Most small business Google Ads accounts are losing money in the first month, breaking even in the second, and finally starting to work in the third. The pattern is so common we plan around it.
The problem isn’t usually the platform. It’s the architecture. Most accounts are built as a single muddled campaign, run for a few weeks, judged too early, then abandoned or scaled before they’re ready.
Get Found, Get Chosen, Get Your Time Back
A free one-day marketing seminar in Milton Keynes · 10 November 2026
There’s a better way. We’ve used the same 90-day campaign architecture across dozens of SME accounts. It produces predictable lead flow by day 90 and a clear path to scale beyond that.
Get your free Marketing Action PlanNot sure where to start? Answer a few questions and get your plan in about a minute. No email needed.
Get my plan →This guide walks through the exact structure, the phase-by-phase rollout, and the common mistakes that quietly kill SME accounts before they get a chance to work.
Why Most SME Google Ads Programmes Fail In The First 90 Days
Three failure patterns show up again and again.
The first is a single bloated campaign trying to do everything. Brand terms, generic terms, retargeting and Display placements all mixed together with no controls. The data is unreadable. Optimisation is impossible. Budget gets wasted on the loosest-matched audience.
The second is judging too early. Most owners check ROAS at day 14, panic at the negative number, and start making changes. Those changes reset the learning. Three weeks later they’re back at day one with even worse data.
The third is scaling before the architecture is stable. Adding budget to a leaky campaign just leaks budget faster. You need better structure, not bigger spend.
The 90-day architecture below addresses all three.
The Architecture: 4 Campaign Types, 3 Phases
We use four campaign types in a layered structure. Each one serves a specific job. They get added in phases over the first 90 days, not all at once.
The four campaigns are.
Brand Search. Always-on, maximum impression share. Captures your warmest traffic at the cheapest CPCs.
Non-Brand Search by Intent Tier. Your conversion engine. Tightly grouped by buyer intent, with strict negative keyword discipline.
Retargeting. Display and YouTube ads to past visitors. Cheap, efficient, recovers leads that abandoned.
Performance Max with Controls. Added last, with brand exclusions, deliberate asset groups and weekly oversight. Brings incremental volume without cannibalising the rest.
The phasing matters. Building all four at once on a new account is the single most common cause of the first-90-days fail pattern.
Phase 1: Days 1 To 30. Foundation
The goal of phase one is to capture the warmest, highest-intent traffic at the lowest possible cost. Nothing fancy. No automation. Just clean Search campaigns earning baseline data.
Two campaigns go live in phase one. Brand Search and high-intent non-brand Search.
Brand Search Setup
One campaign. One ad group per main brand-term variant. Manual CPC or Maximise Conversions bid strategy. Maximum impression share priority.
Negative keywords for irrelevant brand searches like “jobs”, “complaints”, “login”. Sitelinks pointing to your top three landing pages. Phone extensions enabled.
Budget. Around £15 to £30 a day depending on brand search volume.
High-Intent Non-Brand Search Setup
One campaign. Three to five ad groups, each grouped by tight thematic intent. Keywords in phrase or exact match. Strict negative keyword list from day one.
Ad copy with the buyer’s exact intent in the headline. Two RSAs per ad group, each with 8 to 12 headlines and 4 descriptions. Conversion-focused landing pages.
Budget. Around £50 to £150 a day depending on category competition.
What to measure in phase one. CPC trends. Search term quality. CTR by ad group. Conversion tracking accuracy. Don’t judge ROAS yet. Don’t add new campaign types. Just collect clean data.
Phase 2: Days 31 To 60. Expand And Refine
By day 30 you have enough conversion data to refine the foundation and add the second layer. Phase two focuses on two things. Sharpening what’s working in phase one. And adding retargeting.
Refine Phase One
Pause underperforming ad groups identified by week four data. Add 10 to 30 new negative keywords from search term reports. Increase bids on top-converting search terms. Switch high-volume ad groups to Maximise Conversions if they’re now hitting consistent conversion volume.
If you’re seeing real conversion data, this is the phase to start testing single-keyword ad groups for your top 3 to 5 winners.
Add Retargeting
Two campaigns. Display retargeting for site visitors who didn’t convert in the last 30 days. YouTube retargeting for video viewers if you have YouTube content.
LIVELive every Tuesday at 2pm UK. Bring your marketing questions.
Get your Google, local SEO and AI questions answered live.
Strict frequency caps. 3 impressions per user per day maximum. Excluded audiences for converted users (you don’t want to pay to retarget customers).
Budget. Around £10 to £30 a day across both retargeting campaigns combined. Retargeting is cheap and should stay cheap.
What To Measure In Phase 2
Conversion rate improvements from refined Search campaigns. Retargeting CPA versus first-touch Search CPA. Branded conversion rate (often lifts during this phase as more buyers complete their consideration). Total spend efficiency trending the right direction.
Phase 3: Days 61 To 90. Scale With Controls
By day 60 you have a working foundation with predictable lead flow. Phase three adds the volume layer carefully. Performance Max with controls.
Performance Max Setup
One PMax campaign. Asset groups separated by service line or product category. Account-level negative keyword list including all brand terms (this blocks cannibalisation).
Display and YouTube placement exclusion list applied at account level. Standard MFA blacklists are a fine starting point.
ROAS target set 10 to 20% above your actual break-even from real margin data. Start lower if you’re optimising for lead volume rather than direct revenue.
Budget. Start at around £40 a day. Scale weekly based on incremental performance, not attributed conversions.
What To Measure In Phase 3
Incremental conversions versus phase two baseline. PMax channel-level breakdown showing where spend is actually going. Brand search impression share (should stay near 100% because you’re blocking cannibalisation). Total account efficiency trending up.
