The Getting Started Page – Transitioning your users from cautiously optimistic to happy customers.

Anyone that has been around web apps for a while would have noticed that most of them are doing something different UI wise when a user first signs up or when they are first seeing a new feature. Be this an overlay of information within the natural usage of the product or be it a completely separate page.

For example if you sign up for Twitter nowadays you will be sent through a series of pages which encourage you to follow a host of people that you may find interesting. The logic being that someone who signs up and is greeted with a completely blank feed may easily just give up on the service, an engaged user with a full feed is much more likely to stick around.

When you first see a new feature of Facebook you are usually presented with a Guider like overlay which points to the relevant parts of the page and describes how they should be used.

When we first approached a getting started process for AffClicks we modeled it off the UserVoice getting started page, which is a single page (now separated into basic and pro tips tabs) where there is a list of things a user getting started should do with the instructions to complete them. Completing them generally requires navigating to another parts of the app and using the regular interface. When tasks are completed they are marked as being completed by being given a tick and crossed out.

The original AffClicks version is show below:

The old getting started screen.

As you can see it has some fairly dense slabs of text and can quickly become confusing. The left sidebar is there because the getting started page was built into the user accounts/ settings page. This just succeeded in distracting the user (I have seen users the signed up, click on all the side links, then leave). The font was relatively small and the tasks were close enough together that people could skim them without really reading them. They were also ordered wrong (more on that below). By observing user actions my guess would be that only really motivated and fairly technical users were getting much out of it.

Like many who have worked on their startups for a long time I can sometimes delude myself that everyone from our target niche will be highly motivated to get setup in AffClicks. The truth for most startups though is that a large number of users will only have a passing interest in using a product at first, they may give up before they really see any value.

We needed to step back and think about which actions were the highes priority for new users. The first thing we really want a user to do is to connect with an Affiliate Network. This kicks off automated importing, so any sales they get end up appearing in AffClicks. With sales coming in users can get a feel for how AffClicks works and how their information is presented. Getting sales also populates a weekly update we send out, which can remind users that their sales are still being collected and encourage them to come back to AffClicks and continue to use it if they had gotten busy and forgotten about it.

The first tab of the new getting started screen, allowing a user to add an affiliate network directly.

In order to streamline this connection process we duplicated the UI for connecting a network in the getting started process. This is a positive and a negative. A positive in that the first and most important action is even easier to access, a negative in that it could be slightly confusing to the user that adding subsequent networks occurs on a different page to the first. This way a user can complete this task in a matter of seconds from their first signup.

Another important change as you can probably see from the screen above is that the tasks have been separated into tabs, as not to further confuse the reader as they skim ahead.

The next getting started tab, instructions to add the tracking snippet

The second task, adding a tracking snippet to their site is potentially a lot harder for a user to do than filling out a form. They have to add the snippet to their pages HTML and also add class=”affclicks” to their affiliate links. In order to split it up, the user is now asked what method of tracking they use before they are presented with how they can pass the task, as to not prematurely confuse. This could probably still use some more work, not just in explaining how the snippet is installed, but exactly the benefit that the user will get from it. We always have to remember that many users may have only quickly glanced over what we have to offer before signing up, not read all our copy and blog posts. Some good examples  for snippet installation are StatCounter and Google Analytics. StatCounter is probably more interesting because they have gone to pains to make the process as easy as possible for as many users as possible, listing out individual platforms and providing precise instructions. Google Analytics offers a clean and simple interface for the snippet but is probably closer to what AffClicks does right now, provide the code and instructions that users might need to be slightly techy to follow. Something such as a wordpress addon would be definitely on the cards if we find enough users using wordpress with AffClicks.

The third task, connecting to Adwords is currently unavailable, as we must pass Adwords minimum 3rd party app requirements. It is still there though and the PPC features are still prominent in the app but just not available yet. As some users may have signed up because of what they have read about our PPC features we want to remind them that they aren’t far off when possible.

The new Getting Started page has been done in conjunction with improving the signup process, first in removing the need for someone to register interest and then be invited and second in putting a lot of time into the form to ensure it is as intuitive as possible. It makes best guesses on values when possible to reduce the amount of fields the user needs to change.

I think with user acquisition for web applications in general there are 2 main stages that a user goes through.  When they first visit your site, possibly knowing very little about your app, I like to think of things very much akin to an ecommerce site. Generally in that situation a user has very little sticking them to your site, at any second they could get bored, confused or misdirected. This could result in them switching tabs, clicking a link off your site or just hitting the back button. In this stage I think you have to do everything possible to streamline the experience. Kind of bootstrapping in a way, you give the user the minimum amount they need to know to get to the next stage, signing up. Then you give the user just enough information to perform their first in app tasks. Then you can slowly layer on complexity. Once the user is using your app, getting real value from it, be that value time/ money saving, new insights or just plain fun, they transition into the second stage.

