How to make millions online?

I have this ongoing fantasy of coming up with the perfect internet side hustle. After a blinding flash of inspiration, I set it up. It gets people visiting due to my fantastic SEO skills, and it ticks over with some high-value affiliate or advertising business that I don’t have to do anything with. I sip Mai-Tai’s on the sidelines and watch my bank balance increase…
Unfortunately, this fantasy has remained just that, but it doesn’t stop me from being interested in creative ways to make money on the internet.

When I’m online, I look for examples of these to see if there is an opportunity for a localised version for South Africa or something I could do better.

I recently went on holiday and wanted to confirm whether or not I needed a different plug adapter for when I was there. While searching, I saw a model for a business that I thought was genius. When you search for “Maldives power plug”, you’ll see this site comes out as number one in the Google rankings.

Good SEO!

Once you get on the page, it has a dropdown menu which outlines the country you live in and then gives you specific information about the electricity system in the Maldives and how you might necessarily have to adapt what you’re doing to use your appliances. It answers all the relevant questions about the local electricity situation in detail. It is genius that they don’t try to sell you their crummy product at this stage. They integrate with Amazon, your trusted retailer. In doing that, they increase the likelihood that you’ll buy:

  • they’re recommending a product from Amazon, so it doesn’t look like a sales plug
  • it has reviews, so the product must be decent (which would be harder to prove with their product)
  • the customer doesn’t have to worry about buying something from a no-name site that might not arrive. It has the trust halo from Amazon

For the company, this works because they don’t have to manage anything. If you drop ship something, you must control the dropshipping and customer service. You must maintain the website with Amazon affiliates: how you promote products and how many people you get to the website. That’s all you have to do. This business has no marginal costs besides your SEO challenges to get you to the top of the list. That’s great because if you get it right, you sit back and reap the rewards. However, knowing that the top 3 search results get 75% of the clicks means you’re nowhere if you’re not highly ranked.

The website also encourages you to get all the products you require for your holiday by providing a checklist of what you might need. Suppose this is your last Amazon shop before you go on holiday. In that case, this could include your travel pillow, holiday read or any other last-minute purchases (which will bump up your total bill Amazon bill and then provide more revenue for the site through Amazon commissions.) The product niche is excellent for this; you’re thinking about your holiday and what you need, which prompts the shopping process. If you’re buying a $50 product, it’s only $2 in affiliate commission, but the objective is to increase the overall bill and make this off many people.

They’ve extended it as much as humanly possible. If you look down the side, they’ve got all the countries you can find out about relevant power plugs, so you can see it is an extensive resource. The site is localised in numerous different languages, as you can see across the top of the site. They’re making the reach as comprehensive as possible to get as many people into their funnel as possible.

I think it’s a great model. It’s relatively hands-off once you’ve got it set up. Getting into the niches is complex, and maintaining your Google rankings to get the traffic required to make this model viable is a struggle. And it’s not the only player in this market. There are a lot of other people who are looking to do a similar kind of thing if you look at the search results.

Finding a profitable niche is difficult. However, there is a big opportunity if you can find something where there is an information gap that you can create easily templated information across different countries and languages. Once you have an audience, they can be monetised in many ways (but advertising/affiliates is the easiest). I’m not going to start up in competition with the power plug adapter mafia. Still, I think it’s an elegant and intelligent way they’ve approached making money online.

Are there other case studies or examples you can think of to get me to my Mai-Tai dream?

How coaching helps you improve quicker

You’ve heard that your brain gets worse as you age, and your neuroplasticity disappears. As a result, as you grow older, you need to do more Sudoku to keep you intelligent and stop the brain rot. By consistently challenging yourself and learning new things, you keep your synapses firing and stay sharp. 

But learning and doing new things can be a pain in the butt and not as much fun as watching Netflix. There are a lot more people moving towards coaching at the moment, and it seems to help people shortcut the process of achieving their goals.

I have had two recent experiences with trying new skills: in both mental and physical areas. I had a coach and a group of fellow students which made a real difference in how I approached the process and my success.

