<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Ali Hill on Medium]]></title>
        <description><![CDATA[Stories by Ali Hill on Medium]]></description>
        <link>https://medium.com/@ali_hill91?source=rss-62f5386d127c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*DGz8BCMLZk4_0M3RyTqZmQ.jpeg</url>
            <title>Stories by Ali Hill on Medium</title>
            <link>https://medium.com/@ali_hill91?source=rss-62f5386d127c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 18:08:37 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@ali_hill91/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[TestBash Home — Deleting Your Tests, Burnout and Coaching]]></title>
            <link>https://medium.com/@ali_hill91/testbash-home-deleting-your-tests-burnout-and-coaching-3bbb5266058d?source=rss-62f5386d127c------2</link>
            <guid isPermaLink="false">https://medium.com/p/3bbb5266058d</guid>
            <category><![CDATA[testing]]></category>
            <category><![CDATA[mental-health]]></category>
            <category><![CDATA[automated-testing]]></category>
            <category><![CDATA[testbash]]></category>
            <category><![CDATA[ministry-of-testing]]></category>
            <dc:creator><![CDATA[Ali Hill]]></dc:creator>
            <pubDate>Mon, 18 May 2020 17:47:55 GMT</pubDate>
            <atom:updated>2020-05-18T17:47:55.231Z</atom:updated>
            <content:encoded><![CDATA[<h3>TestBash Home — Deleting Your Tests, Burnout and Coaching</h3><p><em>I wrote a review of the TestBash Home format a couple of weeks ago. You can read this here: </em><a href="https://medium.com/@ali_hill91/testbash-home-setting-the-bar-high-for-virtual-conferences-e14a4eb47836"><em>https://medium.com/@ali_hill91/testbash-home-setting-the-bar-high-for-virtual-conferences-e14a4eb47836</em></a><em>. This post will focus on a number of the sessions I found particularly valuable.</em></p><p>I’ve shared, Tweeted and written a blog post about the great virtual community experience that was TestBash Home, but I wanted to focus on what my main takeaways were for the conference by focusing on three of the sessions that really stood out for me.</p><h4>Should We Just Delete It? — <a href="http://João Proença">João Proença</a></h4><p>As a developer, deleting code is one of the most satisfying things about the job for me. Making old code redundant by refactoring and removing hundreds of lines of code — is there any better feeling? Well, apparently a lot of people disagree with that, especially when it comes to deleting tests.</p><p>João began his talk by detailing the reasons people don’t like deleting tests. As humans we have a bias where we feel losses more than twice as hard as we feel gains. We are also susceptible to the sunk cost fallacy (“we’ve already put too much effort into writing these tests!”) and the endowment effect (“well, we wrote this test so we may as well keep it”).</p><p>These are some of the psychological factors that prevent us from deleting tests, so how do we overcome those? João suggests we need to weigh up the value of having those tests vs the cost of having them. <a href="https://www.amazon.co.uk/Continuous-Testing-DevOps-Professionals-Practical/dp/1727132173">There is research</a> to suggest that if bugs are found in the same day they are written they are usually fixed within an hour. With this in mind, we need to keep our test runs short.</p><p>Is there any duplication in our test runs? Can our tests be split into smaller (and therefore faster) tests? Are we spending hours investigating failing flaky tests? These are all questions we should be asking when it comes to asking whether we should delete the test.</p><p>What value do our tests provide? Tests are build in order to target specific risks. Risks can be summarised by the probability of something happening vs the impact of that occurring. Our tests should focus on what the worst scenarios could be for our customers/users.</p><p>Finally, João spoke about alternatives to testing. For example, could we put better monitoring in place instead of a long running test? Could we try disabling the test for a few runs and see how people feel about deleting it after we’ve shown there is no cost to not having the test there?</p><p>João’s talk raised a lot of important points. Teams really need to consider the value of their automated test suite, rather than running tests because “they’ve always been there.” Automated tests are like any other form of testing and should target risks. If the tests aren’t targeting an important risk, or are costing the team a lot of time to investigate, then they should be deleted.