How Long Does It Take To Create An Account On WhatsApp?

It takes 2 minutes and 5 seconds to create an account on WhatsApp.

Here's a slidedeck on the site useronboard.com which shows step by step the user journey.

Useronboard have a series of "teardowns" that show you step by step the onboarding process for new customers for a whole raft of well known sites from Vimeo to Netflix to Trello.

What I like about these "teardowns" is that they not only show the whole journey, screenshot by screenshot, but they also have running commentary about what is good or confusing.

If you're a product manager thinking about how to improve your onboarding experience, going through these teardowns will give you plenty of ideas of what you could test or improve.

Thanks to James Brooke for highlighting this excellent resource on user onboarding.

The Experience of Others

You're sitting in a restaurant having a great time, in fact it's a wonderful evening.  You have a good round table with a bunch of friends, the food is fantastic, the service is excellent, the wine is flowing.  The next day you get an email from the restaurant owner asking to rate your visit and you give it 5 stars.  It was fantastic.

Mrs Restaurant Owner is very happy because her customer was deeply satisfied with the experience.  She also runs a good business so she made some money that night.  Overall, it was a success for everyone involved.  Fantastic guest experience, good business, everyone is happy.  Wonderful. Beautiful. Perfect.

Ah. Experience.  That's a troublesome thing you see because it's actually quite difficult to establish what precisely was it about the experience that was good.  Was it the conversation?  Was it the company?  Both of these things?  Or the food? Or the seating configuration?  In fact, it's all of these and much more.  Experience is an emergent phenomenon.  It's emerges from a set of relationships, objects and timing in a specific moment for a specific individual to feel, well, something.

Product managers of a website have the same challenge as a restaurant owner.  They need to create an emergent set of experiences that support the business objectives.  Those emergent experiences could be for example, ease of use, joy, engagement, desire, advocacy (telling others), typing in credit card numbers, or creating a profile.

A Product Manager is an experience engineer.

Creating something which is emergent is not easy.  It's bloody difficult.  If everybody could do what Apple have done with their products in the last 10 years they would have done, but they didn't.  If something is difficult to do, it's difficult for others to copy.  And creating emergent systems that create wonderful experiences is extremely difficult. It's part art and part science.

In this world of emergence first you need to understand the systems from which your experience will come from.  The restaurant owner looks at all the details, from hiring of chefs, their kitchen organisation, the staffing rotas, the buying of ingredients, the menu design and hundreds of other small details and tries to create repeatable patterns of success that result in emergent experiences that restaurant goers rave about.  It's difficult in a restaurant and it's difficult on a website.

This web of interrelated influences that creates the emergent behaviour is a 4D world.  It's like the web itself, connected nodes of influence that act like the synapses in our brain, triggers firing behaviour from one sphere to the next. The fourth dimension is of course time, because each customer interacts with your company several times in several ways in different states of mind at different locations.

A creative product manager will first try to understand the web of influences, understand the triggers and amplify the ones that work, as well as trying new things that could be new influences and triggers.  All the time, measuring the impact as best she can.

Here's the thing though, experience can never be fully measured.  It's indescribable, intangible.

I can try and explain to you how I felt thinking this through on my walk up Shaftesbury Avenue this morning, Run DMC pumping through my headphones, excited because I'd pulled together some thinking that I'd been dwelling on for weeks, rucksack on my back, thinking of my next ski touring trip.  I could try.  No-one can ever really feel that except me, at that time, in that place with my history, future and present before me.

You can measure many aspects of experience, many dimensions.  Snapshots. Like the restaurant rating from the delighted customer.  You cannot really feel the experience of others, but you need to do your damn best to try.

As a product manager you are dealing with millions of those experiences.  And furthermore, you are building an interface that your customer will experience using a team of other people.

Building great experiences for others, through others.  A tough, but rewarding challenge.

How not to ask for a phone number

A real life lesson in simple usability improvements that can improve conversion rates.

My wife and I love the movies, so to celebrate Valentine's Day this weekend we'll be going to our local cinema to see "The Kings Speech".  I've heard it's very good.

I booked online with at the cineworld.co.uk site (@cineworld) and only just managed to book thanks to a rather horrible usability hurdle.  Here follows what I wrote to them using the contact pages after the experience.