Informal Wordle coaching

As many people are doing, I have been trying to solve Wordle every day for the last couple of months. Unlike many people who have taken it up, I’ve never been able to do Sudoku and crossword puzzles. They frustrate me, and I don’t know how to solve them. Interestingly, Wordle is not just a word puzzle; it’s more of an information theory puzzle you’re learning and implementing through a word puzzle (or at least that’s how 3Blue1Brown sees it). My wife is a poor speller but is much better at Wordle than I am (and I’m a pretty good speller), and my daughter is doing Wortel in Afrikaans (which she has only studied for six months). It’s not all about fluency in a language.

I have managed to improve through a couple of tips from my wife/coach. Suggestions on how to structure your guesses and encouragement have been very helpful. Another part of my (modest) Wordle success is having a Whatsapp group where people share their wins and I have the same daily exchange with my daughter. That spurs me to do it even when I don’t feel like it. I also feel joy when I complete it in 3 or do it in a couple of minutes, and I can show off to people who will be impressed by it.


I’m now at a level where I complete it often, and it’s interesting to see the progress that I’m making in it due to having a coach and an accountability team. I haven’t yet gone into a new word every day to start with, I still rely on “ADIEU” and “STORM” as my first two words, but it’s definite progress.

Using a swimming coach

The other thing where I can see progress is through swimming. Swimming is quite weird because it is one of those practices that once you learn it it becomes somewhat automatic. It’s like walking in that it’s something ingrained and semi-instinctual; you learn it as a kid and then forget about it.

I have always swum like a bear. Very inefficient, like my father. There is a lot of splashing but not a lot of grace and not much progress through the water. I swim every so often at the gym because I think it’s great exercise.

I am a big nerd and reader. I have, in the past, read books, watched YouTube videos, and tried to improve my swimming by taking the best practices and implementing them on my own.

To be honest, this has been variably successful. It’s complicated to conduct an audit of what your issues are and then synthesise all the information to create an action plan.Over the last couple of months, I have had a coach who helps me, and the difference is incredible. A coach can look at you and take you on a structured improvement journey. 

I would swim a kilometre in half an hour. After the first lesson, I dropped five minutes and then an additional one minute in the second week. I was stunned at the immediate impact, and my performance continues to improve.

It’s hard to keep all of the new processes in your mind about what you should be doing and what is essential. My coach gives me just enough feedback. We’ll concentrate on only one improvement at a time until it is improved and I have a “body memory” of it. We work on fitness in conjunction with technique, and he takes videos that we look at afterwards, and he dissects where I’m going wrong. I swim every Sunday with my wife, and walking to the gym is an excellent forum to discuss technique. I also discuss what works and what doesn’t with an old friend of mine. We have healthy disagreements about an approach and what is working for us. I’m invested in the process and have a lot of people to discuss my progress with (e.g. I had a debate on the weekend about whether I should use fins or arm paddles)

Some theory behind coaching

If I look at Google trends, I can see that the idea of coaching has nearly doubled over the last five years .

The idea of coaching keeps popping up in what I’m reading. Some people say that the decline of personal tutoring (replaced by ordinary education) has resulted in the lack of any new Einsteins or decreasing geniuses ( a counterargument is that that was just one element in the overall process, the problems are getting harder and its difficult to attribute everything to tutoring). However I believe that individual tutoring can advance you much quicker (but its not a very scaleable solution)

The number of acclaimed scientists divided by the effective population

BJ Fogg has his behaviour model which talks about your ability to change a specific behaviour.

Using his theory Behavior change = Motivation x Ability x Prompts.

You can see that a coach is really helpful across all of these parameters, they can help change:

Motivation- by encouraging you and having a designed process.

Ability- by its nature coaching is designed to improve your ability

Prompts- they’ll know what stage you’re at, and what you should do next to progress.

Coaching as part of a life or business coaching practise has become a lot more prevalent lately. What interests me in the audit of people who were receiving business coaching was the focus on experience and methodology and the lack of interest in certification.

This corresponds with my perspective: you want to know someone has done it before and has a replicable process for going through the challenges with you.

What we’ve learned about coaching through Now Novel

Coaching is a factor we’ve found with Now Novel, too. When people are writing a book, it’s uncharted territory, people don’t know what to do, and it’s beneficial to learn from someone who has been there before. You can benefit from their mistakes, so you don’t have to make the same elementary mistakes. 