</p><h4>Inspiration and Burnout — <a href="https://twitter.com/maryamumar">Maryam Umar</a></h4><p>The most powerful talk of the conference came from Maryam who spoke about her experiences with burnout. Burnout affects a lot of us in the tech industry and Maryam gave a very honest account of her own experiences with it.</p><p>A few years ago, Maryam found the stress of spinning too many plates (work, friends family etc) too much. Her friends noticed a difference in her but Maryam saw the fact that she was experiencing burnout as a failure in herself. She had always strived for perfection throughout her life.</p><p>In order to deal with this, Maryam spoke about a cycle that she found herself in when she was suffering from burnout. She would push herself to aim high, then aim for perfection, which led to stress, then burnout. She would then recalibrate and protect what she had. She thought that this cycle would give her a system to deal with burnout, but unfortunately it never, as she suffered another panic attack.</p><p>Her response to this latest set back was to recalibrate. Maryam began to put herself first and started going to therapy. She got herself a personal trainer as well as a life coach. She shared that having this in place allowed her not to feel guilty about feeling stressed and burned out, but helped her understand how to take time for herself.</p><p>The main takeaway for me from Maryam’s talk was that in order to do amazing work, we need to feel well within ourselves, in our own mental space. Never be ashamed to ask for help if you need it.</p><p>Maryam’s talk is <a href="https://www.ministryoftesting.com/dojo/series/it-s-ok-testbash-talks-mental-health/lessons/inspiration-and-burnout-maryam-umar">available to watch on the Ministry of Testing’s Dojo</a> platform if you sign-up for a free account.</p><h4>Coaching Panel</h4><p>There were a couple of expert panels hosted during TestBash Home. The first of which focused on coaching. The experts were <a href="https://twitter.com/charrett">Anne-Marie Charrett</a>, <a href="https://twitter.com/testertested">Pradeep Soundararajan</a> and <a href="https://twitter.com/AWGHodder">Aaron Hodder</a>. I want to share a couple of gems of advice that came out of this panel.</p><p>Anne-Marie highlighted that you need to be flexible in your coaching style. What worked for one person may not work with someone else. In the age of Covid-19 and remote working we need to show more empathy and also understand that working on Slack may not give us the full picture. Coaching also isn’t just about sharing what you know yourself. Are there any doors you can open for your mentee? Can you put them in touch with someone who knows more?</p><p>Aaaron shared his experience that coaching is a lot about asking “what if the person I’m coaching is right?” It’s about learning from them as much as it is about coaching them. He also shared that “I don’t know, let’s find out together” is a very useful phrase when coaching and that you need to be comfortable not knowing the answer to everything.</p><p>Pradeep highlighted that you need to show vulnerability when coaching, in order to build relationships with our mentees. Learning from failures is also extremely important, especially in your first management role. Learn how to tell when someone is comfortable being thrown in at the deep-end and when they need a little bit more support. As has been mentioned before, not everyone is the same.</p><p>I really enjoyed listening to three experienced panel members share their coaching experiences. As someone who is new to coaching and leadership, I found their advice invaluable.</p><p>======</p><p>I’ve focused on three particular sessions for this post, but TestBash Home provided too much value to write about. The virtual conference organisation and the sessions were superb.</p><p><a href="https://medium.com/u/a41ab7fe385c">Ministry of Testing</a> have moved all of their conferences online due to Covid-19. If you have training budget remaining, then why not check out <a href="https://www.ministryoftesting.com/events/testbash-netherlands-online-2020">TestBash Netherlands Online</a> on 15–16th October 2020?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3bbb5266058d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TestBash Home: Setting the Bar High for Virtual Conferences]]></title>
            <link>https://medium.com/@ali_hill91/testbash-home-setting-the-bar-high-for-virtual-conferences-e14a4eb47836?source=rss-62f5386d127c------2</link>
            <guid isPermaLink="false">https://medium.com/p/e14a4eb47836</guid>