Dear Cineworld

"...I wanted to share with you a usability problem I had with the payment pages.  If my experience is anything to go by it will be seriously hampering your conversion rates and if you fix it should improve your bottom line immediately...


It's to do with the validation of the mobile phone number.


I initially put in my number as follows;
077XXXXXXXX


The error message told me to add the country code. So I did;
+4477XXXXXXXX


The error message told me to only use digits.  So I did;
004477XXXXXXXX


Then the was another error message.  Can't remember exactly what it said.  To get through the validation process I needed to actually type;
4477XXXXXXXX


4 attempts!
From someone who works building websites for a living and has done so for the last 11 years!


It's quite possibly the worst validation experience I've ever  had.


Suggestions;
1. If you want to capture the data in that format tell them so in the form up front 
2. Alternatively have a drop down of country codes to select from with UK as default.  Then ask people to enter their number as normal in the field and get your programmers to remove the first zero and any spaces.


In essence - you're making it way too complicated.


Hope you find this useful..."

What's happened here?

I've seen this happen many times.

The developer says, "In order to send SMS confirmation messages, my system needs clean and correct numbers.  I need the country code, then the mobile number in the following format; [XXYYYYYYYYYY] (where X is the country code, no zeroes, no symbols and Y is the number, no spaces, no zero up front)".

The developer does not say, "How do people usually type in their mobile numbers?  How can I help them enter it in a way that is obvious for them and still helps me get the data in the format I need"

Don't blame the developer though.  The producer / product manager should be on top of this during the design process and the interaction designer should instruct the developer.

Cineworld, to be fair have built a pretty intuitive website and it's very easy to find and select a movie to book.

Funnily enough, Cineworld did populate the form with a suggestion format in the field.  I didn't see it, the light grey was so light it didn't register.  Then, when I got it wrong, I didn't have obvious instructions on what the format should have been.

The only way top really test this is to watch real people using it.  People with nothing to do with the design process. Learn from idiots like me, tweak the form, then measure the results.  How many more visitors converted into bookings did I make as a result of that change? Sounds obvious.

Be your own customer.

Don't Be Mediocre, Be A Generalist

Here's a problem. Most e-commerce websites suck. They're average, sub-optimal.

They suck. If they were high street shops you'd see what I mean. Some would not be signposted. My mother would not be able to find anything she wanted in the store, let alone figure out how to pay. They might not even be found if you called directory enquiries. Lighting would be poor, there would be no space to park and you'd be unsure if you were being served by rip-off merchants or a legitimate business.

Sure, there are some great websites out there. A few. However, because web customers are in many ways "invisible" to the businesses that are trying to serve them, site managers running these sites get away with mediocrity.

Small businesses or large businesses, the problem is the same.

In small businesses you need an all rounder running your web business: he or she needs to be able to do a bit of everything and do it really well to be effective. These all-rounders are few and far between. So - good websites for small businesses are few and far between.

In a large web business, you end up with different "specialists" taking on different roles. You get the design team, the merchandising team, the online marketing team, the tech team, the product planning team, the analytics team, etc. They are all brilliant at what they do. Brilliant.

Just why does the site that you end up with suck if they are brilliant? The really great websites are the ones where the teams do "joined up thinking". The site design team know something about SEO. The analytics team knows something about PPC. The product planning team understand affiliate marketing. If they all understand the basic principles behind all of the main disciplines, you'll do fine. If not you can end up with mediocrity. Division of labour brings pockets of excellence and general mediocrity.

The solution to me is clear: every person involved in delivering the website and running it should be familiar with the basics of how all the main disciplines work.

I will give somple examples of what are these "key disciplines" that are needed to overcome mediocrity and suggest 5 tips on how to acquire those skills. I don't for a moment suggest that anyone can be an "expert" in all of these areas. What I am suggesting is that everyone in an ecommerce business that can affect the business results needs to be a "generalist" and be fluent in the basic principles.

This is true whether you are a one-person site manager doing everything or whether you are a specialist in a big company.

Consider ecommerce websites to be like high street shops. If you are running a traditional "bricks and mortar" business, you need to do three things...

1. Attract customers
2. Persuade them to buy when they come through the door. (And ideally, maximise this so that they
buy more than they would normally have done).
3. Encourage them to come back to shop again and become a regular "repeat customer".