We find that people who have a coach are much more likely to complete a book. Perhaps there’s a causation/correlation element in that you’re willing to commit to a coach if you’re serious. Writing a book is also different from swimming in that it’s a creative enterprise, there isn’t a clear path that you have to follow to achieve your aim (you and I can both draw a frog, but we don’t have to do it in the same way). The fact that it is a personal journey and the path is open to interpretation means that it’s uncharted territory (and can be worrying to know how to achieve it). If someone tells me I am a crap swimmer, it’s less offensive than someone telling me my novel sucks as your creative output reflects on you personally. There is some form of therapy to helping people achieve their creative goals, it is a way of defining who you are. Having a coach allows you to share the germinating ideas in private and give you that motivation. It’s great to have an accountability partner that can show you what to do and help you understand the process in more detail. When you’re stuck, they can give you the necessary solution to let you take the next step. We’ve found that it’s helpful to have a group of likeminded people to work with. Its great to know that there are other people in the trenches with you and this group helps spur you on when you don’t feel like it.

What I take from this analysis is:

  • personal coaching is becoming more of a thing (which is probably a fashion as it once was more prevalent)
  • its great to get help from someone who has done it before and knows the processes
  • qualifications aren’t as important
  • having a forum/people to share the journey with is helpful
  • coaching helps with your motivation, ability and coaching sessions trigger you to practise

You can do a lot more if you have a capable and wise coach who has a process for advancing your skills. And I’m intrigued to see how I can apply this to other areas of my life.

Why delivery is crucial for ecommerce, and what to test

Even though I’m nearly 50 years old, I still have a childish enthusiasm for trainers.

Over the last year, during the lockdown, when I couldn’t even wear them anywhere, I bought a couple of pairs unavailable in South Africa (because there is extra joy in having something unique).

I bought these from two different platforms, and there was a distinct difference in the way they messaged and provided delivery that I found pretty insightful.

Buying from Farfetch was excellent:

  • They messaged me consistently over the process.
  • My delivery was via DHL, and it took 3/4 days to arrive even though packages were coming from the US and Italy to South Africa.
  • Returns were painless. I had something that I needed to return, and I could send them an email to process that. DHL came and picked it up, there were no questions, and there were return postage stickers in the box to make it easier.
  • They even made me feel like I was doing something for the environment (or at least offsetting any damage) by taking advantage of their climate conscious shipping

It was smooth, painless and easy. I really couldn’t fault the Farfetch experience, see examples of their communication below.

Buying from StockX was a completely different experience.

Stock X is closer to eBay (in that they have a network of sellers) but they’ll review and ensure it isn’t fake, and then they’ll send it to you. But they set no expectations about when they will deliver things (as you can see in the below email).

As a consumer, when you want something, you want it now.

I was waiting impatiently for my product to arrive at StockX to be validated. Unlike Farfetch there was no information, but they said it would take five to seven days. It took longer than that… (about 3 weeks to get to me) .There was no messaging in the interim. It was painful.

And then once it came, it wasn’t delivered via DHL but it was a normal budget delivery, with no easy returns associated with it.

There is a huge difference in these ecommerce delivery experiences.

A classic Jeff Bezos quote says; “It is difficult for us to imagine that in 10 years from now customers would want higher prices, less selection or slower delivery. I believe in the durability of these pillars gives us the confidence required to invest in strengthening them.

Amazon has ruined everything in terms of setting expectations around delivery:

  • that it will be free
  • that it will arrive quickly
  • And that returns will be painless

Those ideas have been internalised by customers, and they have high delivery expectations.

I have run numerous A/B tests across retailers in different countries and in various verticals around delivery.

Here are some principles that are useful for an understanding of eCommerce delivery and opportunities for testing.

1. The strip that you run across the top of your page.

The value strip is essential to message your delivery promise and explain your offer. It is a persistent message that visitors see on every page of the site, so it really reinforces your message. It give you the opportunity to explain your free delivery proposition.

Many customers are interested in how click and collect works (for omnichannel retailers) and your free delivery threshold (i.e. how much do I have to spend to get free delivery).

Having this communication in a prominent position that runs sitewide can be very helpful.

