Ad analytics.

What marketing analytics is

Analytics is the collection, processing and interpretation of data about what is happening with your promotion. Where people come from, what they do on the site, which enquiries turn into deals and how much each one cost.

The point is not the reports. The point is that decisions are made on facts rather than gut feeling. Without data, “the ads are working” is an opinion, and you can argue about it forever.

For a small company this is not about complex systems. It is enough to know what an enquiry costs in each channel and what share of enquiries make it to payment. That alone changes where the budget goes.

What it does for you

First, understanding the audience. The data shows who is coming: from which regions, on which devices, via which queries, which pages they look at. That changes the copy, the offers and the choice of platforms.

Second, evaluating campaigns. Clicks, leads, cost per enquiry, the share that reach a purchase. It becomes clear what works and what burns money for nothing.

Third, allocating budget. Once you know which channel brings cheaper enquiries, the money moves there instead of being split evenly “just in case”.

Fourth, planning. Accumulated data reveals seasonality and recurring patterns, so the next period is planned more accurately.

Fifth, finding bottlenecks in the interface. The pages people leave most often and the form steps where they abandon are found exactly this way.

Where to start

The sequence is simple, and you start not with tools but with the questions you want answered.

• Formulate the questions — what a lead costs, which channel pays off, where people leave the site.
• Choose the metrics — three to five that genuinely reflect the result, not everything at once.
• Set up collection — a tracking counter on every page, goals for key actions, tags on advertising links.
• Link it to sales — enquiries must reach your records, otherwise you see clicks but not money.
• Set a schedule — a regular short review matters more than a bulky report twice a year.

The last point is where most people fail. Analytics that is set up but never opened is exactly as useless as no analytics at all.

Which metrics to watch

Start with the metrics closest to money, not to traffic. Visits by themselves say nothing: a site can collect plenty of visits and produce not a single enquiry.

The basic set: number of enquiries, cost per enquiry, share of enquiries that became deals, average order value and the final return on investment. Everything else is auxiliary and explains why those numbers are what they are.

It helps to look at metrics by channel and over time. A single monthly figure means little, while a six-month series already shows the direction.

How to read the data

Data does not answer questions on its own. A figure shows what happened; the cause has to be put forward as a hypothesis and tested.

The most reliable way to test is a comparative experiment: two versions of a page or ad on the same traffic. It answers not “on average across the market” but specifically for your site and your audience.

Visualisation helps. A chart of the trend over six months reads in seconds, while the same information as a table of numbers usually goes unread.

Common mistakes

Poor data quality. The counter is not on every page, ad links have no tags, goals are set on a page visit rather than a form submission — and conclusions are drawn from a distorted picture. Check that collection is correct before building decisions on it.

Ignoring context. A dip may be explained by the season, holidays, a broken site or the loss of one large client, not by ad “fatigue”. Numbers are always read together with what was going on around them.

Decisions on small numbers. Five leads in one channel and three in another is not a difference, it is chance. While data is scarce, it is too early to draw conclusions.

Analytics detached from strategy. If reports live on their own and influence nothing, they are a waste of time. Every regular summary should end with a decision: what we change.

Frequently asked questions

Do I need analytics if the business is very small?

Yes, all the more so. The smaller the budget, the more it costs to spend it in the wrong place. A minimal setup takes little time and immediately shows which channel is surplus.

Which tools do I need at the start?

A free web analytics system, tags on advertising links and a simple spreadsheet where enquiries and deals are brought together are enough. Paid services make sense once the free ones stop being sufficient.

How often should I look at reports?

Advertising — weekly, the overall picture by channel — monthly, strategic conclusions — quarterly. Daily checks more often do harm: you end up reacting to noise.

What if the data contradicts experience?

First check the data collection, then the context, and only then change your mind. But if collection is correct, trust the data.

What to do next

Check one thing: can you name the cost per enquiry for each channel last month? If not, that is where to start, not with expanding the budget.

We set up data collection and run advertising by the numbers — that is advertising setup. Next, our guides on how to calculate ROI in marketing and how to run A/B tests will come in handy.