Strategy before tactics: choosing the next move that matters

Adding a channel is the easiest decision in marketing. It feels like progress, it can be started on a Monday, and nobody has to admit that the last one is not working yet.

The harder decision, and usually the more valuable one, is what to stop. This is a method for making that decision rather than an argument that you should.

Adding a channel is the easiest decision

New channels are attractive because they are unspent. Nothing has gone wrong on them yet. The existing ones carry the accumulated evidence of everything that has not worked, which makes them feel tired even when they are the ones producing the work.

It is worth noticing that pull before acting on it. The question “should we be on this?” almost always gets a yes in the abstract. The useful version is “instead of what?”

Capacity is the real constraint

Most small teams do not have a strategy problem so much as a capacity problem wearing a strategy costume. Five channels run at a quarter of the attention each will nearly always lose to two run properly.

Every channel carries a standing cost that does not appear on any invoice. Someone has to produce for it, watch it, and respond to what it produces. A social account needs replies. A newsletter needs a next issue. A paid campaign needs someone checking that spend is going somewhere sensible. That cost continues long after the initial enthusiasm has gone.

So the first question is not what to add. It is what you can actually sustain in a normal month — not a good month.

Three questions worth answering honestly

• Where did our best enquiries in the last year actually come from? Not the most enquiries. The ones that became work you wanted.
• Which channel would we notice within a fortnight if it simply stopped?
• What are we doing because it is working, and what are we doing because we started it?

The third one is uncomfortable and usually the most useful. A channel that nobody would start today, but that nobody wants to be the one to kill, is the most expensive thing on the list, because it consumes attention without anyone defending it on the merits.

Answering the first question honestly needs records rather than recollection. If you cannot tell where your best enquiries came from, that is the first thing to fix, and it is a smaller job than it sounds — usually a source field on the enquiry and the discipline to fill it in.

Score them, then cut one

Once you have the answers, sort your channels on two axes. Not a matrix on a whiteboard for its own sake — two columns on a page is enough.

• Evidence that it produces good enquiries. Not attention, not impressions, not how it feels. Enquiries you were glad to receive, in the last twelve months.
• Cost to sustain properly. Hours per month, from the person who would actually do it, in a normal week rather than an ambitious one.

Four groups fall out of that, and each has an obvious instruction:

• Good evidence, low cost. Protect it. This is the one that quietly gets neglected because it is not the problem.
• Good evidence, high cost. Keep it, and ask what would make it cheaper to run rather than whether to keep it.
• Weak evidence, low cost. Leave it alone for now. It is not what is holding you back.
• Weak evidence, high cost. This is the one to stop.

Then actually stop it, for a full quarter, rather than half-running it. Half-running is the worst of both: you keep the cost and lose the chance to find out whether it mattered. Tell whoever needs to know, write down the date you stopped and the date you will review it, and put the recovered hours somewhere specific rather than letting them evaporate.

Pick a move you can judge

A good next move has a visible outcome and a sensible time horizon. “Improve our brand” is not a move. “Rewrite the three service pages that get the most traffic, and see whether enquiry quality changes over the next quarter” is a move: you know what you did, when to look, and what would count as it having worked.

Decide the measure before you start, and write it down, because the temptation afterwards is to pick whichever number moved. In Google Analytics 4 the thing to set up is a key event — an event you have marked as particularly important to the business, such as a completed enquiry — rather than watching traffic and hoping. Note that in GA4 the word “conversion” now refers specifically to the Google Ads side of the integration, so if you are reading older instructions the terminology will not match what you see on screen.

That is where analytics and reporting does its real work: not proving that marketing works in general, but telling you whether the specific thing you changed did anything.

Sequence beats breadth

If organic search is already your strongest source of qualified enquiries, SEO probably deserves attention before a new channel does. If you have demand but the site does not convert it, the website is the constraint, and paid media will simply buy more traffic for a page that already loses people — which is a fast way to spend money confirming something you could have found out for nothing. If that might be your situation, where visitors actually drop out is the more useful place to start.

The order is the strategy. Everything else is tactics.