The above strip promises at least two weeks until I receive my order; that is a lifetime for the hungry consumer.

Contrast that with the Book Depository below consistently reminding me that I’ll receive free worldwide delivery wherever I go on their site.

Messaging on the value strip is an area of messaging where you can get great rewards by clearly explaining your promise.

I have run tests with different nuances on the messaging and the use of icons that have very different results. Iterative testing is useful as you have a small amount of real estate to communicate, but a lot of opportunity for reinforcement of your message, so small differences can have a big impact.

2. Click and collect

If you offer click and collect as well as delivery this is an interesting opportunity.

If you compare South Africa to e.g. the UK, there are big differences in the customer’s ecommerce maturity.

In South Africa, there’s more of an education component around what click and collect is, but it is something that people are interested in. Customers can secure the product they specifically want from a potentially broader range from an online shop and manage the delivery process when they want it. With click and collect, they can be guaranteed of getting the exact product they want, and if they don’t want it being able to exchange it is a practical option.

In South Africa, it can be difficult in non-lockdown/work from home times for people to accept deliveries. Addresses can be challenging to find, and people may not be at home to get packages.

You need to understand your consumers’ feelings about delivery methods, costs, and click and collect. Understanding why they might want a specific option will allow you to message them and reflect their feelings about it.

From a business perspective, click and collect is excellent because you don’t have to pay for a delivery, and this lack of cost can be passed on to the consumer or to your bottom line.

You should do a lot of testing about messaging, education, and which delivery options you make the most prominent.

3. Site delivery information

Site delivery information is a testing opportunity. I’ve had a lot of assistance from savvy copywriters in doing this.

Basic questions are:

  • Who are you writing it for? (i.e. their level of sophistication, what are their questions about delivery? what are their concerns?)
  • How do you simplify it? 
  • Can you make it easily scannable?
  • Can you put it into tables?
  • What is crucial to read first? 
  • How do you talk about returns? 

You need to test these things to optimise what you’re doing. It’s only people who are really committed to purchasing that read the sitewide delivery copy, and they’re reading it because they haven’t found the information they’re looking for elsewhere. If you can give these people the information they need to help them convert you can sway people who are on the fence.

4. Delivery information on the product page

There’s no need to go into too much detail about delivery on the product page. This needs to be only enough to allow people to quickly understand what is available rather than going into deep detail about everything. Too much information that isn’t prioritised to help with customer’s questions is worse, because they can’t quickly scan to find out what they need to know.

You’ll need to determine how prominent this information should this be, should the delivery tab be open by default for people to read and what the layout should be in order to provide information that visitors require?

These are all considerations and tests.

There are a number of pieces of delivery information that it can be good to summarise and surface on the page:

Making it immediately obvious that you have free delivery has been shown to increase conversion (at the risk of making this like a Christmas tree I have also showcased free click and collect in this position and achieved uplifts)
Giving a delivery date expectation is great because it tells me I can get it in my paws immediately.

5. Free delivery threshold

Research I’ve conducted on numerous websites points to customers always wanting a lower free delivery threshold. It stands to reason; nobody wants to pay for shipping.

Any changes you may make in decreasing the threshold come at a cost to you because visitors will have a smaller basket size, and you will earn less, but still have to pay for shipping.

It’s pretty hard to test the free delivery threshold on a website because you have to change your cart threshold details and how that gets managed by your payment system (in addition, it is a legally shady area as you’re not supposed to offer different prices to different people). I think you can get around it by testing messaging further up the funnel that your free delivery is less (e.g. on your product page or in your value strip) and then giving everyone the same delivery price when they checkout.

In my house, there are always discussions about what else you might need to buy from a certain website to help us achieve the free delivery threshold. Do you have something in a wishlist that you didn’t yet purchase because the price was too low? This is how consumers think; what else can I buy in order to get this free delivery threshold, because no one wants to pay for delivery.

The testing that I did on free delivery thresholds was interesting because we tested decreasing the threshold from £50 for free delivery to £40 and then down to £30. The assumption would be that you would sell a lot more, your average order value would decrease, and your shipping costs would go way up.