In the second stage a user is going to be a lot more sticky, you can throw some more complex concepts at them without alienating them. You can reasonably expect them to come back without having to put special effort into reacquiring them. So a big part of user acquisition is going to be what strategy you employ to transition users to the second stage.

For AffClicks a lot of time and effort has gone into this recently and the early results are promising, seeing more users signing up and completing some in app actions. This will be something we will continue to focus on as we grow.

I am interested though, what kind of strategies others are running to best hold onto new users and sell them enough on your concept before they navigate away for any number of reasons in that first stage.


Aquihires and Product Shutdowns

A growing trend that I have noticed over the past year are startups quickly shutting their service shutting down after asking users to buy into their vision and investing a lot of time to make the most of the product/ service. This has commonly been through aquihires (When a company buys another almost purely for talent and has no intentions of moving forward with that companies products) or startups just giving up on their idea (sometimes due to circumstances/ funding and others due to the team pivoting the product to a new market or even starting an entirely new product).

This is a controversial subject, in particular in reguards to aquihires. Views are often split between the crowd that congratulates the founders on their efforts and wishes them well with their new role and the crowd that feels that they are selling out on their products users.

2 examples that have come up in the last week are Posterous being acquired by Twitter and Kevin Rose’s company Milk abandoning their first iphone app, Oink.

Posterous is a blogging platform, originally slated as the simplest way to blog online. Simply send them an email and and get your blog started, then either post through email or in an online interface. It is one of the more prominent companies launched through Y Combinator []. For a lot of it’s life it has been in somewhat of a battle with Tumblr [] for users, not that the blogging space is a zero sum game but both operated on revenue models which required immense scale to create a largely profitable and sustainable company. Tumblr had gained the upper through being less of a traditional blogging platform and a bit closer to a social network of blogs and had a lot of users that just shared photos. Posterous has always been about more tradition blogging with a smaller user base. Sensing this trend Posterous launched their own take on the more social/ photo sharing concept called Spaces.

Some may call the eventual Posterous acquisition a smart business move, returning a decent amount to investors and giving the founders a nice pay day. Many people though that have come to depend on the platform for their blogs though see it differently. Twitter will likely be making no effort to extend the life of the platform and the Posterous employees heading to Twitter will be working on other Twitter products.

The Oink case is an interesting one, the parent company Milk is a case of investors throwing money behind an internet celebrity founder with a past success (if you can call ultimately Digg a success). The idea behind Oink was to rank things around you, pretty much anything, whether it can actually be attached to a location like a Coffee at a Cafe or something that doesn’t just exist in one place like a laptop. It was given 4 months after launch until it was shut down, one Hacker News commenter likens it to business ADD. Given Kevin Roses popularity the app quickly shot up in downloads, hitting 150,000 in just over a month. I guess it was likely not meeting expectations in adoption, engagement or product direction. Milk also indicated when they started that they would be experimenting with lots of ideas when they started.

It is more the trend as a whole that worries me rather than any specific example. Time and time again we see companies that are one day selling customers on their products and exciting futures and the next annoying that they have been acquired and the product is shutting down. This time a fair bit of discussion is spawning such as this post drawing an analogy to real estate landlords (discussion: I worry that in the future users will be jaded on investing their time into getting started and making the most of new services. They will remember the last time they uploaded all their photos to that “cool new service”, only to have them shut down either require them to export the photos or potentially lose them and any content that has been generated around them.

This is all related to the interesting post “dont be a free user”. Basically it states that when you pay for a service you have a higher chance that that service will stick around, that the service will be generated real revenue and not be so beholden to the next VC round or the VC board that now controls the company. Sure there is still the chance that these companies will be acquired but there is a far higher chance that the a company will either survive without needing addition funding or to be purchased or that when they are brought their product is actual valuable for the acquirer to continue to run.

Many companies run a model where they try to delay revenue generation for as long as possible to build up a critical mass of users. This works for some companies, Twitter being a notable example. Others though, you feel that they would have been better off working out the revenue model before they reached the point where they could no longer continue without outside assistance.


A while back I started playing around with a service called Ifttt (If this then that). It basically aims to glue the internet together. There are loads of online services out there that we use. Usually though they are siloed in that unless they have expressly decided to integrate with other services we can only use them in isolation. At first this isn’t apparent because everything is integrated with the usual suspects, such as sharing/ liking something on Facebook.