Quite simply, running a website / web service is all about customer acquisition, conversion and retention.

Acquisition

Retail is all about selling the right product to the right person at the right time. So to get this right, there's some preparation needed in terms of customer insight / research, developing a brand, building a proposition, product planning & strategy, business case modelling. Then once you have a product you need a communications plan, (including, but not excluding online PR). On a technical side, you need to understand search engine optimisation (3 core pillars; site structure, content and reputation), search engine marketing (paid search), affiliate marketing, social media optimisation, RSS, blogs, ad serving.

Conversion

Here, the most important discipline is user-centred design, also information architecture, followed closely by merchandising techniques and understanding effective calls to action. Think about good error handling, optimising shopping "funnels", accessibility, compelling copy writing, tactical promotions and you might just have scratched the surface.

Retention

Plenty to learn about here too, from email marketing to direct mail (CRM), use of confirmation emails, special offers, customer database management and that's just scratching the surface.

And more...

Underpinning all of the above is a good understanding of web analytics (for traffic sources, keyword analysis, conversion tracking - and a whole lot more).

And to get stuff done, to make it all happen they also need to understand a little about technical things like technical architecture, software development methodologies, project management techniques, be a great tester.

To add some real flavour to the mix, key principles of many academic disciplines come in useful, economics, sociology and psychology.

And to be effective in business generally, you need to know about HR issues for the countries in which you operate, be a great manager, have good negotiation skills, know a little bit about contract law.

If you find someone who does all of the above well and them let them loose on your website - now you're talking! Also, point them my way, I'd love to meet them!


So - if you're just starting out...

5 tips to become a great E-commerce "generalist"

1. Work with great people

We learn so much from working with other people. Put yourself with the best you can find and learn from them. Soak it up.

2. Apply yourself

This industry is only about 10 years old. Most of the so called experts became experts by just applying themselves and learning from their mistakes. You can too.

3. Attend training

If you can, go on specialist training courses. There are few places such as e-consultancy.com where you can get a kick start

4. Read, read and read

You can never read enough. Some of my favourite books are listed here if you need ideas.

5. Have a go

For me this is the most important advice of all. Have a go! I didn't know anything about setting up RSS feeds until I had to do one for this blog. Figure it out, and just try.


One day, most websites will be great. Until then, if you are a "generalist", go make some money!

"The General"

I Must Be Getting Old

According to recent research by Jakob Neilsen, older web users take longer to complete web tasks. In fact, he's actually calculated a decline of 0.8% per year for every year you get older after the age of 25.

He says "The human aging process starts around age 25 and causes erosion of cognitive resources, loss of visual acuity, degraded reaction times, and reduced dexterity. People need more time for the same mental operations; they have less memory capacity and take longer to process the same perceptual input."

That makes me feel really great as I approach my 39th birthday. Ouch.

So, am I 8% slower than I was 10 years ago? Wow, I wonder if that's true?

Well, I just know I'm faster overall. Maybe my brain is slower, but I am sure that 10 years of internet usage have brought massive gains from experience. Slow brain maybe, but a very efficient one!

See Middle-Aged Users' Declining Web Performance for the full article.

Experience Is All That Matters

It's a great saying: "Be Your Own Customer".

It's only when you actually experience being a customer that you really know what it is that you are selling. OK - you might be able to hold the product you are selling. You might be able to go into the shop and see the shop. It's not until you actually try and find the shop for the first time, find what you are looking for, buy it and take it home to use it that you truly experience what you are offering your customer.

The key point is this: you never sell a product - you always sell an experience.

Let's take an example. The iPod. The iPod has been phenomenally successful. You hear people extolling the virtues of the "user interface", the "design", the "packaging", the "marketing".

Really though, it's the sum of all of these parts (and more) that is the iPod experience. The marketing campaigns convey a simplicity and "cool factor". If you go to buy one from an apple store (online or in the high street), you are wowed by state of the art presentation. You may get good service in the shop, or - if online - you get site that is well designed and easy to buy from. The packaging is so slick, neat, clean. You unwrap your iPod and just "feels" great. You just plug it into your computer, it charges fast. You download iTunes - nice and easy. In fact, iTunes is just as much a part of the iPod experience as the player itself. It's all so easy. Then of course, there's the user interface - the simple controls. On top of that it just "looks" cool.