The results were strange:

  • The conversion rate went slightly up.
  • The average order value went slightly down with each dip in the threshold (but not hugely).
  • Shipping costs went up.
  • Profitability went down.

This case study wouldn’t apply to everyone, but it was good to validate it and there is more research and testing that should be done.

6. Returns

As part of the delivery process, the other thing you need to think about is returns.

As a customer you don’t want to think about returns when buying something, especially if you’re looking at something like clothing. It’s tough to know how something’s going to fit when looking at it online as everyone’s size and shape is different. The thought of having to go through the pain of ordering something, having it delivered and then having to return it is enough to put some people off actually ordering something.

In your on site messaging you need to strike the right balance between being too obvious and explaining how your returns policy will work. You also don’t want to make your returns messaging too upfront as this can promote doubt in a customer’s mind. As a result, it will be a painful process, so maybe it’s easier not to buy anything. (as a case study I actually increased revenue for a retailer by decreasing the size of their guarantee).

The ideal scenario for a customer is that returns will be picked up from them, they’ll be quickly refunded and there won’t be any drama involved. Your job is to communicate something that shows that.

7. Delivery messaging in the cart

Delivery messaging in the cart is crucial. The classic thing to do is have a strip that lets you know when you have achieved free shipping on your order. Messaging can be communicated and stretch back to the mini-cart and through to the payment page.

Everyone (not just cheapskates like me) love to know when they have achieved free delivery, and if you reinforce it throughout the buying journey, you can aid motivation to complete the purchase. 

There are a few pointers for testing around delivery; make it obvious, manage expectations, communicate your benefits, understand and manage people’s concerns and showcase the information clearly and succinctly throughout the customer journey.

Anatomy of a scam

A (smart) friend of mine sent me this Whatsapp the other day.

whatsapp scam

It looked like a scam to me, but I was interested in understanding a couple of things:

  • how smart people get taken in (i.e. what marketing/persuasion tactics were used)
  • what benefit the person who set it up gets

As I went through it I had discussions with my wife who managed to decode the tech stuff behind and (spoiler alert!) find the culprit.

Marketing persuasion tactics

The initial message is good:

  • they use the official logo
  • the domain looks legitimate (but actually it’s a subdomain of
  • it’s short and punchy
  • it contains the magic word “FREE” (twice)
  • Urgency (as Cialdini says) is a great method to encourage people. (though I think they’ve tripped up here: “Hurry up! Collect your FREE voucher” is not the words of a respected retailer giving away R5000)

Unfortunately the site is now down, so you’ll have to look at these desktop screenshots to puzzle through the rest of the process.


Screenshot 2020-04-18 at 10.46.59

The page you click through to is on the left and the Woolworths homepage on the right. If you compare them you can see quite distinct branding differences (i.e. the left hand version doesn’t have great branding: small logo, lots of colours, “cheap” design). The image of the shop on the left hand version is also a bit strange, it looks like someone’s taken a picture and it doesn’t look like they’d put a graphic designer on a promo that would be likely to cost them a fortune.

Step 1:

There is a countdown from 85 remaining vouchers on the page (again Cialdini’s principle of scarcity). It almost seems believable that Woolworths would be giving away big vouchers (especially in the first days after lockdown). The precision of the 2 day expectation of when you’ll receive the vouchers is a nice touch, it sets expectation and makes the offer less nebulous. Having to jump through some hoops to qualify also adds trust. It show it’s not some lame freebie, one actually has to work for it.

Step 3:

Step 2 was an age verification, another trust marker to prove your eligibility. Step 3 gives you more hope if you’ve never won a voucher previously, that they’re using an algorithm to ensure it must be your turn this time.


More Cialdini magic here, some social proof to cement the fact that you’re definitely going to win as evidenced by these real people (or three people from the Smith family). You can’t click on any of the profile links to check that they’re real, but having them there as proof is enough for most people.

Unfortunately this is where the wheels fall off a bit. The tool they use is a template that they’ve used for other offers and they didn’t bother to change this offer for a free pair of shoes to the Woolworths voucher.
Technical stuff

If you view source (and are cleverer than me) you can find out there is a script that does most of the hard work:

There are a couple of sites referenced and if you do some digging through domain registrars you’ll find a clear link to this guy in Delhi:

Screenshot 2020-04-18 at 11.43.24
It seems that he isn’t a developer and is using a script that you can buy for perpetrating scams like this (Google voucherf0c9.js if you’re in the market for some low level scamming).

