I ll start with 3 advices 1 Start with reading the subtitle2 Then read the subtitle3 And finally read the subtitleThis book is not about analytics in general or even analytics in lean production development It s just only about analytics in terms of building developing a Lean Startup If you re fine with this assumption, you ll love it Why TONS of examples, all of them under the actual names and we re talking about really known brands here VERY comprehensive approach to various start up scenarios practical approach you can easily utilize in your situationMy two favourite bookmarks about growth hackers the practical difference between correlation causability about the role of revenue stage in the LAIt may sounds odd well talking about the analytics book, right , but it was a very captivating read Recommended. The one metric that matters OMTM is the key measurement you need to be be making of your current endeavour My endeavour to learn all about building products and the business that build products seems to be measured right now by the number of books I m reading on the topic, so 1 to that Lean Analytics builds on the Lean Startup and presents some quite prescriptive ways to views various types of internet businesses It s an easy quick read but I don t think it will stand the test of time, it s too rooted in today s view of the importance of mobile apps and SAAS businesses, which is sure to be transient.But, as a tool for today it s a great read Thoroughly recommend it.It has a great section on intrepreneurs as well.One of my favourite parts is the rule of 3 s which goes something like this at a certain size you will likely have 3 levels of management board, executive and operational you likely have 3 big assumptions in your current business model be clear what they are pick the top 3 these kind of assumptions will be ones paying your pay role or internally keeping your stakeholders at bay they are fairly static ie not changing month by month they are agreed at the board level and communicated to the entire company at the executive level you agree the 3 main tactics to make your assumptions reality ie what actions are you taking to move those metrics in the right direction new features, new marketing campaign , this is like your sprint goals in agile on a daily basis everyone in the company is performing individual tasks related to the tactical actions, these are experiments you run to attempt to move the metric, each is a small hypothesis, experiment and learn cycle, for each of these tests, what 3 experiments are you running Three assumptions monthly at board level three tactics weekly at exec level 3 experiments daily at staff level Another disjointed analysis with specifics about my company that I shouldn t tell people.Lean Analytics10 Ratios are good for comparing factors that are opposed or have some kind of tension A good metric changes behavior.12 Types of Metrics Qual Quant, Exploratory Reporting, Leading Lagging, Correlated Causal If you find a relationship between something you want and something you control, you can change the future 14 We care about active users, because it probably lead indicates our churn rate This requires a lot of different queries to diff DBs 18 Analyze patterns of engagement and desirable behavior, find commonalities.16 Knowledge Matrix Known Knowns are facts that may be wrong and should be checked against data Known Unknowns are questions we can answer with automated reporting Unknown Knowns are intuitions we should quantify Unknown Unknowns are located through exploratory reporting and where we can locate unfair advantages and new insights.18 Pivot hard or go home, and be prepared to burn bridges 19 Churn is a lagging indicator Cohort analysis comparing customer groups over time is the road to leading indicators 20 Are the metrics we track helping us make better decisions faster 22 A company assumed active was 4x a week, when it turned out to be only 1x a week to great success Things they tried Clarified signup flow, added explanatory copy Daily email notifications, transactional emails tied to actions on site.24 A segment is a group that shares a common characteristic Segment visitors then compare segments to each other to understand differences in metrics Look for disproportionate relationships.26 Cohort analysis allows patters to emerge across customer lifecycles.27 We can test anything, but focus on the critical steps and assumptions.34 Venn Diagram Expertise, Desire, Monetizable E D Learn to M E M Improve D D M Learn to say no All 3 Victory 52 Use Google Analytics multi channel conversion visualizer to see which referral sources are combining to influence visitors57 Find days where unsubscribe rate is high, then find out why Need to tweak the unsubscribe process to get better resolution on this Subs expire some time after cancellation, so a decrease in sub numbers on a given day is not necessarily indicative of anything Find the action, not the result.67 When you subscribe to QS, you re subscribing to a slice of our personality.92 Backupify focuses on monthly recurring revenue They watch churn, but are not going to focus on it until they hit the 10MM revenue level.97 Properly calculating churn Select a time period Average the number of customers at the beginning and the end of the period Divide the number of cancellations by this number To increase data integrity, measure churn daily using the method in 57 125 User Generated Content Use the forums as a source of UGC and find ways to repackage it and syndicate it.154 QS is in Stage 5, we have revenue and we are beginning to branch into new verticals We need an ecosystem to help us cross the gap from niche site to industry staple.159 Find out what s actually important to people Get inside their head Delve for this information aggressively.211 Refresh the 3 year plan every 18 months Align the entire company around the vision.213 The best companies warehouse every possible data point about their site s interactions and use only the data they need Rally a software co records everything from kernel level performance to HTTP based user gesture interactions between the browser and software They can then correlate changes in site performance to user behavior and vice versa 256 metrics for stage 5 scale Attention is a precious commodity Don t waste the visitor s attention on stuff that doesn t matter Internally, compare the metrics that matter across channels, regions, and other segments to find efficiencies and inefficiencies.258 Get a better understanding of what new visitors really do 260 Limit the company s vision to a 3 pronged strategy.261 