How much does marketing cost a small business in Israel?

The rule, and why it is only a starting point

Most companies put between 5 and 9 per cent of turnover into marketing and sales. It is a rule of thumb, not a law: some spend less, some more, and it depends heavily on the industry and the kind of company. B2B companies usually spend a little less than companies selling to consumers, because consumer marketing leans much harder on online spend. I work in B2B, and I will not give you consumer numbers.

The percentage gives a first feel to someone with nothing else to hold on to. The real budget comes out of the go-to-market strategy, not out of a percentage. If a strategy exists, we talk about it; if not, we build one. Once it exists you can estimate the components it needs and price them. The big money comes later: after the plan has been validated in small experiments and the tools are in place, at the stage where you multiply what has been proven.

And two kinds of company have to be kept apart. A company whose marketing is entirely online and a company that meets its customers face to face, at conferences and roundtables, live in two different marketing worlds with different costs. Television, radio and print are a third kind, and most B2B companies never go near it.

What costs what: the numbers in Israel

These figures are for Israel only, they move over time, and they depend on the quality of service you buy. Every one of them can come in lower or higher.

• Paid search on Google: a floor of 5,000 shekels a month in media spend. In theory you can do less, but then it is hard to tell signal from noise.
• Managing the campaigns: at least 2,000 to 2,500 shekels a month, or 20 per cent of the media spend, whichever is higher.
• Content: starts at 5,000 to 6,000 shekels a month for very little and climbs to tens of thousands once video production and serious writing come in. Good B2B content is expensive even with AI tools, and in fields like defence technology especially.
• A roundtable: 5,000 to 10,000 shekels per event.
• An SDR, the person who calls and books one-to-one meetings: 7,000 to 15,000 shekels a month, depending on the firm supplying them.
• A website: 10 to 30 thousand shekels, sometimes 50, as a one-off. Market price lists offer a basic showcase site at 5 to 15 thousand and template builders at a few hundred a year; the difference is between a site that presents and a site that runs a funnel. Maintenance: 200 to 800 shekels a month.
• Tools: a few hundred dollars a month for marketing tools. Sales tools licensed per seat cost more.
• And there is always more: a graphic designer, photos and stock video, gifts and holiday deliveries.

Order of magnitude: 12 to 25 thousand shekels a month before people, 20 to 40 thousand with an SDR, plus the website. It is flexible and it moves, and it is still the number to hold in your head at the start.

What to buy first

In marketing that centres on the web, the website is the keystone: every campaign runs through it. Paid campaigns lead to it, social campaigns use it, landing pages hang off it, all the way to a marketing-qualified lead handed to sales. Every company should have a website, but some websites are part of the structure of the strategy itself, and those are laid first.

Templates are good mainly for inspiration. A website has to be tailored to one company's strategy, audience, message and funnel. The analogy I use in meetings: the website is one more employee in the company, and it deserves a job description like any other. What is its job: bringing leads, supporting sales, recruiting? Who does it talk to, and in what tone? What does it say in the first meeting, and where does it lead the person who walked in? It has character and a way of talking, and like any employee, a detailed and exact job description is what lets it do the work. An employee asked to serve investors, candidates, products and services all at once gets confused and does nothing. That is exactly what weakens most B2B websites.

Spend the budget, do not save it

The CEO's hard decision in the first year is how fast to spend. My answer: forcefully and deliberately, not conservatively. At the start you want answers quickly, and quick answers need quick experiments in real volume, each one ending in a clear yes or no. A budget trickled out over the year never produces a signal sharp enough to decide on, and a company that moves slowly can run out of money before it reaches the answer. A healthy company gets to the point where marketing and sales spend is less than the revenue the sales bring in, and from there everything gets easier. You want to reach that point as early as you can.

A marketing budget is not carried over from one year to the next. A marketing manager who ends the year with half the budget unspent is not a good manager. The job is to use every asset at their disposal.

How you know it is working

Campaigns are not measured in the short term. At the start you mostly see noise: small numbers that say nothing. Then something starts to resonate with the audience, the signal clears and strengthens, and only then do you fine-tune, and only after that do you scale.

The experiments in the validation stage take a month or two each. In paid search you can see within two to four weeks whether a campaign is capable of producing anything, after a warm-up period, because the platform's algorithm has to learn, and with small search volumes that takes longer. A campaign shows results after three months at the least, and a marketing plan holds for six months before you change it. Switching strategy every month yields nothing. Repetition is one of the key success factors: video and social content build a following, engagement and traction only when they keep coming.

Where the money burns

The mistake I see all the time is marketing several ideas to several audiences through one channel. Shoot at the average direction of three birds on a tree and you hit none, and they all fly away. Choose a bird, and aim at it.

The second mistake is vanity posting. A company gives its channels to photos from the retreat, the event, the holiday in the office. If the audience is job candidates, that works. If you are trying to bring in customers, then beyond a residual "we have money to spend on our people", most people do not want that in their feed, and they unsubscribe. Content has to answer one question: what is in it for me? A viewer or a subscriber has to get something that meets a real need of theirs. That is the only way they come back.

What you pay a campaign manager for

That a good campaign manager brings in more than they cost is trivial. That they lower the cost per click, also trivial. What you are really buying is experiments and judgement: A/B tests, trying new things, reading the data as it comes in, isolating one change at a time, ideas for new campaigns, and ads and graphics of real quality.

And AI

Nobody has exact figures for this yet, and I want to be straight about that. My estimate is that AI infrastructure lands in the same place, 5 to 8 per cent of turnover, so with marketing and sales you are looking at 10 to 16 per cent. On one condition: that spend is justified only if it saves money and earns more than it costs. Which is why you start by augmenting people rather than replacing them: a step that does not pay off has cost you a step, not a capability.