He has Google Tag manager set up and a number of services to track events, generate random social profile pictures and an AI tracker

It seems that it is set up specifically for mobile and has a load of affiliate links/cookies on it (+/- 134 cookies!!)

You can see an example of another promo/scam that they had on their site below.


The big question, why do this?

People want to get a great/FREE deal so they’re motivated to fill it in, but what about the person that is perpetrating it.

My assumption is that they dump a whole load of affiliate cookies on your device, you share it with all of your friends so there is a large viral spread of this. Eventually someone will purchase somewhere and there is an opportunity to make some money off you.

Simple and effective if it spreads far.


I had the pleasure of receiving a new version yesterday, as you can see exactly the same template:

The power of nudges

I recently upgraded my Balsamiq subscription and I thought this was quite a good example of what I mean by a nudge:

  • a consistent always on reminder of the payment option and the benefit that you’ll receive.

Screenshot 2020-03-05 at 11.42.52

I think it works best with industries like SAAS where you are on the site fairly frequently, but paying isn’t always obvious, and as the site owner you want to keep it top of mind,

I’ve written previously about when we started testing nudges on Now Novel, and since then we’ve broadened out the complement so that they drive over half of our sales.

We’ve tested a number of different areas where it is possible to reinforce benefits at different junctures :

  • modals that cover the screen
  • text links near your profile/sign-in
  • inline boxes and
  • top of the page persistent strips

I wanted to share some of the results that we had with a test that we were running. This was at the end of the freemium section where you have created your character. You have seen the value of the product at this stage, have some momentum and increased motivation,  therefore it’s a good point to encourage purchase.

This is the control/default option we had been showing:

Screenshot 2020-03-21 at 16.31.06

and we tested two variants, one with a focus on a testimonial:

Screenshot 2020-03-21 at 16.30.16

and the other with a focus on the benefits that people said they received from the service (so reflecting benefits back at them that we learned from surveys)

Screenshot 2020-03-21 at 16.31.45

We measured the impact of this by visits to the payment page (a useful proxy metric for progress to sales) and sales conversion. We had over 90% significance on both variants for sales conversion, but a higher uplift on the testimonial version. This is the one that we’re rolling out as our new winner.

Some next testing opportunities:

  • make the review more obvious (star ratings, who powers reviews to add more 3rd party accreditation, “What our customers say” to label it effectively)
  • combine some more of the benefit messaging into the testimonial version
  • can we make it shorter, easier to scan and punchier?
  • Is “Upgrade” a good call-to-action or could we use something like “Get access” (upgrade sounds like you’re spending a lot of money)

Headspace’s cancellation funnel- teardown

My Headspace subscription was coming to an end and I wanted to cancel . As I went through their cancellation funnel I wanted to look for dark patterns and see if there was anything I could learn from what they had done.

Overall I thought it was OK, but I think trying to rescue a cancellation at this stage is hard work. Even if you do, without finding a method to fix why people were cancelling in the first place (through some sort of re-engagment/re-on-boarding) would be quite difficult.

Step 1:

Screen Shot 2019-05-20 at 12.05.32

I thought this was pretty good, a good graphical representation of the features that they offer (having never used Headspace for kids I’m not sure why it got such a high placement, but maybe that is a key feature for most people). I am surprised that these are all feature driven, and not benefit focussed  (I’d expect them to be focussing on improved concentration, reduced stress, better sleep etc). The green “Never mind” is a strange call to action, I’m not sure exactly what it should do/why that is the exit option.


Step 2

Screen Shot 2019-05-20 at 12.05.45

This is an interesting screen. “Changing your mind” is a good turn of phrase, especially in the context of what meditation can offer you. The first two (customer support and payment plan changes) are quite mundane.

Helping you find time to meditate is a great option. We find that it’s a key challenge for people wanting to write a book using Now Novel too. Self directed long projects are challenging for people to commit to. They don’t have an idea how long they’ll have to do it for (which is scary) and they’re unsure if they have the gumption to stick with it. In addition watching Netflix or scrolling Instagram is a lot easier after a day working. I would expect that this strategy would be better employed when onboarding rather than at this juncture. People have low motivation when they’re cancelling, so asking them to try and start a time saving process may be difficult.

