Data vs Noise: How to Stop Wasting Your Media Spend

There is a version of marketing that feels productive but isn’t. Google Ads is running, LinkedIn is posting, a retargeting campaign ticks over in the background, and the dashboard shows thousands of impressions and a cost per click that looks fine. Then someone asks which of those activities produced revenue last quarter, and the room goes quiet.

That is the noise problem. The spend is real, the activity is real, but the link between the two and the sales ledger is missing. Without it, budget decisions get made on gut feel, on whoever argues loudest, or on a win someone remembers from six months ago.

This guide covers where media spend usually leaks, a worked example of finding the waste, how to audit each channel, and what to measure instead. It is written for owner-managed UK businesses spending somewhere between a few thousand and a few tens of thousands of pounds a month on paid media.

Where media spend leaks

Marketing waste rarely looks like waste. It looks like activity, and each piece of it is defensible on its own. The leaks we find most often fall into three groups.

Vanity metrics

Impressions tell you how often an ad was shown. Clicks tell you how many people arrived. Neither tells you whether the business grew. If reporting stops at click-through rate and cost per click, the campaigns are being judged on the wrong numbers, and the platforms will happily optimise for them.

Too many channels at too little depth

A modest budget spread across Google Ads, Meta, LinkedIn, display retargeting and programmatic at the same time is almost always weaker than the same money concentrated on two channels run well. Each channel needs enough spend to produce data you can act on. Split it five ways and none of them gets there.

Poor campaign hygiene

Ads nobody has reviewed in weeks. Keyword lists never pruned of irrelevant searches. Landing pages never tested. These are slow bleeds rather than dramatic failures, and they compound. Google’s own guidance is that negative keywords “let you exclude search terms from your campaigns and help you focus on only the keywords that matter to your customers”.¹ Most accounts we audit have a short, stale negative list.

A worked example: finding £2,190 a month

The numbers below are a worked example, not a client’s figures. A B2B services firm spends £6,000 a month: £2,400 on non-brand Google search, £400 on brand search, £1,500 on LinkedIn, £900 on display retargeting and £800 on a Meta reach campaign. A 90-day audit of search terms, audiences and CRM records finds four leaks.

• Irrelevant search terms: 30% of non-brand search spend went on queries such as jobs, training and free templates, or £720 a month.
• Stale retargeting: 60% of the retargeting budget was reaching people who bounced more than a year ago, or £540.
• Reach with no lead goal: the Meta campaign was optimised for reach and produced no enquiries; we would keep a small test and move £480.
• Over-broad LinkedIn audience: 30% of impressions went to job functions that never buy, or £450.

That is £2,190 a month, about 37% of the budget, or £26,280 a year. None of it needed new campaigns to fix. It needed someone to read the search terms report, tighten two audiences and change one campaign objective.

How to audit your channels in 90 days

Ask one question of every active channel: can you trace a line from this activity to a qualified lead or a closed deal in the last 90 days? If the answer is no, or “probably, but we’re not sure”, the channel goes on probation. You are not cutting it. You are pausing new spend until the measurement is fixed.

1. List every channel and its monthly cost, including agency fees and tools, not just media.
2. Pull 90 days of leads from your CRM and record the source of each one. If the source field is empty for many of them, that is your first finding.
3. Ask sales to grade those leads: qualified, not qualified, or closed. Platform-reported conversions are not a substitute.
4. Read the raw data: search terms, audience breakdowns, placements and frequency. This is where the waste in the worked example came from.
5. Decide per channel: keep, fix, or pause. Consolidate onto the two or three channels where buyers have shown intent.

Fix measurement before you fix campaigns

Most of the waste above survives because the tracking cannot see it. Google describes conversions as the actions you choose to identify as valuable, “like a purchase, sign-up, or phone call”, and says they help you “learn which keywords, ads, ad groups, and campaigns are best at driving valuable customer activity”.² The catch is the word valuable. Counting every form view or PDF download as a conversion teaches the bidding to find more PDF downloaders.

Connect ads to closed deals

B2B sales rarely close on the website. Google’s offline conversion imports exist for exactly this case, where an ad “starts a customer down a path that ultimately leads to a sale in the offline world, such as at your office or over the phone”.³ Import qualified leads and closed deals from your CRM, give them values, and Smart Bidding starts optimising for revenue rather than enquiries.

Build consent in, not on

UK law shapes what you can measure. The ICO’s guidance on storage and access technologies is clear: “If your service uses storage or access technologies for the purposes of online advertising, you must get consent.”⁴ That includes ad measurement and performance. Analytics used only to improve your own website can fall under a statistical exception, but not once the data is shared to link visits to adverts.

Google’s consent mode is how the tags respect that choice. When visitors deny consent, consent-aware tags do not store cookies and instead send cookieless pings, which Google uses to model the gaps in your conversion data.⁵ Set it up properly and you keep a usable picture of performance without breaking the rules.

What to measure instead of clicks

Every platform reports on itself in its own terms, which makes channels impossible to compare. Measure all of them against the same downstream numbers.

The infrastructure behind this is unglamorous: UTM parameters applied the same way on every link, GA4 key events that reflect real outcomes, and a CRM that stores the source of every lead. None of it gets celebrated like a campaign launch, but it makes every later decision cheaper. If you want a quick read on where your own marketing stands, the marketing score calculator takes a few minutes.

What good PPC management looks like

Pay-per-click is one of the easiest channels to run badly. Results are delayed, a click today might convert in three weeks, so it is easy to keep funding a campaign that is not working before the data catches up. Four things separate a well-run account from an expensive one.

• Separate campaigns by intent: brand, non-brand search, competitor and retargeting each get their own budget, bidding and success measure. Mixing them hides where cost per acquisition is rising.
• Tight keyword control: review the search terms report weekly and keep the negative list current.
• Conversions tied to revenue: separate a genuine enquiry from a download, give values where you can, and import offline outcomes.
• Landing pages built for the search: one clear offer and one action per page, never the homepage.

This is the core of how we run performance marketing: structure and measurement first, spend second. For the targeting side, choosing who sees your ads in the first place, see our guide to targeting UK B2B buyers without wasting budget.

How to scale what works without breaking it

Scaling is not the same as spending more. Plenty of businesses have raised a Google Ads budget and watched cost per acquisition climb with it, because more budget amplifies whatever inefficiency is already there. We use three rules.

1. Earn the increase: a campaign gets more budget only after four or more weeks of stable cost per qualified lead, and sales confirming the leads are worth having.
2. Step up gradually: our rule of thumb is increases of around 20% with a review point after each, so the bidding has time to adjust. A doubling overnight behaves like a new campaign.
3. Reinvest in the system: put part of the return into things that lift every channel, such as better landing pages, fresh creative and cleaner tracking, not just the same campaign.

Good data has a way of making the right decisions obvious. The hard part is building the data.

Where to start

Pick the channel with the largest budget and run the 90-day trace on it this month. Read its raw data, not the summary dashboard, and ask sales to grade every lead it produced. You will usually find a leak within an hour.

Then fix the measurement: one conversion that reflects a real enquiry, consent mode in place, and closed deals flowing back from the CRM. Once that exists, cutting noise stops being a debate. The numbers make the case, and every pound of media spend becomes accountable to revenue. If you are planning the quarter around it, our marketing checklist for SMEs sets out the order to do things in.