You see - the iPod is not successful for any one of these individual factors. The total "experience" is what counts. I'm not saying that everyone has a great experience every single time - I know of people who have a had batteries that fail after just a year. Some of these are genuinely annoyed and will never use Apple again (Apple, take note - "must improve here"). Equally though, there are fans that will just dump the old player whose batteries failed and just buy a new one. Their experience to date was SO good, that the simple fact that the product was substandard did not put them off buying another one. Kudos to Apple.

For any business therefore, every point of detail matters. The message, the communication, the product features, the sales process, the purchase experience, the convenience, delivery / pick up experience, customer service, product reliability. None of these details ever stand alone.

Unless you are a company of just one person, you always have different staff members dealing with different parts of the experience. As a result, none of them get to see the overall picture from a customer's perspective.

The only way to see what you are really delivering, is to be your own customer.

Go buy your own product. See what it feels like. Is it good enough? Is it better than your competitors? Go test their products too. Now you're getting somewhere.

Bigger fonts come of age

Increasingly, I'm seeing more sites designed with bigger font. Nowadays, 10pt font is considered "small", and we're seeing more and more 12pt fonts as standard. Even if paragraph text is still relatively small, headlines and titles are getting bigger.

I love it.

Bigger screen sizes now mean that we have the space. What used to be readable on the old 800 x 600 screens of yesteryear now requires a magnifying glass if you put it on a 1280 x 1024. Only a tiny percentage of users now have the small screens.

Silver surfers are driving the demand for simplicity. When you get older, your eyesight deteriorates. Larger font sizes it make it easier to read what's going on. Usability is now a common place discipline in e-businesses (although not common enough in my opinion), and time after time, usability studies show that small fonts can create unnecessary hurdles.

It's not just older folks that benefit. The rest of us can see what's going on from a distance, in poor light, on the move.

How many sites have you got frustrated with because the font was too big? I can't think of any. What about sites where the font was too small? Now, I could think of a few there.

To illustrate why big works, here's some examples of sites that use larger fonts...

The new rightmove.co.uk property search site
BBC news
Amazon
37 signals
Jakob Neilsen's alertbox

Note that you don't need to use large fonts throughout. There is still a place for smaller fonts, although they shouldn't be standard.

Set your browser to "larger" font and start looking at sites with smaller fonts. You soon realise how much easier many of them would be to use if they increased their font size.

Checklist For Reviewing Webpages

I'm often asked "what do you think of this web page"?

I immediately have a surge of mental activity that leads to a whole load of things I'd change. Increasingly though, the same old things keep coming up. So here's my basic checklist.

What is the user journey?
This is the most important question to ask first of all - what is it we are trying to get people to do here? Remove yourself from the page, the design and the clutter and write it down in simple steps. Go back to the page now and review the design against those steps.

Using font that people can read?
At least 10 point. Maybe 12 point if you can. Small sucks. Make it easy!

Is the contrast OK?
Text should always be either light text on a dark background or (preferably) dark text on a light background

Are text links underlined?
Ideally they should be. Exceptions are place where it's obvious that this is navigation (tabs or menus for example).

DON'T USE CAPS, Use sentence case
It's easier to read

Use only the words you need
On the web, less is more. Look at every detail of the page - forms, buttons etc - and only use the words that are needed

Is it obvious what this site is about?
If the site is well known, this is less important (e.g. Dell, Amazon, Google), everyone else should explain what the site is about. Either in simple terms (e.g. Bargainholidays.com - "more holiday than you bargained for") or in a sentence near the top of the page.

Is it obvious what this page is about?
Page title, big, saying what the page is about.

Is it obvious what I should / could do here?
The call to action, whether it be a price point, form button or text link, should be obvious. A few obvious calls to action are more effective than many obscure ones.

Is is clear where I am on the site...?
(...relative to the rest of the site). Which section am I in? How deep am I? How do I get to other sections/products like this? How do I get to the homepage?

Does the page use standard conventions?
Form buttons that look like buttons, text links that look like links (in blue), click able images having text appear to describe the link destination (alt tags).

