiled down to me needing to pursue some opportunities with a more international footprint to help me realise my near-future goals and aspirations.

That said, I don’t see myself leaving Cape Town permanently just yet. I am taking a short trip to the US and Bulgaria (the former to attend SXSW and a friend’s wedding, both while working remotely, the latter for some much-needed time off in Bulgaria with my girlfriend).

It has been a very difficult and emotionally complex decision, but I’m feeling very positive about the future (mine, sure, but mostly PayFast’s) and hopefully I’ll see you at one of the many awesome Cape Town tech events!

It’s been a blast.


Startup Christmas

Excerpt from a mailer that went out this week (not composed by me):

Startup Christmas

#StartUp Christmas was conceived by Jens (Shopstar), Alex (StoneTree) and Werner (PayFast) over a glass of wine at an end-of-conference party.

Conversations, which started with the classic “Sjoe! What a year its been!“, ended with a serious discussion on how to best bid it farewell.

The biggest challenge –okay, not exactly the biggest, but a significant challenge none-the-less– is that its really hard to have a big end of year party with just three people. Just adding beer to the people you have be staring at every day doesn’t really do much.

But – add a few like-minded startups, a pool, a DJ, some fairy lights, a Santa – and BOOM: #StartUp Christmas is born.

We hope that you can make it. It’s going to be fun – much more fun than you, the two other devs and shot of tequila each. Promise.

Go check out the website: RSVP, while you’re there. I can’t wait.

Separate your transactional and marketing emails

Companies send out two very distinct, very different types of emails (or at least, they should).

Transactional (or business) emails relate to the specific individual: welcome emails, invoices for payment, lost password requests, scheduled system downtime and the like.

Marketing (or newsletter) emails relate to things that the individual might find interesting such as new features, opinion pieces from your blog and new features.

It’s about reputation – baby!


Mail providers uses something called a sender reputation score to determine whether or not to display your emails to the recipient. Just because an email displays as “sent” on your side, doesn’t mean that the recipient actually received the email. The reasons behind this are all rather logical: they are in place to make sure that spammers (and dodgy operators in general) will lose reputation and you won’t be bombarded with unwanted mail (which will either never appear or show up in your spam box).

I won’t go into it too much, but your reputation is based on two things: IP reputation (or server reputation) and domain reputation. Some ISPs look focus more (or exclusively) on either of them, so it’s a good idea to think of your reputation holistally.

Things you should consider

Different mail providers have different standards, but there are a few things that are very important to keep your reputation high:

A squeaky clean server


Don’t ever bother signing up with a dodgy mail operator. If you are on a shared hosting environment and use the same servers for email, you’re looking for trouble, since other people might send spam from other websites on that server.

If you don’t know where to start, sign up with MailChimp for marketing email and Sendgrid for transactional email. They know what they’re doing, provide good support and have thorough documentation to help you do things right.

No spam complaints


While you’ll never be able to completely control this, you need to try and avoid spam complaints as far as possible. This means two things: 1. Write meaningful content and 2. Make it easy to unsubscribe.

The more relevant (or value-adding) your marketing messages are going to be and the more control you give your user, the fewer spam complaints you’ll have.

You should also give your users the option to opt out of certain types of transactional emails, as Twitter allows you to do:

Twitter email controls



Sending out large attachments (or certain types of attachment), having suspicious subject lines (or content) and badly formatted HTML emails all could lower your reputation.

If you’re using a quality email service to compile your emails, providing valuable content and not trying to push Viagra on people, you should be good.

Segment, segment, segment


Segment your email recipients as much as is reasonably possible.

If you sell three types of products (say cloud backup software, booking software and e-commerce software), you don’t want to tell hotel owners (using your booking software) that there has been a version update to the e-commerce platform.

You don’t want to send a newsletter about changes to online shopping delivery laws in your country to everyone/if it doesn’t relate to the cloud backup users.

You could –depending on your business— further segment your subscribers (for marketing mail) and users (for transactional mail) by location, industry and the like. This will help greatly in refining the value-adding message for the recipient.

Use different email addresses (but don’t go overboard)


Some companies use different domains to distinguish between their marketing and transactional mail, but this is, in my opinion, a bad idea.

Simply create separate email addresses like (for transactional mail, which/if it has to be delivered) and (for marketing mail).

With modern mail applications you can set separate reply-to addresses (so if someone replies to your newsletter email, that it goes to something like

This helps your customers to filter your emails. Users might want add an email filter on your newsletters, since that kind of information isn’t critical to operations (I like to keep a clean inbox and get to my newsletters when I have downtime).

Don’t make the mistake of sending emails from too many “from addresses) like using the first names of every newsletter contributor. It might give it a personal touch, but if the recipient has added to her address book, particularly when it’s filtered it so that all emails from “Tom” should go to her newsletter/social inbox, you’re going to piss her off when you clutter her inbox with an unfiltered email from and 

Authenticate! Use DKIM and set an SPF record


I should have mentioned this first, as it is one of the most important things to help show the recipient’s mail provider that you’ve authorised certain servers (say, those of Sendgrid) to send emails on your behalf.

You’ll notice that some emails (usually marketing emails) you receive have the “sent via”. Here’s a blurb by Google on it.

