Advertise My Web Site

I always start with a single platform. If the website sells something people search for – a product, a service, a specific solution – I choose Google Ads. Search intent is already there; your job is to match it with a tight landing page. For a brand new site with no brand recognition, Facebook or Instagram can work better if you are targeting a demographic or an interest, but expect higher friction and more testing.

Here is the exact path I followed for a client selling leather notebooks. We set a budget of £100 on Google Ads, split across five ad groups. Each ad group had three headline variations: one mentioning price, one mentioning quality, one mentioning free shipping. I used a [SEO vs PPC](/seo-vs-ppc/) principle: the search terms that converted on paid ads became topics for organic content later.

After four days, one ad had a 4% click-through rate and a cost-per-acquisition of £2.50. I scaled that ad to £50 per day while pausing the rest. Once the campaign had two weeks of data, I pulled the search term report. The long-tail keywords with decent conversion rates – “embossed leather notebook”, “A5 leather journal” – went straight into my content strategy. Those pages now rank organically and the paid spend dropped to a maintenance level of £10 a day.

The principle is: test small, measure conversions, then let the data inform both paid and organic. If you go straight to Facebook with £500 and a broad audience, you will burn cash without knowing what works. [PPC and SEO Working Together](/ppc-and-seo-working-together/) is the fastest way to turn ad spend into long-term asset growth.

A counter-argument I have heard: “Why not run both Google and Facebook at the same time to compare?” Because you will split your attention, your budget, and your learning. One platform at a time, three ad variations per ad group, a single conversion action. That is the path that has never failed me.

Do not measure by clicks alone. I managed a campaign that delivered 500 clicks at £0.30 each – looks cheap. Zero conversions. The click-through rate was a lie because the ad promised a free guide but the landing page was the homepage. Conversion tracking was set up but the goal was a page visit, not a guide download. Clicks are vanity; cost-per-acquisition is the only number that pays your bills.

Do not send traffic to your homepage. The ad says “20% off running shoes today”. If the landing page shows your entire catalogue, the visitor lands and has to hunt for the offer. They leave. I now build a separate landing page for every ad group. It costs an hour of development time and saves weeks of wasted spend.

Do not launch without tracking. I once spent £300 on Facebook ads with zero conversion tracking because I skipped the pixel setup. I had no idea whether the 2,000 clicks came from bots or buyers. [PPC Reporting](/ppc-reporting/) is not optional; it is the feedback loop that tells you what to kill and what to scale.

Do not spend big upfront. A test budget of £50–£100 is enough to see signal. The trap is thinking you need £500 for statistical significance. False. You need five to ten conversions in a single ad group. If the CPA is £20, you need £200, not £500. If the CPA is £50, your product price probably cannot support paid advertising at all. That is the hard truth.

Watch out for ad fatigue. Running the same creative for three weeks on Facebook or Instagram causes cost-per-click to rise 20–30%. Refresh creative every two weeks. On Google Ads, add negative keywords weekly to stop irrelevant searches from draining your budget.

Finally, do not ignore organic. Paid ads give you speed; SEO gives you margin. If you spend £500 and get sales, that is good. If you use the data to rank those keywords organically, that is better. Read [How SEO and Adwords Work Together](/how-seo-and-adwords-work-together/) to see how to align the two.

I made a classic mistake with a B2B SaaS product. The client sold project management software for small agencies. I set up a Facebook campaign targeting “small business owners” as an interest. The audience size was 12 million people. I thought I was being broad enough to find customers. I was being lazy.

We spent £300 in two weeks and generated three leads. Each lead cost £100. The trial conversion rate from those leads was zero. I had ignored the most important rule: audience targeting is a precision tool, not a broad net. Facebook’s algorithm cannot read your intent; it shows your ad to people who matched “small business owner” in their profile but were freelance graphic designers, sole traders, or retired consultants. None needed a multi-user project management tool.

What I should have done: use LinkedIn Ads with job title targeting – “operations manager”, “agency owner”, “studio manager” – and pair it with a lead magnet like a workflow checklist. Or, better yet, start with Google Ads on search terms like “project management software for agencies”. The intent would have been crystal clear.

I also failed to use the [SEO and PPC Integration](/seo-and-ppc-integration/) tactic. The search terms I could have captured from Google Ads would have revealed the exact language buyers used. Instead, I made up audience segments. I now run a small Google Ads experiment for every new client before spending anything on social display. It gives me a keyword list I trust.

The lesson: do not guess your audience. Let search intent tell you who is looking and what they want. Every penny spent on a guess is a penny you cannot get back.