            <category><![CDATA[ministry-of-testing]]></category>
            <category><![CDATA[testbash]]></category>
            <dc:creator><![CDATA[Ali Hill]]></dc:creator>
            <pubDate>Sun, 03 May 2020 12:33:37 GMT</pubDate>
            <atom:updated>2020-05-03T12:33:37.136Z</atom:updated>
            <content:encoded><![CDATA[<p><em>After the resounding success that was TestBash Home, I wanted to focus a post on the way the virtual event was structured. I’ll write a separate post covering the sessions themselves.</em></p><p>On 13th March, just over a week before the UK went into lockdown due to the Covid-19 pandemic, it was announced that the Ministry of Testing’s (MoT) TestBash Brighton was due to be cancelled.</p><p>As someone who was due to attend this event, I was gutted. I wasn’t just gutted to be missing out on the conference itself, but I was gutted for the amazing MoT team who had put a lot of hard work into organising the conference over the past year.</p><p>However, instead of taking it lying down, it didn’t take long for the bosses at MoT to react to the ongoing crisis and pivot. At the beginning of April, <a href="https://www.ministryoftesting.com/events/testbash-home-2020">TestBash Home</a> was announced. Home was going to be a virtual conference, running for 24 hours from 8pm(BST) on Thursday 30th April right through until the evening of the 1st May.</p><p>The idea was that this conference would span all timezones, although some did see it as a challenge to stay-up for the entire 24 hours! When the time finally came for me to sleep (I forced 4 hours between 4:30am and 8:30am), I could rest easy in the knowledge that anything I missed would be available for me to watch on MoT’s Dojo. Adding to the accessibility of the event, the low ticket price of £50 meant those who may not have been able to attend a TestBash in the past, could now do so.</p><h4>A Unique Event</h4><p>Anyone who has attended a TestBash before will confirm that the MoT community is unique. The conferences are the most inclusive events I have attended and the energy and enthusiasm that is shown by organisers, speakers and attendees alike is hard to describe.</p><p>I logged into TestBash Home intrigued (and slightly apprehensive) as to how the MoT team were going to replicate this community feel with a virtual conference. I didn’t need to worry. As soon as the first host, <a href="https://twitter.com/TesterFromLeic?s=20">Vernon Richards</a>, appeared on screen you could tell that the MoT team weren’t going to let the barriers that come with hosting virtual events from dampening that community spirit. Every single host after Vernon (<a href="https://twitter.com/TristanLombard2">Tristan Lombard</a>, <a href="https://twitter.com/JamesEspie">James Espie</a>, <a href="https://twitter.com/Btknaack">Bart Knaack</a>, <a href="https://twitter.com/SheWrestlesTest?s=20">Jenna Charlton</a> and <a href="https://twitter.com/gwendiagram">Gwen Diagram</a>) did a great job of keeping the energy and conversation going for the entire 24 hours.</p><p>There were some really nice touches throughout the conference, that made the event feel unique. Hallway crowd noises were played during the breaks, applause noises were played after talks, a PA announcement sounded five minutes before the next session was due to begin, hosts interacted with attendees through Crowdcast and Slack chats, the bosses at MoT had “Meet the Bosses” sections, attendees were able to present 99 Second Talks and there was even a virtual game created by the MoT team which allowed you to adopt an avatar and interact with other attendees.</p><p>At first, I was worried that all the different methods of communication would be overwhelming. We were told about the MoT Slack workspace, the chat and question areas in Crowdcast and The Club (MoT’s forum). But each of them added their own value. Crowdcast was great for interacting with people during the talks and for asking questions for the Q&amp;A sections, Slack provided a way to interact with people during the breaks and Slack’s replying and threading capabilities meant people could have longer conversations than on Crowdcast and The Club was a place for unanswered questions for the speakers to be answered at a later time.</p><p>Of course, with it being MoT’s first time hosting a virtual conference, there were a few very minor technical hitches. For me, these technical hitches added to the raw energy and enthusiasm that came from the event. In a strange way, it wouldn’t have felt right without them.