In some cases the mail might even end up in your spam folder with a warning from your mail provider saying that they couldn’t verify the sender of the email. The following email (from the otherwise superb Buffer)

subscription without SPF records

It’s a remarkably simply fix, you’ll simply need to reach out to your email (or hosting) provider and ask them (or do some light browsing through their documentation).

Two blurbs by Google: the first on Sender Policy Framework and the second on DomainKeys Identified Mail.


You can check your reputation free on SenderScore (should be taken with a grain of salt, as not all mail providers use the same formulas)

To test a whole range of things associated with your domain, check out the excellent MX toolbox.

In closing

Authenticating your email can be a little complicated, but it is a necessary step to make sure your emails get delivered. Make sure that you use a reliable/clean server to send messages from and always give people a way to opt out of your marketing messages (and if possible, a way to control the amount of transactional messages).

Lastly –and this should be at the core of what you do; not just your emails—only create and share things if they add value.

Email is a tricky thing to get right and you can go down lengthy wormholes to get things right, but it is far cheaper to get it right right now, than to spend the extra time and money (and lost resources) by having mails go unread or servers blacklisted.

Lessons learned at the Lean Startup Machine

The Lean Startup book, by Eric Ries has had a really big impact in the way I think about and approach various projects since the day I first read it. It’s applicable not just to work projects, but also for one of the many side projects, charitable causes and ideas I’m tinkering away at in my spare time.

Lean Startup

At the crux, it is about focusing on getting things done quickly: shortening the development by doing clever experiments and pivoting (or exiting) early to avoid spending a very costly amount of time (and money) working on something that is likely to fail.

It’s not as much about failing –something I do think needs to be seen in a different light—but it is about failing faster.

There’s obviously more to it: good management & accounting, building, measuring, learning…just go ahead and read it (email me, I’ll lend you my Kindle copy). I can’t recommend it enough.

Lean Startup Machine

I was lucky to attend the very first Lean Startup Machine in South Africa here in Cape Town (and plan on doing so again in future, hopefully in other cities). LSM are weekend workshops, held all over the world, where you are taught how to build something people (other than you) actually want. You get to build (or break) an idea or business in three days.

The majority of new ventures fail because they build something that not enough people find valuable, followed by running out of money (but the one feeds into the other). There are 101 startup post-mortems, worth a glance, here.

For me it ties in nicely with any idea you might have (not just founding a tech startup).

I was part of a superb team and we worked hard all weekend on refining our idea on how to create a simple platform to help bridge the gap between the demand and supply of the superb art this city (and country) has.

I’m very proud to say that our team was chosen as the winner of the event, but I’m getting ahead of myself…


This isn’t much about the actual event (one I highly recommend to anyone working in a fast-changing environment) or about us winning (the takeaways would be the same), but rather about what my thoughts are as an attendee.

Your team matters


Our team consisted of two people working for startups (myself and Nic, in e-commerce), a bright artist (Clara), a developer with a few small businesses (Nathan) and someone working in finance (Jason). There was a lot of synergy in getting to our mutual goal, but we were diverse enough to constantly challenge each other’s ideas and ensure that we didn’t fall into the groupthink trap.

The best thing you can do is to get out of the building


There are a lot of valuable lessons you can take from the weekend, but by far and beyond the best one is to get out of the building (and asking people the right questions). We learned an incredible amount by just talking to potential customers and artists. Knowing who your (potential) customer is something you should

I know that Steve Jobs said that “people don’t know what they want until you show them“, but that is only great advice if you are actually Steve Jobs. If you’re not, get out of the building and talk to people.

Timing, luck and hard work matters


Yes, once you have a winning idea (and a winning team) you have to work your ass off. It’s relatively easy to do so over a weekend of camping out at a central meeting point, so not really relevant for LSM, but new ideas and products take a lot of hard work. Something they also take is a large element of good timing and pure dumb luck. Sure, you might have worked hard your entire life and maybe have the accolades to show for it, but you were still born, by stroke of luck, into the circumstances that allowed you to get educated, work and meet the right people to carve your way to the top.

I was lucky to get into the right group, we were lucky to have a bunch of great, hardworking members and we were lucky that we could get the right idea and right advice from the right people.

Mentors matter


I’m a firm believer of learning from those who have already learned the lessons (and bumped their heads a few times). Obviously at an event like this there will be mixed advice, since every person’s experience up to that point of time was unique to their story, but there were a lot of superb feedback and ideas from them.

This is something I’m definitely going to spend more time incorporating into my own career, but more on that later.

Winning at LSM doesn’t (really) matter


Again, I was absolutely stoked at our team winning the event (just look at our goofy grins), but the lessons would have been as relevant if we didn’t win.

We got great press out of the event, a couple of publications even reported on it (and there was a lot of interest from the art community), we won some cool prizes but in the end we still had to pull the plug on the entire thing. As much as we wanted to pursue our platform –The Artery– we all still had our day jobs and couldn’t contribute enough time to build it into what we all envisioned it to be.

But that’s OK! The incredible value from the team building, challenging your ideas, learning how to ask the right questions and just hanging out with a diverse group of attendees and mentors make it more than worth the admission price.

If there is ever a Lean Startup Machine in your city, go ahead and join it.