Anonymous Analytics Club or Dow Johns Analytics

Hi, when Julia came to our company, we realized that in order to optimize everything - speed, cost, efficiency, and experiments - it seems to be time to return to the roots: the GA code, which is buried on our sites and landings.

Below we have collected vivid illustrations of how Google analytics is underused, and the types are written off from real, but all of them who wished to remain anonymous projects;)
')

Uff, it’s good that I don’t need it yet - the project already exists.

And in general - there is plenty of work, there is no time to climb into it

Let's recognize that there is such a type among marketers, and among techies, and among leaders and project owners, PR people, copywriters and other great people who are afraid of GA. Because it is difficult. Not always clear. Who to ask?

Often they skilfully bypass the question of analysts with a side - remember the Dannig-Kruger effect, when the most ignorant are most of all convinced of the correctness of their own and only their opinion.

To understand what is happening on the site or in the application - this is not a clock. With skills and knowledge, this is 5 minutes. And this skill is suitable not only for beginners, but also for experienced ones.

I am the creator of the content project, but I do not analyze the content in GA

And why - we all and our friends like wildly

In general, texts and pictures are the most difficult moment from the point of view of analysis.

Well, when there is a developed artistic taste. Bets on “audience loves fan and cats” or a compound editorial view is also accepted.

It is good that services like “Glavred” appear (although it is sharpened for just one style) for analyzing texts; it is good that there are numerous services of a / b and multivariate testing - however, most of them cost some money.

The first results for reflection can be obtained in the account GA.
From grouping by content type and comparing group scores, to evaluating virality, it is interesting to follow this with the _trackSocial function - analysis of mailings and bold a / b experiments with headers and buttons.

By the way, they say, it will be possible to do multivariate testing too soon.

I made a mobile application. Waiting for a miracle

I wrote to friends - that's the core of the audience!

Great, but the name is legion. Now wait for instals, returns and fame to happen on their own.

The first thing you can do while you wait: agree and remarketing, if the application is the first; if not the first, use your base. And start tracking channels and revenues from installations through GA - as a useful addition to specialized tools.

Plus, try not to forget about the limited possibilities.

Speech about your users. A simple example, although others are far from straightforward: when there are about 18,000 unique models of gadgets on Android, it is worth looking closely at the reports by audience. To understand the technical “stuffing” of key user groups and adapt to them.

New gadgets, which you expected, can not afford everything.

We are all responsible for analytics

And in general - we are in the evenings together waiting for the statistics update for the day.

Often, several people have access to an account in a company. To make everything happen, as in collective correspondence, do not assign anyone specifically responsible and do not train anyone (in order to better achieve the effect of entropy, it is better that this person is NOT a marketing specialist at all).

This is the first thing we did in #tceh - we assigned a person in charge and began to train slowly (although the others in GA also drop in). At the same time, it helped to unload tasks and quickly find answers to many business questions.

I am the creator of the e-commerce project, but I do not consider the effectiveness of my people and the performance of competitors

What for? If they start to bypass me at the turn - I will make them an offer that they cannot refuse!

You can closely follow every bit of information that a competitor will write on a blog, social networks, or a presentation. Even worry when an interview with him comes out on that same site. If nobody has yet told you about benchmarking in Google analytics.

You can watch Dow Jones - style graphics - though, those who need selling blood from the nose are needed, usually go further than the first screens.

Goals, offline conversions, events, segmentation based on reports by audience - google analytics are already sharpened and perekatocha under e-commerce.

And well, if you use it. If so, then super reports are a relatively new interesting feature, so I would advise her to pay attention.

It will help to collect disparate information under your KPI - and it is better to understand how the reports of departments or contractors (if they exist) coincide with reality.

I made a landing page. I drive traffic and wait for a miracle. So what?

You read the previous paragraphs, right?

And I'm waiting for the developer to be released.

I can not distract him on trifles.

Layfkhak: to netechnary today you can add the lion's majority of the necessary codes (the same services of optimization, personalization, and at least confirmation of the link of the site and Youtube, so that you can put external links from videos) with your own hands.

Via tag manager. Without a headache, knowledge of administration or programming. And without SMS.

At the same time, from the tag manager, you can customize and rebuild the main Google Analytics codes, originally located on the site.

I don't know who all these people are

And to hell with them - the main thing is to go!


Multichannel sequences and a bunch of User ID sounds scary. Scary interesting :)

While large corporations can spend a year and 100 million to eventually enter the market “with a new handy spoon, which made it easier to pour powder into a washing machine,” a small company will have to spin like a snake.

To understand customers thoroughly need. If you use customer development approaches, you can get users to interview from Google Analytics - while clearly understanding what kind of segment we want.

This is more effective than throwing a call and questionnaires for social networks. And N times more effectively than in everything to make a start from opinions of those who appeared at hand.

I do not know what to choose - mobile site or application

The dilemma that we recently presented to the employee.
You guessed right, the answer - again - lies in Google analytics, see:

1. We take the current picture of the site or landing page, do the division into categories desktop / tablet / mobile.

2. Compare the percentage of hits, failures and conversions for each category, check for possible obvious problems.

3. We make a breakdown by channel groups.

4. We analyze the ratio of visits on paid and free channels, conversions, refusals and time on the site.

5. We consider the cost of attraction plus the final cost - how much we are willing to spend without loss, so that the client can bring us the desired money.

6. We consider the resources for scaling such a group of customers - are you ready to invest in growing the number of paying users from this category?

7. If you are ready (and if not, go back to the previous steps and look for it), then we pull out “live” representatives of the target group from the analytics — modern GA tools allow us to do this — and interview them about needs.

8. Understanding the use cases of the paying audience, which we want to scale through the mobile channel, carefully we make a bet in the race “application vs adapted site”.

Analytics - like a swiss knife.

And to solve most problems at once ...

We have invited to #tceh Alexander Kulachikov, an expert of Google’s Russian Google analytics, to conduct a course where all the new products, complex functions and solutions will be considered comprehensively.

What do we do?
To study problems in a complex and, in fact, to take and customize what our projects need - with the support of a teacher.

We will study in the evenings, with breaks between the days of classes - so that we can evaluate the results of previous experiments and the effectiveness of changes.

We start on February 18. And until the 10th, you can not only calmly stake out a ticket, but also use the option of payment by installments - that is, the second part will be paid in March.



Tell us in the comments, what's stopping YOU from sorting out Google analytics?