5 Disciplines That Make This Architecture Work
Discipline 1: Weekly Search Term Reports
Pull search term reports every Monday. Add 5 to 15 negative keywords per week minimum. This single habit prevents 30 to 50% of typical Google Ads waste.
Discipline 2: Strict Phase Discipline
Don’t add Phase 2 campaigns before Phase 1 has 30 days of clean data. Don’t add Phase 3 before Phase 2 has another 30 days. Most accounts fail because someone got impatient and added everything at once.
Discipline 3: Brand Protection On Every Campaign Type
Brand cannibalisation by Performance Max is the single biggest hidden cost in SME accounts. Account-level brand negatives on PMax are non-negotiable.
Discipline 4: Real Conversion Tracking
Click-only conversion tracking with proper offline conversion imports if your lead-to-sale cycle is longer than 7 days. Without this, you’re optimising for the wrong metric and the whole architecture wastes spend.
Discipline 5: Monthly Margin Calibration
Recalculate your true break-even ROAS monthly based on real margin data. Adjust your campaign-level ROAS targets accordingly. Don’t optimise to a number from three months ago.
3 Mistakes That Kill SME Google Ads Architectures
Mistake 1: Skipping Phase 1
Tempting to launch all four campaign types on day one. Don’t. The foundation phase exists because without clean baseline data, every later decision is guesswork. Phase one feels slow because it is slow on purpose.
Mistake 2: Judging ROAS Too Early
Most account-level ROAS metrics need 4 to 6 weeks before they stabilise enough to be meaningful. Reading them at day 14 and making big changes is one of the most common ways to destroy an otherwise healthy account.
Mistake 3: Trusting Smart Bidding Without Conversion Quality Controls
Smart Bidding strategies optimise for whatever you tell them is a conversion. If your conversions are unqualified leads, Smart Bidding will deliver more unqualified leads. Layer in offline conversion imports or lead-quality scoring before scaling.
Your 7-Day Architecture Audit Plan
- Day 1: Map your current account structure. Note which campaign types you have, which you don’t, and whether brand and non-brand are separated.
- Day 2: Pull your last 90 days of search term reports. Count the negatives you’ve added per week. If it’s under 5, you have a discipline gap.
- Day 3: Check your branded search impression share. If it’s below 90% and you’re running PMax, you have brand cannibalisation.
- Day 4: Audit your conversion tracking. Is it click-only? Do you have offline conversion imports if you need them? Are unqualified leads being counted as conversions?
- Day 5: Review your campaign budgets. Is anything over £30 a day with under 30 days of clean data? If yes, that campaign needs more baseline before scaling.
- Day 6: Calculate your true break-even ROAS from real margin data. Compare it to the ROAS targets currently set on your campaigns.
- Day 7: Build a phased 90-day rollout plan using the architecture above. Identify which phase you’re actually in versus where you thought you were.
That’s a complete account architecture audit. In one week. Most owners running Google Ads have never done it.
Frequently asked questions
What is the right Google Ads campaign structure for SME lead generation?
A four-campaign-type layered architecture. Brand Search always-on at maximum impression share, non-brand Search by intent tier as the conversion engine, retargeting for past visitors, and Performance Max with strict controls. Add them in phases over 90 days, never all at once.
How long should I run Google Ads before judging the results?
Around 30 days minimum for individual campaign-level decisions. Around 90 days minimum for account-level ROAS judgements. Most accounts that fail are accounts that were judged at day 14 and changed before the data was meaningful.
Why aren’t my Google Ads generating leads?
Most often it’s one of three causes. Single bloated campaign structure with no segmentation. Lack of search term discipline and missing negative keywords. Brand cannibalisation from PMax stealing credit from organic and branded search traffic.
How long does it take to see ROI from Google Ads?
Most well-structured SME accounts hit break-even at around day 45 to 60 and start producing predictable positive ROI by day 75 to 90. Earlier than that is unusual. Later than that usually means the architecture or conversion tracking has issues that need fixing first.
How do I scale Google Ads without breaking what’s working?
Increase budget by 20% per week maximum, not in big jumps. Add new campaign types only after existing ones are stable for 30 days. Measure incremental performance, not just attributed conversions. The fastest way to break a working account is to scale too aggressively.
Should I run Performance Max as my main campaign type?
No, especially not for SME lead generation. PMax works well as a volume layer added after phases one and two are stable. Running it as your main campaign without controls is the single biggest cause of waste in SME accounts.
How much budget do I need to make this architecture work?
The minimum that produces meaningful data for a 4-campaign-type architecture is around £80 to £150 a day total spend. Below that, you can run a simplified two-campaign version (Brand plus high-intent Search) and add layers later as budget allows.
The Architecture Is The Strategy
Most SME Google Ads programmes don’t fail because the platform is broken. They fail because they were built without an architecture, scaled before stability, and judged before the data was meaningful.
The 90-day plan above is unglamorous. It’s also reliable. The accounts we run on this structure produce predictable lead flow, controllable costs, and a clear path to scale.
The work isn’t complicated. It just requires discipline. Phase one for 30 days. Phase two for 30 days. Phase three for 30 days. No skipping. No scaling before stability.
By day 90 you’ll have a working machine. By day 180 you’ll be scaling it intentionally. By month 12 you’ll wonder why everyone else is still treating Google Ads as a black box.
Start today with the audit. The architecture follows.
Written by Uzair Kharawala
Partner at SF Digital Studios, a Google Premier Partner agency. Uzair has been helping businesses grow with Google Ads since 2002 and teaches it every week on YouTube.
Get our best Google Ads tips by email
New articles and videos, straight to your inbox. Unsubscribe any time.