Google Analytics Is Free for Everyone Now. No More Excuses.

Something changed quietly in mid-August: Google removed the waiting list from Google Analytics. Anyone can now create an account. No invitation, no fee, no cap that matters for most sites.

Google had rationed access since buying Urchin last year, because demand swamped them at launch. The rationing is over. Serious web measurement is now free, and the vendors charging four figures a year for it should be updating their pitch decks.

We have been installing it for clients since the invitation era. So consider this both a welcome and a field guide, because a free tool does not make the measurement automatic.

Setup takes ten minutes, with one caveat

The mechanics: create the account, paste a small JavaScript snippet into every page, wait a day for data. On any decent CMS, this is a ten-minute job.

The caveat: JavaScript tracking counts people with JavaScript turned on. It will not see search crawlers and bots, which is actually a feature, since your old log files counted them as visitors. Your new numbers will usually be lower than the old reports. They are also closer to the truth about actual humans. Expect the question from whoever reads the old reports, and have this paragraph ready.

Goals: the difference between counting and measuring

Out of the box, the tool counts visits and pageviews. That is activity. It earns its keep when you define goals: the completions that mean your site did its job. A donation thank-you page. A request-information form. A newsletter confirmation.

Define up to four, and every report reorients around outcomes. Which sources send visitors who convert. Which pages come before success. Where the form loses people.

This is a business conversation, not a technical one. What does this site exist to produce? We have sat in rooms where that question, asked so a tool could be configured, surfaced disagreements the organization did not know it had. Useful tool.

Three reports worth a monthly habit

• Referring sources, with goals attached. Knowing which sites send traffic is trivia. Knowing which send traffic that converts is strategy. One client’s prized directory listing sent hundreds of visits and zero completions. An unglamorous industry newsletter sent forty visits and six inquiries. Budgets moved.

• Top content, read as demand. Your most-viewed pages are your audience voting on what they came for. When the vote contradicts the homepage’s priorities, and it usually does, that is free research.

• The funnel on your key goal. Where people abandon your form or checkout, step by step. Every leak is a fixable page.

Week-one configuration

A few choices separate trustworthy data from noise, and they are painless now and annoying to retrofit later.

Filter out your own traffic. Your staff reloading the homepage forty times a day is not audience behavior. Track the actions the tool cannot see on its own, like PDF downloads and clicks to a donation platform on a vendor’s domain. Each takes a small addition to the link. And tag your campaign links, especially email, so visits announce where they came from.

Two more habits worth starting now. First, keep a simple changelog: when a campaign launches or a redesign ships, write down the date. Traffic charts without context become archaeology within a quarter. Second, create a backup profile with no filters at all. Filters throw data away permanently, and every analytics veteran has a story about an overzealous filter silently discarding a month of truth.

The privacy conversation

One thing deserves adult attention before the code goes in: this data lives with Google. For most organizations that is an acceptable trade for an excellent free tool. But acceptable should be a decision, not a default.

Say honestly in your privacy policy that you use third-party analytics. Sites serving sensitive audiences, like health information, should think a step harder about what page URLs themselves reveal. Free tools are paid for somewhere. Know the terms and choose knowingly.

What free measurement changes

Here is the part that outlasts the product news. Until now, ‘we don’t really know what’s working’ was defensible, because measurement cost money. That defense expired this month. When the tool is free and setup is an afternoon, not measuring becomes a choice, and choices get questioned in planning meetings.

Our recommendation for the fall: install now, define goals honestly, and let a few months of data pile up. Then walk into 2007 planning with evidence instead of recollection. Does the expensive directory listing send anyone who inquires? Is the homepage carousel, the one four departments fought over, clicked by anyone at all? What share of donation-page visitors complete the gift?

Each answer retires an argument that has been running on anecdote. Some answers will be unwelcome. Those are the valuable ones. One discipline keeps it honest: decide what question a report should answer before you open it. Without a question, data is just a hobby.