I was thinking if I was looking at this first screen through a different lens how I might test a change. I have no user feedback so it would be entirely based on my perspective (so it probably would be 100% wrong :)).


I have put together a very rough wireframe.

Here are my thoughts:

  • starting meditation can be hard, so reflecting that back to them and explaining that they’re not alone with the challenge can make them more receptive.
  • Focussing on the benefits, not the features
  • The marginal cost of one extra month for someone is so negligible in terms of cost, that you might as well try offering it
  • Getting people to start onboarding (or speak to customer service) may resurrect them
  • Putting in as much social proof as possible (60 million users, high ratings) may convince people that its them and not the service
  • you could be sneaky and put the cancellation in a text link, hide it elsewhere or make the button text something like “I’ll quit please”, but I don’t think that kind of dark pattern UX helps you in the long run.

Using data science to get insights from reviews

Review Analysis Wex Canon (1)

Machine learning and AI are frequently used buzzwords. In terms of application I’m not sure whether we’re in the killer app territory yet. I read someone on Twitter saying something to the effect of most AI can be replicated with some good SQL queries (which is a nice soundbite).

From a conversion perspective I use Eyequant quite a lot. It mimics heatmaps of user interaction from a design (either on live websites or on heatmaps/wireframes) so you can get an idea of how visitors will perceive a website and where they will focus on the design.

From their website:

“EyeQuant fuses leading neuroscience research with AI to accurately predict how people will react to digital designs.”

I have been working with a data scientist on a couple of research projects to try and parse reviews to get an idea of underlying perspectives. “Text mining” of reviews is not something unique, this article goes into some detail about how to use R to extract the data, but more importantly how to display it so that quantitative insights can be gleaned from the voice of consumer.

Here are two examples of how I’ve used it:

  1. A company was selling the same products that Amazon stocked and my thoughts were to understand both the reviews and the questions asked by customers (as Amazon has such a volume of feedback it made this process easier to look at Amazon rather than on their site). One of the key products looked at was a bundle. If I use the example of a keyboard, computer and mouse I was surprised to see that most reviews focussed on the mouse, rather than on the core product. This lead to testing exposing more details about the mouse as we knew it was such a focus for visitors.
  2. The other was also to look at reviews for a high-tech product, but to understand what features people were most interested in. Once I found out what the focus was this helped to prioritise what to surface more and how to modify the architecture of the page used to promote this.

These are some pretty straightforward applications, but its interesting to understand that there is a wealth of data available, and if we can learn to unlock it efficiently we can be more efficient.


How jobs to be done can aid your first user experience


I’ve drunk the jobs-to-be-done Koolade and there is no coming back. Jobs-to-be-done is a framework which allows for uncovering customer requirements in more depth, so that you can ensure what you build maps to what customers want. through the process you’ll conduct open-ended interview to understand:

  • what solutions customers are currently using
  • how they feel about them
  • what their ideal solution would do (even if it doesn’t currently exist)
  • what job they would employ your product to do

Your role is to then take all the responses, annotate and create high level requirements from this. We’re looking to understand what job the customer is looking to “hire” your product to do. Its not asking your visitors for product specs or to design the product for you, they don’t know ( the apocryphal quote from Henry Ford fits here: “If I had asked people what they wanted, they would have said faster horses.”)

However ensuring that you’re solving the right problem and building solutions to help with customer needs means that you don’t waste resources, saves time and results in happier customers. I’ve used this framework across a number of industries and it is remarkably robust.

I believe jobs to be done are more effective than personas. Most personas that I encounter seem cliched. A JTBD allows you to stack numerous jobs on a single person irrespective of their demographic or psychographic background. Others would say that a well sketched out persona should already include the JTBD and the additional detail in a personas allows you to have more empathy for the subject. I suppose it’s a matter of personal preference.

A colleague recommended some helpful books recently about Jobs to be Done:

  • Cindy Alvarez’s Lean Customer development
  • Alex Osterwalder’s Value proposition canvas (for the second time)

both of these are not just theoretical models about how to get insights, but offer actionable step wise processes to gain insights.