The first one I setup is to post my blog posts to Twitter, I’m sure there is probably a WordPress add on to do this but nowhere near as quick and easy as with Ifttt. This was a while back but I got thinking about it again when I quickly added the #b03 hashtag to the auto generated Tweets for the March blog challenge. The possibilities are really endless with the channels that they have available and they seem to be adding more all the time. They have a list of recipes that is full of actions that others have come up with to give you ideas. The most used one right now is to synchronise your Facebook and Twitter profile pictures, saving you the time to update both. Some other cool ones are: Star an email in gmail to send to Evernote, Backup pictures on Facebook into dropbox, or save email attachments to Dropbox.

Looking through the recipes, it seems like the RSS feed channel is a big enabler right now, as anything being put into any kind of RSS feed can be acted upon.

Currently upon making new posts for AffClicks I am spending a fair amount of time pushing them out through all the social channels and relevant social news top lists like Hacker News, Reddit and, while I don’t think Ifttt is quiet up to the task yet of automating that pipeline it looks to be heading in that direction of taking that work you would usually have to do and fully automating it. Getting the right words down to create valuable content when blogging is the important part, pushing things out to get maximum exposure (what I am usually aiming for with AffClicks based stuff) is the repeatable bit that just takes up a chunk of time.

Investors seem to agree that ifttt has a bright future, recently investing $1.5 million dollars to further fund development. As developers, we tend to look to automate everything in our workflows that we are able to, giving us more time to focus on the parts that matter. So a service like this naturally appeals to us. What will be interesting is whether this can cross the chasm and gain a mainstream following with regular internet users, as the TechCrunch article mentions:

The simple, big-text interface has managed to guide a wide variety of users through it and get them hooked but some pieces of it. But it still has some rough edges. Some pieces, like the curly brackets section for setting up detailed actions, can be confusing to people who aren’t familiar with programming.

The 30 day blog posting challenge

I have been watching some of the buzz online about Steve Hopkins 30 day March blog posting challenge. While I haven’t really committed to it anywhere and have missed a few days at the start of it it has got me thinking more about the kind of topics that I want to write about, things where I feel I either have something to add or something interesting to share. I don’t think I will be able to keep up to a post a day but want to try more own version for it.

Steve has talked a fair bit of late about creating a habit calendar and keeping up your chosen habit every day in order too see how long a chain of days you can create and cement the chosen pursuit as a permanent habit. In Steve’s case and for the purposes of the March challenge, one of the main habits being formed is that of blogging. In my case last month I made 2 blog posts, one here and one on the AffClicks Blog. This month, while not aiming to post every day I am aiming to post a lot more. Basically any time I have something that I think I can order my thoughts well around a blog post and also to get through a few more of the Affiliate Marketing topics I want to cover on the AffClicks blog.

So what I want to find out is not wether I can post every day, which I feel with startup demands in might be hard, but whether I can make slower but sustained improvements to my blogging habit.

Also I will be following along and referencing in posts what others are posting for the challenge. I was tossing up the idea of trying to make a small web app to track peoples progress before it got underway but didn’t have the time to get stuck into doing it.

(As an aside, the wordpress gods weren’t on my side tonight, spent way longer making things work than writing this post)

Web Application Cycles

After following web applications for a while you end up seeing cycles. A new product comes along which promises to be dead simple, straight to the point, giving you the user a much better experience than you currently have. The product is lacking features but it’s developers claim that the lack of features is a feature.

As a product gains traction though it attracts a diverse set of users, each bringing their own expectations and use cases to the table and many giving feedback of “It’s great but I’d love if it also did…”. Many of these suggestions in isolation sound great, who wouldn’t want that extra functionality.

Generally things can go one of 2 ways, many of these features can be added and the product can continue being iterated and mature. Or most of all feature requests can be knocked back in order to keep the product as simple as the first day it come out. This tends to result in the project stagnating though, things work better when they are constantly moving and come to a grinding halt when they aren’t.

Either way, the product will eventually be ripe for disruption, either it will now be the slow moving behemoth due to years of feature creep or it will be a simple product that has failed to include many things over time that the disruptive competitor can be based around. Things move so fast that the simple product over time may have to be refocused and mostly rebuilt in order to both stay simple and to cover just the right subset of use cases to remain relevant.

While something new may look new and innovative, maybe it is more like something from the previous cycle than we think. An example is how Facebook greatly simplified the social network from the mySpace experience. Facebook identified exactly what people wanted to do on a social network: easily find and chat with their friends, share pictures and organize events. As time has gone by though there are now many aspects to Facebook and it is becoming more complicated all the time, one look at the myriad of privacy options or the different types of content that can now appear in your feed can convince you off this.

We now have the next way with social apps like Path, which is a lot closer to the original Facebook than Facebook is now. Path focuses on the people in your life you want to connect with the most, while your Facebook “friends” list may have ballooned out Path is all about cutting it back. Path is also riding the mobile wave, being predominantly mobile only. In a few years time will we be able to add the next way to this cycle? Will todays simplistic products have either grown to look like todays incumbents or stagnated and forgotten about?