Is the page design consistent with the rest of the site?
Thinking here about layouts, colours, imagery and copy

Do links describe what you're going to get on clicking?
("Click here" is absolutely banned in my world)

Do forms state what's optional versus required?
Make it easy using asterisks or other well recognised devices

Does error handling make sense?
Try and break the page - see what error messages you get

Is the URL meaningful, canonical and unique?
It should describe the content, avoid parameters & dynamic URLS

Does the page load fast?
Simple to say, simple to test

Does the HTML validate against W3.org standards?
Build it right first time

Does it work in all browsers, in all of the main sizes?
IE 6 and 7, Firefox 1.5 and 2 as minimum, ideally Safari too, on Windows and Apple Mac.

Can we track the activity that we want to improve?
Install web analytics if need be, and use it.

The User-Centred World

I predict a move from user-centred design to a user-centred web, and eventually to a user-centred world.

Imagine a world where all of your relationships, communication, interests and interactions are wrapped into a single digital communication portal. This portal would save all of your contact details, conversations & messages, whether these be phone calls, instant messages, emails photos, videos, music & other media.

Importantly, this portal would transcend any one device and any one software or hardware provider. It would not be just on your PC. It would not just be on your mobile phone. It would not be just through your TV. It would not just be through Google, Microsoft, Yahoo! or Facebook. This portal would sync with any device that you use, at any time, for any reason.

Imagine also that this portal could interact with you in your real world as well as your virtual world. It would track your position through GPS on your mobile phone (say) and give you directions to reach your friends in the vicinity. Maybe you would exchange contact details with new contacts just by bluetooth, and this would automatically sync with your online services such as your email, blog or birthday reminders list.

Why would we want this?

Why not? The digital era is getting ever so complicated. I've lost count of how many logins I have for different sites and services. Even keeping my contact lists up to date gets a little crazy... Mobile phone (personal, work), Outlook (personal, work), social networking site A, social networking site b, webmail a, webmail b, webmail c, address book at home.

Why should I have to keep of these devices in sync? As a human being my life is being carved up arbitrarily by the fact that different companies supply me with different devices, software and services. Do I care? My friends are my friends. My family is my family, and my work colleagues are my work colleagues. Real life is not divided into mobile phone, email, IM and social networking providers.

I want to interact with the real world as easily as possible. I want to have everything in one place. That one place is ME!

YOU are the centre of your universe and I am the centre of mine. (OK, some wouldn't agree, so don't get all philosophical on me now - let's keep this simple shall we!?)

I want to interact with people I like, need or can help. I don't want to interact with the rest of the world - life is too short!

I want to interact with companies that offer me things that I am interested in. I don't want to interact with companies that are polluting my life and time with things that I don't care about.

(Don't try and sell me a bed shorter than 2 metres in length (I won't fit in it). Don't try and sell me a steak (I'm vegetarian). Don't try and sell me shoes unless you stock size 12 - you're wasting my time. However, if you offer me a DVD subscription service better and cheaper than the one I have now, I might just be interested).

Anyway - if I could interact with all these people and companies through one profile, in any place, on any device, in a way that suits me, well that would be the user-centred world.

User-centred design can deliver commercial success. I predict though that the user-centred world will be absolutely more commercially successful than our current world. It's the ultimate "long-tail" in action.

This is already happening and I believe it will accelerate over the next few years. What example trends are there in this direction?

Users requesting information (pull, not push)

Google's iGoogle (personalised homepages)

RSS feeds

Customisation of desktops in Windows Vista

Tivo TV / Sky Plus

Podcasts

Contact across multiple devices

IM on mobile phones and PC

Email on a Blackberry

Mapping relationships and sharing interests

Social bookmarking (e.g. del.icio.us, Reddit)

Social networking (Facebook, MySpace, Bebo and others)

Business networking tools (Plaxo / Linkedin)

Infrastructure

Open-source software (e.g. Linux, Android)

Open source knowledge bases (e.g. Wikipedia)

Web enabled phones

GPS

Fingerprint based payment systems

Retina scanning security systems

Wi-Fi

10 steps to the user-centred world

1. User interests will be saved against user profiles and will be updated in real time in response to user activity
2. User relationships will be saved, categorised and constantly modified
3. There will be eventually a single user profile for each user
4. User locations will be tracked in the real world as they move about
5. Profiles in the digital world will transition to real world
6. All communication devices will centre on the same user profile
7. Content will be delivered to the user at the user's request
8. Advertising will be requested by the consumer, (pull not push)
9. "Popularity" will become increasingly more important in driving brands
10. Globalisation will allow faster networking of individuals and companies in a single digital reality that is entwined with the real world

Consumer demand for simplicity and convenience will ultimately take us down this path.