For each C Level strategic assumption, what are the 3 line level tactics that can be used to survey, test, prototype, then fill kill quickly 262 Enable both emotionally and technically anyone on staff to run a split test Give the line level a wide range of flexibility 263 Scale stage summary Focus on the health of the ecosystem and ability to enter new markets Pay attention to compensation, API traffic, channel relationships,and competitors These are no longer frivolities or distractions.274 Hypothesis to test most people unsubscribe beacuse they don t need our service any, not because we re crappy Limited data bears this out but we need to conduct FAR better exit surveys 281 Product pricing has nothing to do with cost, and everything to do with what the customer pays and how they derive value from the product 283 Try an intentionally absurdly priced package to anchor high prices, as well as to see if anyone actually bites.286 We have no real way to measure virality We can try to use the affiliate program but I suspect many will refer by word of mouth if given an easy way to do so How many do they refer and how quickly 287 Sharing by email accounts for 80% of social sharing, usually between small groups of people Remember Emerson Spartz theory about bridge nodes Which groups of people are likely to be conduits between other peer groups 326 288 3pm is when people are most likely to open an email If software permits, time newsletters on a per user level, based on signup time.290 Site load time matters a great deal Spend a lot of time to get this down 302 Negative Churn is the long tail of brand awareness We might convert some Long time listeners, first time callers after a year, or longer Focus on customers that have been on our mailing list for a long time but are not subscribers 304 Top SaaS companies increase revenue per customer by 20% a year Can we upsell the Buylist product enough to hit this target Can our price increase for new subscribers help us hit this level 324 What kind of content do different traffic sources expect Twitter Time on Site is disproportionately high Why is this 325 Find outlier content and promote it heavily Unlocked Insider articles are great for this, as are nexus pages like our BOTG page.326 Most sharing is intimate Each share generates an average of 9 visitors, per the book s data Can we find this number for our site Book data 5 1 Twitter, 36 1 on reddit Sharing happens from a groundswell of small interactions between colleagues and friends rather than a one to many broadcast See above 287 about bridge nodes 334 Engagement rations 90% lurk, 9% contribute sometimes, 1% engage heavily Make participation easy, and a side effect of site usage. Needed to read this for the Growth Tribe academy.It s very insightful and gives a lot of examples However, there s a load of different kinds of information, therefore I have a feeling I didn t remember everything. Marc Andreesen Once Said That Markets That Don T Exist Don T Care How Smart You Are Whether You Re A Startup Founder Trying To Disrupt An Industry, Or An Intrapreneur Trying To Provoke Change From Within, Your Biggest Risk Is Building Something Nobody WantsLean Analytics Can Help By Measuring And Analyzing As You Grow, You Can Validate Whether A Problem Is Real, Find The Right Customers, And Decide What To Build, How To Monetize It, And How To Spread The Word Focusing On The One Metric That Matters To Your Business Right Now Gives You The Focus You Need To Move Ahead And The Discipline To Know When To Change CourseWritten By Alistair Croll Coradiant, CloudOps, Startupfest And Ben Yoskovitz Year One Labs, GoInstant , The Book Lays Out Practical, Proven Steps To Take Your Startup From Initial Idea To Product Market Fit And Beyond Packed With Over Case Studies, And Based On A Year Of Interviews With Over A Hundred Founders And Investors, The Book Is An Invaluable, Practical Guide For Lean Startup Practitioners Everywhere Lean startups core concept is build measure learn Learn analytics focuses on measure stage.The authors believes that there are six business model e commerce, SaaS, mobile apps, media site, user generated content, and marketplace They describe these models in details and introduce the important metrics for each one I think, in these business models media site is simplest and marketplace is most complex model.After introducing business models book introduces Lean analytics stage that include 1 Empathy 2.Stickiness 3.Virality 4.Revenue 5.Scale For each stage, there are certain rules and tasks that must be implemented There are many good case studies, advices and guidelines in the book and it is useful for entrepreneurs, web developers, and data scientist Also, I recommend this book to anyone who directs the business But I believe this quote In theory, theory and practice are the same In practice, they are not Anonymous Amazing Period Full of meat , contentful and actionable from the next business day.If you re into start ups or at least anything connected to cutting edge digital marketing you need to read it It s one of these books like Lean Start up , Rework , Business Models Generation and Startup Manual It s a must read and I m surprised that it took me so long to get to it.Even though I knew about 90% metrics, it was still useful to put them into the right context Especially, adding the case studies.The book is not the one you would read before going to bed You need to be focused to get 100% value of it Initially, I had it borrowed from the library, but decided to buy my own copy to be able to make my own notes 5 5 Instincts are experiments Data is proof it is a good summary of Lean Analytics, the book about data driven decision making The book describes 2 areas stages of a startup development and a couple of scalable business models Each part is full of case studies and business insights If you already read Lean Startup or The Startup Owner s Manual probably the first part would be less interesting The abundance of examples makes Lean Analitycs helpful for each job which requires entrepreneurial gene because the book shows how decisions should be made. Pretty dry at times, but definitely worth the read for any project manager, product manager, scrum master, or product owner The insights here when put to good use are probably worth their weight in gold. The book floats between principles and recipes for lean Analytics very well It provides nicer use cases, segmentations of business model and examples On the other hand, the examples could be concrete , showing in a didactic manner how the data was used and the decision was made.