</p><p>I really liked the pacing of the event. MoT took full advantage of the 24 hours they had. There were plenty of breaks which gave me an opportunity to grab food, stretch my legs, or take a power nap. There were also extended Q&amp;A sessions with the speakers, which was an added bonus.</p><h4>Next Year?</h4><p>Over 1000 people signed-up for the inauguralTestBash Home, and at Crowdcast’s peak, 750+ people were watching the event live. For the majority of attendees, it was their first ever TestBash event. Going by the reaction on <a href="https://club.ministryoftesting.com/t/what-is-testbash-home-like-for-people-who-are-new-to-conferences/37998/22">The Club</a>, it was as great as an experience for them as my first TestBash was for me in Brighton in 2018.</p><p>There are already people asking if the event will be reoccurring next year. We can’t predict what conferences will look like next year, or whether in-person events will resume in the near future, but I believe that there is a place for the pivot of TestBash Home to be included in MoT’s permanent conference calendar.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e14a4eb47836" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The 3Rs of Risk]]></title>
            <link>https://medium.com/globallogic-uki/the-3rs-of-risk-aca137d75ac3?source=rss-62f5386d127c------2</link>
            <guid isPermaLink="false">https://medium.com/p/aca137d75ac3</guid>
            <dc:creator><![CDATA[Ali Hill]]></dc:creator>
            <pubDate>Sat, 22 Feb 2020 15:35:37 GMT</pubDate>
            <atom:updated>2020-02-22T15:35:37.398Z</atom:updated>
            <content:encoded><![CDATA[<p>After writing about using the Test Automation Pyramid to identify at <a href="https://medium.com/@ali_hill91/risk-automation-pyramid-76bfbb666998">which layer we can mitigate a risk </a>and our team’s current <a href="https://medium.com/ecs-digital/describing-our-automated-test-strategy-a3c390c76065">automated test strategy</a>, I thought I’d explore a model that helps us ask questions to identify the risks we are trying to mitigate. The mitigation of these risks does not have to occur through testing, but it may help fuel test ideas.</p><p>**Firstly, I’d like to say that this model is a work in progress, so all feedback is welcome!**</p><h4>What is the Purpose of the Model?</h4><p>Before I share the model itself, I’d like to describe the thought process behind creating the model and why I believe it is important.</p><p>The main purpose of developing a product is to provide value to “someone”. This “someone” is usually our customer, but it can also mean providing value to our company, in the form of monetary income, for example.</p><p>When considering what risks there are to the value of the product I am working on, I often rely on my own past experiences. What issues have been raised in the past that have caused harm to either a) our customers or b) the company? What is the worst possible thing that could happen when the software/functionality we are working on goes live?</p><p>Based on my own answers to the above questions, I’d like to share my opinion that there are three main areas to consider when discussing which risks we need to mitigate throughout the development process. These are:</p><ul><li>(Lost) <strong>Revenue</strong></li><li><strong>Relationship</strong> (with customers)</li><li><strong>Reputation</strong> (of the product)</li></ul><p>These categories are very much business focused. To put it bluntly, in the majority of cases, if the company we’re working for isn’t making money then it doesn’t matter how high the quality of the software we’re developing is. On a more human level, our Relationship with our customers is vital for retention. The last category of Reputation can massively impact growing our customer base and have financial implications.</p><p>I’ll explore each of these categories in more detail after sharing the model.</p><p>Of course, not all of these three areas apply in every case. For example, if your customer is an internal team or you’re operating as a not for profit organisation then you may not have to consider Revenue as a potential risk area.</p><h4>The “3Rs of Risk” Model</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Xs0aH3enzI8aD3r58w3knA.jpeg" /></figure><p>The model itself uses what I’ve identified as the three main areas of risk. The overlapping circles signify that risks to Revenue, Reputation and Relationship can overlap. For example, there may be risks identified which could result in lost Revenue that also result in damaging your Reputation. Poor app security, resulting in a security breach, may result in a damaged Reputation in the media and also damage your Relationship with customers.