Remind me to read this post in 15 years time to see if we've done it.

Next Generation Breadcrumbs - Internal Linking For SEO

Five steps to internal linking heaven...

Breadcrumbs are a useful navigation tool indeed. Nowadays they have been given a "best practice" vote by the customer experience minded folks. Indeed Jakob Neilsen, king of usability says in a post Breadcrumb Navigation Increasingly Useful.

I know I like a page to have breadcrumb navigation. It makes it easy to jump up a level or up a couple of levels without going back to "home".

I often think of pages as being "parents" or "children" in a site. If the home page is your Matriachal Great Grandmother, the category homepage might be the grandfather (her son), the product group page the mother (his daughter) and the product page the child (her son)...

Home > Category > Product Grouping > Product

or...

Great Grandmother > Grandfather > Mother > Child


What have breadcrumbs got to do with SEO?

Breadcrumbs have everything to do with SEO. The search engine spiders will use your internal linking structure to determine which pages you think are most important. If you link more often to a page yourself, you must think it's important and search engines take that logic onboard when mapping out your site. Given that a breadcrumb consists of many links on all pages, it's a good place to start.


Step 1 - The basic state

Imagine the following (hypothetical) site with 4 levels in it's heirachy:

Home (1 page) > Category pages (8 pages) > Product grouping pages (30 pages per category) > Product pages (200 per category grouping).

This gives me a site with 6,249 unique pages.

Assume that the homepage does not have a breadcrumb, but that all the others pages have one do according to the site structure given above.

The total count of links into each page type created by breadcrumbs will be
Homepage: 8 + 240 + 48,000 = 48,248
Product category page: 6,030
Product grouping page: 200
Product page: 0

In this example, the homepage gets a huge number of links. After that, the page that gets most links from breadcrumbs is the product category page. The product page itself is not getting any links.


Step 2 - The medium state

You may not be happy about sending all of your links to your category pages. Perhaps on your website these pages are not the ones that you want to rank highest. Maybe your aim is to get the product grouping pages to rank higher. What can you do?

The next step is to try and add in some peer to peer linking. By this I mean linking from every page to all of it's "brothers and sisters". This could be done using say a side navigation bar. From a a usability point of view it might make sense as well - your users will be able to browse in a horizontal fashion. We do all the time in the real world. When we look at a pair of trousers in a clothes shop, we look at many pairs before trying some on - you wouldn't expect to have to go to the aisle end every time you'd looked at a pair.

The lower down pages in the heirachy have more sisters and brothers than the ones above them in the heirachy.

New links from "brother / sister" linking:
Homepage: 0
Product category page: 7
Product grouping page: 29
Product page: 199

Let's also assume also that your site template has "tabs" in the header that link to all of the category pages, plus the homepage.

New links from primary navigation (tabs)
Homepage: 6,248
Product category page: 6,248
Product grouping page: 6,248
Product page: 6,248

The new totals for internal links now stand as follows...

Homepage: 54,496
Product category page: 12,285
Product grouping page: 6,477
Product page: 6,447

We're getting there, but our category grouping pages need more help.


Step 3 - The advanced state

I'm still need more links to my product grouping pages. These are the ones I want to see top of Google's rankings.

What can I do?

Here's a simple idea. Link to every product grouping page from my site footer. This appears on every page of the site, that's another 6248 links (one from every page on the site) to each product grouping page.

New totals:

Homepage: 54,496
Product category page: 12,285
Product grouping page: 12,726
Product page: 6,447



Step 4 - The enlightened state


Here's the crunch point. Here's where you might want to mess with your nice neat site structure to benefit SEO. It could comprimise your usability, maybe not.

Here's what you do.

Change from:

Home > Category > Grouping > Product

To

Home > Category

And

Home > Grouping > Product

The Category page still has home as it's parent. However, if you navigate from this page onto the Grouping page your breadcrumb does not show the category page as the parent, it will still show Home as the parent.

