Marketing strategy.

Where a strategy begins

A marketing strategy is the answer to four questions: who you sell to, how you differ, where these people will find you and by which numbers you will know that it is all working. Everything else is tactics, and tactics change far more often.

Most companies skip this stage and go straight to channels: launch advertising, start social media accounts, write articles. The result is predictable — the money is spent, and there is nothing to compare the effect against, because no one agreed in advance what counts as success.

The good news is that a strategy for a small company fits on two pages. A multi-page document with pretty diagrams usually means no one is going to read it.

Who you sell to

Start not with an imagined portrait but with the people who have already bought from you. Pull up the last thirty to fifty deals and look: who these people are, what task they were solving, what they asked before buying, why they chose you.

Then the data from the website and the ad accounts: where people come from, from which devices, which pages they view before enquiring. This refines the picture but does not replace a conversation with real customers.

Then divide the audience into segments by task, not by age. “Needs it urgently and is willing to pay extra” and “takes six months to choose and compares prices” are two different groups that need different messages, even if both are thirty-five years old.

There should be few segments: two or three. For each one, note the main objection — that is what all your marketing will be answering.

How you differ

The second block is the offer. Formulate in a single sentence what a person gets and why it is better than next door. Wordings like “high quality” and “individual approach” will not do: everyone writes them, and they carry no information.

The wording is easy to test. Substitute a competitor's name for your own. If the phrase is still true, it is useless.

To find where the free space is, you need to look at competitors: what they promise, what their websites look like, what they emphasise. The method is covered in our piece on competitor analysis.

Goals and metrics

The goal must be expressed as a number and a deadline. Not “increase sales” but “receive forty enquiries a month at a cost per enquiry of up to two thousand roubles by the end of the quarter”.

Metrics are chosen to fit the goal, and these are what you will use to track progress. For most companies five are enough: number of enquiries, cost per enquiry, share of enquiries that reach a deal, average order value, return on investment. The rest are details that are convenient to hide behind when the main thing is not working out.

It is worth separately agreeing what counts as an enquiry. If marketing counts every contact and sales counts only qualified ones, an argument about results is inevitable. How to bring this to a single scale was covered in our article on linking marketing and sales, and the way to calculate returns in our piece on ROI.

What the work consists of

Search engine optimisation. Delivers traffic that does not stop when payment stops, but takes time — the first results come months later. It starts with selecting the queries you are actually able to compete for; how to do that is described in a separate article. For a company with an address and a showroom, local search results and map listings are separately important.

Content. Answers the questions people ask before buying and removes objections before they even speak to a manager. It works only with regularity, so a publishing plan is needed. Details are in our piece on content strategy.

Social media. Worth running where your audience actually is, not everywhere at once. One platform run regularly is more useful than five abandoned ones.

Email campaigns. The cheapest way to bring back those who have already shown interest. Requires a collected database and segmentation.

Paid advertising. Delivers traffic immediately and is well suited to testing hypotheses: before investing in long-term work, you can find out in two weeks whether the audience responds to the offer.

Analytics. Not a channel but a precondition for the others to work. Without the “source — enquiry — deal” link, you will not know which of the above pays off.

Launch order

There is no need to switch everything on at once. A sensible order is as follows.

• Set up analytics and agree on the metrics. Otherwise the very first results will be impossible to interpret.
• Get the traffic destination in order — the website or a page. If there is nowhere to leave an enquiry or the form does not work on a phone, the rest is pointless.
• Launch one fast channel to test the offer on real people.
• In parallel, start the long-term work — search and content, which will pay off later.
• After a month, review the numbers and reallocate the budget towards what worked.

Set aside a reserve for tests — ten to fifteen per cent of the budget. Without it, the strategy turns into a plan that cannot be adjusted.

What most often goes wrong

The strategy is written and filed away. A document lives only if you come back to it once a month and check it against the numbers.

Too many channels at once. The budget is spread thin, no channel gets enough data for conclusions, and three months later it is unclear what worked.

No timeframes for the slow tools. Expecting results from search engine optimisation within two weeks ends with the work being stopped right before it would have started to pay off.

Changing course at the first bad report. Decisions should be made on accumulated data, not on weekly fluctuations.

And the most common one: marketing works, but sales do not pick up the phone. A strategy ends not at the enquiry but at the deal, and the stretch between them needs to be described as thoroughly as the advertising.

If you start from the foundation — visibility in search and a flow of organic traffic — that is the job of SEO promotion. We build it together with analytics so that the effect shows in enquiries, not only in rankings.