</p><p>I’m now going to cover the three areas in more detail and provide some examples so I can illustrate why I’ve identified these areas as threats to software value.</p><h4>Revenue</h4><p>Losing Revenue has to be one of the biggest risks for the majority of businesses that we will work for throughout our careers.</p><p>When identifying what types of tests we need to run, we should consider what risks there are to losing Revenue. Whilst exploring the three areas of risk, we will imagine we’re working for an online retailer. Based on that scenario, I’ve identified a few risks to Revenue in the circle below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5Q1_kpbSop1sTsIjZF-cLw.jpeg" /></figure><p>Losing sales is probably your biggest threat to revenue. What tests/types of testing can you run to identify that there are no major issues resulting in lost revenue for the customer? If completing the checkout process is identified as one of your main critical flows, can you automate a test at one of the layers of your application to flag if there are any changes in this area?</p><p>Does your product function correctly on multiple platforms (desktop, mobile etc)? What percentage of customers are using each platform? Will that help you decide which risk is more important to mitigate?</p><p>How easy is your software to use? Can you ensure your product is observable? If customers aren’t completing the checkout process then is it a usability issue? Or is it a performance issue? Do we need to identify more performance tests to run in our process to cope with peak loads and stop people abandoning transactions?</p><p>The aim of the Revenue circle is to initiate conversations around what risks there are to losing Revenue with the software we’re developing. Lost revenue impacts the value we are providing to the business.</p><h4>Relationship</h4><p>Our Relationship with our customers is key. We need to be able to connect and empathise with them in order to meet their needs and provide them with valuable software. Having a good Relationship with customers is vital to retention.</p><p>Let’s again use our example of an online retailer. What risks are there to our Relationship with our customers?</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*ijndoufSmaCZnIdXOhyg-w.jpeg" /></figure><p>The first word that comes to mind for me is trust. Do our customers trust us with their data? How secure are we? In order to try and mitigate this risk could we use the <a href="https://en.wikipedia.org/wiki/STRIDE_(security)">STRIDE</a> model?</p><p>How well are we communicating with our customers? Have we tested confirmation emails are being sent once transactions are complete? Do we have automatic communication when the website is down (a temporary web page for example)? What tests could we add to cover that?</p><p>How usable is our website? Is it accessible? Are those (all too popular) annoying pop-up adverts getting in the way of people using our site? Are people frustrated?</p><p>How about our loyalty schemes? Have we tested these? If loyal customers were to lose ‘points’, how would they react?</p><p>Identifying and mitigating risks to our Relationships with customers (even if they are internal) is vital if we wish to provide a valuable and reliable product.</p><h4>Reputation</h4><p>The third and final area of risk in my 3Rs of risk model is Reputation. What issues are likely to effect your company/product’s Reputation and prevent you from delivering a valuable product?</p><p>Our shopping application/web page may have a number of risks associated with Reputation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fEXDUwu3wOObyjqURapLuQ.jpeg" /></figure><p>Security breaches are possibly the most reported software bugs within the media and social media, and for good reason. As well as damaging your product/company’s Relationship with customers, security issues can also have devastating effects on your Reputation. What are we doing to mitigate security risks?</p><p>If a customer is looking for a nearby store, are we hooking into Google Maps (for example), or have we created our own version of a interactive map? How would we test the design and usability of the map? The reason I’ve added this to Reputation is the disastrous rollout of <a href="https://www.silicon.co.uk/mobility/mobile-apps/apple-maps-launch-embarrassment-196350">Apple Maps in 2012 </a>and the media coverage that followed.