"No no!", I hear the purists cry. I agree that this is less than optimal for usability, but just remember this - that your category pages are still accessible through the tabs.

Why do this? Well, if I do some calculations on the resulting link count per page it now ends up like this:

Homepage: 54,496
Product category page: 6,256
Product grouping page: 12,726
Product page: 6,447

Apart from my homepage, my product grouping pages (all 30 of them) are the highest linked pages on the site.

Job done. Or maybe not. You could still do more...


Step 5 - Linking nirvana


You might want to consider adding some "children" to your product pages. How about "product reviews", "colour options", and "sizing options". That's 3 pages per product page (6000 products), so that's another 18,000 pages you've just added to your site. That's 18,000 footer links, 18,000 header links and 18,000 breadcrumbs from pages you're not too fussed about whether they rank or not. What does that do to your link count?

Homepage: 72,497
Product category page: 6,256
Product grouping page: 48,726
Product page: 24,448

Now we're rocking.


Summary

Every page should have it's position in the hierarchy (family)

The URL does not need to reflect the hierarchy

The hierarchy does not always need to reflect the user journey

The hierarchy should create more links to your most important pages

I hope I've got the maths correct in this post. If you find any errors be sure to let me know, but either way, the principles behind it do work.

Online Copywriting Best Practice

So how should you write for the web? Here's some top tips to get you started.

1. Keep sentences short

Short sentences are easier to digest. They require less brainpower. Your visitors are one click away from your competitors, so don't make it difficult. If your sentence can be written simply, write it simply.

2. Get the message across quickly

You can get the essence of your communication across in the first few words. Don't make someone read through a whole paragraph to get to the point - they'll not get that far.

3. Use the words of your audience

Your vocabulary needs to use the words that your readers are familiar with. There are 2 good reasons for this. One: it's easier to understand. Two: these are the words most likely to be typed in by your visitors to Google. If you put these words on your page, your copy will already be on the way to being optimised for search engines. Use Google's keyword suggestion tool to research what words to use.

4. Keep paragraphs short

A good example of this is the BBC news site. Each paragraph is simple, short and easy to digest. Long paragraphs are just hard work.

BBC example article: Facebook's developer platform

5. Use good grammar and spelling

It sounds simple, and it is. I don't claim to be perfect either. I do however see sites that lose credibility at the last hurdle with a spelling mistake.

6. Be consistent in your style

By style I mean the way in which you use numbers, abbreviations, dates and terminology. It just makes your site more coherent.

7. DON'T USE UPPER CASE

It's so much easier to read sentence case!

8. Use headings, subheadings and bullet points

This gives your article structure, makes it easy to scan read. Website visitors rarely read every word - they need to be able to scan through quickly and select what copy they feel like engaging with. Ask yourself the question: "can I see at a glance what this article is about from 5 feet away?"

Related articles

• Interesting article from Jakob Neilson: Passive Voice Is Redeemed For Web Headings
• Fantastic site for online copy writing: Nick Usbourne's Excess Voice - copy writing tips for the web

Making a Decision Without All The Data

"You don't need a weatherman to see which way the wind is blowing" - Bob Dylan

I heard this on my ipod the other day and it got me thinking. In business you can often get to "analysis paralysis". By this I mean where leaders only make decisions when they have a full set of data, a fully developed business model, risk assessment models, etc etc.

In many cases making decisions on a well research data set is the right thing to do.

Yet there comes a point where gathering more information is not going to get give you significantly more insight.

Example: for usability testing, Jakob Neilson has shown that it only takes five users to give you enough information to work from. See his article Why You Only Need to Test With 5 Users. Five users can identify 80% of the usability problems. To get 100%, you need to test 15 users. You can get working quicker on resolving the issues if you just test 5, fix them, and then re-test.

Going back to Bob Dylan. The point is this... you can't predict the weather without specific data, but you can predict the seasons. Summer always follows spring, and it's usually warmer. You know this from your experience. Your experience feeds your intuition.

The role of your experience in decision making is considered in more detail by Malcolm Gladwell in his book "Blink". See my recommended reading post for more details.

If you have experience in an area, sometimes you just need to take a view and go with it. You'll probably be right.