A new project we undertook was to use JTBD to aid onboarding and create a better first user experience. We have one onboarding sequence for everyone, irrespective of their requirements. As you can imagine this isn’t perfect. Once we outlined the key jobs that were required we went further with a survey to get even more insights. We asked a number of questions, but a couple were useful for this area:

  • finding additional challenges
  • finding satisfaction points

We could then segment this based on paying vs non-paying users as well as the core job they were interested in. Once this was analysed we could find insights that would match the specific job with the best part of the product as well as the additional challenges they had with parts of the product that help with these requirements. This would aid them and give them a more tailored first user experience.

More testing and development needs to happen in this area, I will report back on results.

Feature request: Intercom analytics

webaroo-com-au-shr_Xn8S8QU-unsplashWe’ve been using Intercom for Now Novel since 2012 and have watched it turn into the behemoth it currently is. It is a great tool that helps us communicate personally at scale.

Intercom doesn’t leave a lot of extra customer value. They know the value of their service and what people’s propensity to pay is and are not afraid to map those two quite close to each other (i.e. its not cheap :)). This could be an additional service they offer that could add more to their offering.

If I look at the tools that we use to get insights from our customers (mostly Hotjar and Google Analytics) we don’t have a tool that focusses on users. Google Analytics is a great tool, but it’s a little hamstrung by being session rather than user based. Sessions are interesting, but on a user level you get a lot more of the user journey and see where they interacted with what portions of the site on an individual level.

We’ve already set up Intercom already with a lot of information that would be great to use to understand our users further. Our Intercom integration has specific questions that are set up at key points in the customer journey to message to customers. This means that you’ve already got the tagging part of the equation sorted out. Intercom prompts are usually set at key parts of the customer journey too (e.g. end of onboarding, start or at payment junctures) Being able to interrogate these would be very helpful. We have already tried to use this before to look at the data, but it isn’t possible to look at it in an easily accessible manner. You have to go into each user and manually look at their progress, you can’t e.g. export a csv with all of a user’s activities.

I’d ideally like to be able to look at a tool that would allow me to:

  • look at users over a specified time period
  • who have achieved a specific aim or goal (which would be one of my triggers) and
  • examine the data around this criteria

The data around this criteria could be other goals, visits, length of time, communication acted upon etc. This starts to be even more interesting when you can import and export this information into other tools (e.g importing my Google Optimize test variation into Intercom for segmentation on how a specific test performs or a specific segment I have created in Intercom that I can start to view more information about in GA).

Intercom must have thought this, so when they need an early beta-tester, count me in.

Books read 2020

  1. Lean customer development- Cindy Alvarez

My first completed book of the year was very useful. I’ve read a fair amount in this area (the book references the Lean Startup and Steve Blank), but this was a useful addition to the canon. The best thing about this book was the  practical /tactical method it was written in. It taught you the theory, but then it shows you exact methods on how to implement your customer learning process and how to quantify and use your learnings. There is immediate practical value I am taking from it to use in my process.


2. The big nowhere- James Ellroy

I have read (and watched) LA Confidential and am a big Raymon Carver fan, so I was prepared for the hard-boiled cop fiction. It is quite unrelenting however; brutal murders, wolverines, communists, crime bosses. It doesn’t let up for 400 odd pages and goes through numerous twists and turns. The heroes aren’t really apparent and the main characters all have their own struggles. However it is definitely worth a read, the perspective on marijuana and homosexuality in the 40’s s especially interesting knowing how in modern day California they aren’t even an issue.


3. Grace Jones- I’ll never write my memoirs


Thought this isn’t the cover, its hard to talk about Grace Jones without talking about her image (and my favourite of her images are the ones she created with Jean-Paul Goude, he talks about this one here).

I am a big Grace Jones fan; I listen to her music, I watched her documentary so this was interesting to hear straight from the horse’s mouth. Some of it was a little repetitive; pushing against the vagaries of the established music system for 40 years must be challenging, but can also be much of the same. There were flashes of her personality that came out that sound amazing and tiring (if you had to behold them). A perspective of a unique artist.