</p><p>Are we compliant with industry regulations? If not, it could result in a high profile court case or disciplinary procedures. What tests (or monitoring) do we have in place to mitigate that risk?</p><p>Is our app highly available? How will downtime be handled? How quickly can we recover? Do we carry our war games or chaos engineering tests?</p><p>Are we an ethical company/product? How do we deal with customer complaints? If we advertise that “x% of purchases goes to charity” have we ensured that this is <em>actually</em> the case?</p><h4>Final Words</h4><p>I hope that you will find this model useful when deciding which areas of risk to focus on. Like all models, I don’t believe it is complete, but I hope you can use it to facilitate conversations around the risk of lost Revenue, customer Relationship, product Reputation and how to mitigate those risks through the design or testing of your software.</p><p>As always, feedback on this model is welcomed. Does it make sense in your context? Have you found it useful? What would you change?</p><p>If you would like to read more about testing and risks then I’d recommend Dan Ashby’s <a href="https://danashby.co.uk/2019/05/01/risk-based-testing-part-one-talking-about-risks-over-types-of-testing/">Risk Based Testing series</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aca137d75ac3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/globallogic-uki/the-3rs-of-risk-aca137d75ac3">The 3Rs of Risk</a> was originally published in <a href="https://medium.com/globallogic-uki">GlobalLogic UK&amp;I</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Describing our Automated Test Strategy]]></title>
            <link>https://medium.com/globallogic-uki/describing-our-automated-test-strategy-a3c390c76065?source=rss-62f5386d127c------2</link>
            <guid isPermaLink="false">https://medium.com/p/a3c390c76065</guid>
            <category><![CDATA[test-automation]]></category>
            <category><![CDATA[software-testing]]></category>
            <dc:creator><![CDATA[Ali Hill]]></dc:creator>
            <pubDate>Mon, 17 Feb 2020 08:04:58 GMT</pubDate>
            <atom:updated>2020-02-17T08:04:58.207Z</atom:updated>
            <content:encoded><![CDATA[<p>After my previous post describing <a href="https://medium.com/@ali_hill91/risk-automation-pyramid-76bfbb666998">how I use the Test Automation Pyramid</a>, and a <a href="https://twitter.com/FriendlyTester/status/1228463658960924672">Twitter discussion with Richard Bradshaw</a> who pointed me in the direction of his talk <a href="https://www.youtube.com/watch?v=83aHKC5C14Q&amp;feature=youtu.be">“Pyramids are Ancient — Test Automation Strategy”</a>, I decided it was about time that I reflected on my current team’s automated test strategy.</p><p>I’ve asked myself the following questions: what automated tests do we write? Why do we write them? What value do they provide us?</p><h4>Background: What Does Our Team do?</h4><p>We are a team of consultants who are responsible for writing security controls in AWS. These controls are set up in every AWS account used by teams within the organisation we currently work in. Our customers are anyone who uses AWS within the organisation.</p><p>The majority of our controls utilise AWS Config Rules which trigger AWS Lambda Functions, written in Python, and then report a compliant or non-compliant status depending on the AWS resource the rule is evaluating. We deploy this infrastructure using Terraform, which adds another layer of complexity.</p><h4>What Automated Tests do we Write?</h4><p>Our automated test strategy can be summarised with the following illustration (yes, I’m using my favourite, the test automation pyramid):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*M6XeNtnKo6l6KkjrYa4_Tw.jpeg" /></figure><p>Let me explore these in more detail:</p><ul><li><strong>Bottom Layer: </strong>at the bottom layer of our pyramid we have our unit tests. As I mentioned above, the majority of our code is written in Python and is deployed in the form of AWS Lambda Functions. We ensure that the critical functionality of each function is tested using Python’s <em>unittest</em> framework.</li><li><strong>Middle Layer: </strong>here we have two different types of tests. Our Terraform Tests ensure that we have configured our Terraform correctly to support the functionality of our controls. We also have what I’ve described as <em>boto3</em> tests. These tests make a <em>boto3</em> API call to AWS to ensure that the resources we have deployed are running correctly.</li><li><strong>Top Layer: </strong>our top layer, and by far our most expensive layer of tests, are our end to end tests. These tests typically trigger an action in AWS that will trigger one of our Lambda functions to run. We then have to analyse the results that our function produces and assert an expected outcome, depending on the action we have performed.</li></ul><h4>What Value do These Tests Provide?</h4><p>I’ve explained what tests we write and where they fit into the Test Automation Pyramid, but <strong><em>why</em></strong><em> </em>do we write them?</p><p>I decided to scribble my thoughts on why we write the different types of tests we do within our team. What value is provided by each of the blocks of our pyramid?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g91cihHs9bcf9i0zWsi2gw.jpeg" /></figure><p>At the top of our pyramid, our End to End tests are slow-running and are difficult to maintain. We tend to only write these for our most critical pieces of functionality. I’d argue that AWS has not been built with testing in mind (throttle limits on APIs, time it takes for logs to appear), but that is an entirely different topic.</p><p>They do still provide value. They test each and every layer of our individual security controls and they simulate the experience that we are providing our customers. The information they provide us (and auditors!) is really important.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZjoVdCqMWhl9pkZ3GJ1rPw.jpeg" /></figure><p>Next up is the middle layer. Our Terraform/<em>boto3</em> tests allow us to ensure that we have configured our declarative Terraform code properly. Have we created our own resources to be compliant (e.g. not exposed our EC2 instances to the public)? Has Terraform Enterprise deployed our infrastructure correctly? This is arguably the first layer that Terraform can be tested at, so these tests act as unit tests for our Terraform code.</p><p>By using <em>boto3 </em>we can also ensure that all of our individual Terraform resources are working together correctly. These are non-intrusive checks that allow us to (resonably) quickly determine the state of our deployed features. For example, has CloudWatch been enabled, for monitoring purposes, on one of our EC2 instances?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vNNtBsQ0OWZTd13baiFDKA.jpeg" /></figure><p>I don’t think I need to spend too long describing the benefits of unit testing in this blog.</p><p>In the context of our team, we practice TDD which allows us to ensure that we’re developing the functionality we have agreed as a team and with our stakeholder. The unit tests act as scaffolding to our functional code, meaning any functional changes are safeguarded by these tests. We also have a relatively slow deployment process, so having these checks to run before we kick that off is often invaluable.</p><h4>Final Words</h4><p>I hope you found this insight into the value that the different layers of tests provide to our team useful. Test strategies like this cannot be copied and pasted and they are heavily dependent on your team’s context.</p><p>If you’re looking for advice on how to develop a test strategy of your own, whilst considering the different layers you should be testing at, then I’d recommend giving <a href="https://www.youtube.com/watch?v=83aHKC5C14Q">Richard Bradshaw’s talk</a> (which I also linked above) at SauceCon a watch.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a3c390c76065" width="1" height="1" alt=""><hr><p><a href="https://medium.com/globallogic-uki/describing-our-automated-test-strategy-a3c390c76065">Describing our Automated Test Strategy</a> was originally published in <a href="https://medium.com/globallogic-uki">GlobalLogic UK&amp;I</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Risk Automation Pyramid]]></title>
            <link>https://medium.com/globallogic-uki/risk-automation-pyramid-76bfbb666998?source=rss-62f5386d127c------2</link>
            <guid isPermaLink="false">https://medium.com/p/76bfbb666998</guid>
            <category><![CDATA[software-testing]]></category>
            <dc:creator><![CDATA[Ali Hill]]></dc:creator>
            <pubDate>Fri, 14 Feb 2020 19:55:41 GMT</pubDate>
            <atom:updated>2020-02-14T20:00:27.881Z</atom:updated>
            <content:encoded><![CDATA[<p>I know what you’re thinking; “not another Test Automation Pyramid article!”. Well, I’m afraid it is. Seb Rose recently wrote an article titled <a href="https://cucumber.io/blog/bdd/eviscerating-the-test-automation-pyramid/">‘Eviscerating the Test Automation Pyramid’</a>. This article got me thinking about how I use the Test Automation Pyramid as a model to inform my testing.</p><p>To illustrate my thinking I’ll begin by sharing the original Test Automation Pyramid, first published by Mike Cohn in his book “Succeeding with Agile” in 2009:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Rx9DKfsrXxhnTsax2GKV8w.png" /></figure><p>The pyramid is simple. It argues that we should write more automated Unit tests than any other type as they are easier to maintain and provide fast feedback. Automated Service tests exercise a service layer of the application, but avoid the complexity of UI frameworks. UI (or end to end) tests often exercise every layer of the system under test but are notoriously brittle and difficult to maintain.</p><p>There have been many debates around the Test Automation Pyramid and its usefulness. A quick search engine query will return multiple articles about the model (there’s a certain irony of me adding yet another one!). Some people find it too limiting and believe it leaves too much unsaid and oversimplifies testing. It doesn’t cover exploratory testing, for example.</p><p>I like the Test Automation Pyramid. I like it with the caveat that it is <strong>ONLY</strong> used to consider which <strong>automated tests </strong>to<strong> </strong>write. I don’t believe one model can cover an entire testing strategy, although I’m happy to be proven wrong!</p><h3><strong>How I Use The Test Automation Pyramid</strong></h3><p>I <a href="https://twitter.com/ali_hill91/status/1227923966452490241?s=20">wrote a Tweet</a> about how I have used the Test Automation Pyramid in the past. I identify the risks of the functionality being tested and work my way from the bottom of the pyramid (lowest possible layer) to the top (highest possible layer). I ask myself at every layer: <strong>“can we mitigate the risk here?”</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DYLsOIgkAdJRFm-HsOzXDg.jpeg" /></figure><p>This is a simple addition to the original model, but it helps us to ensure that we are mitigating the risks we have identified at the lowest layer possible.</p><p>Here’s some notes I wrote using an example of a checkout page for an online retailer:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*r7ZUZ8Xq8m51HPvePFZAiQ.jpeg" /></figure><p>I have quickly identified a number of risks that we may identify when testing a checkout page. Heuristics for identifying risks is not something I’m going to go into here, but it’s definitely an idea for a future blog post.</p><p>A big risk to the business, in the above example, is that a <strong>customer cannot</strong><em> </em><strong>complete their order/checkout</strong>. This would result in a frustrated customer and lost revenue for the business.</p><p><strong>Can we mitigate this risk at the Unit layer?</strong> Definitely. A number of unit tests can be written with a number of scenarios (hopefully spawned from a Three Amigos session!) to ensure that the component of code will successfully complete the checkout.</p><p><strong>Can we mitigate the risk at the Service layer?</strong> Again, yes. We will want to ensure that any API and database calls that are required to complete a transaction can be completed successfully.</p><p><strong>Can we mitigate the risk at the UI layer?</strong> Given our previous tests, we may want to add a test to ensure the “Checkout” button is there and can be selected. We’ve covered the majority of the functionality at the other layers, so an end to end test may not even be necessary. Either way, we’ve reduced the number of testing needing done at this layer as we’ve tested the functionality below.</p><p>I would then follow the same process with the other risks we have identified.</p><h3><strong>Final Words</strong></h3><p>Again, I want to highlight that the Test Automation Pyramid should not be used as a model for <strong>every </strong>test that you write. It should make up part of your test strategy, not all of it. It is an extremely useful model for when you’re considering the addition of automated tests, but exploratory testing ideas, for example, should be considered outside of the pyramid.</p><p>I welcome feedback on whether you use the Test Automation Pyramid and and how you use it.</p><p><strong>TL;DR</strong>: Identify the main risks of the system you are writing automated tests for and aim to mitigate these risks at the lowest possible layer.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=76bfbb666998" width="1" height="1" alt=""><hr><p><a href="https://medium.com/globallogic-uki/risk-automation-pyramid-76bfbb666998">Risk Automation Pyramid</a> was originally published in <a href="https://medium.com/globallogic-uki">GlobalLogic UK&amp;I</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>