How I Decide Where The Next $100,000 Of Marketing Budget Should Go

After a good quarter, the question I hear most from CEOs is some version of “We have another $100,000. Where should it go?”

Most of them already have an answer. It’s usually more of whatever worked last quarter, which tends to mean paid search, because paid search has the cleanest dashboard.

Sometimes that’s the right call. More often the money lands on the part of the business that’s already working, and the thing holding growth back goes unfunded because it doesn’t belong to any one channel.

So, I run the same process every time. It takes two to three weeks, and the first step has nothing to do with channels. To show how it works, I’ll use a hypothetical company built from patterns I see across B2B SaaS clients. The company is made up. The process is the one I run with real ones.

Meet Fieldline

Fieldline is a Series A B2B SaaS company with about $8 million in ARR. It sells scheduling software to home services businesses with 20 to 200 employees. Average contract value is $24,000 a year.

Marketing spends $90,000 a month: $45,000 on paid search, $20,000 on content and SEO, $10,000 on events, and $15,000 on tools and contractors.

The board just approved an extra $100,000 for next quarter. The VP of Marketing wants $60,000 for paid search and $40,000 for content. Here’s everything else that could compete for the same dollars:

• More paid search.
• More content.
• Conversion rate optimization on the site and demo flow.
• Product work on onboarding.
• Customer research.
• Retention and lifecycle marketing.
• AI search visibility.

Seven options, one budget. The only fair way to compare them is to ask which one moves the stage of the business that’s limiting growth right now.

Step 1: Map The Whole Funnel, Including After The Sale

I start with one spreadsheet row per stage, using the last 90 days of data. I don’t stop at closed-won. Here’s a typical Fieldline month:

Two numbers jump out. A 20% win rate is low for a $24,000 product with a sales-led motion. And losing 22% of customers in year one means nearly $1.8 million in annual revenue walking out the back door, most of it in the first 90 days.

Pull that retention number from your billing system, not your CRM. They rarely agree.

Neither number was on Fieldline’s marketing dashboard. That showed traffic up 14% and cost per demo flat, which is exactly why paid search looked like the obvious bet.

Step 2: Price Each Option Against The Constraint

For every option, I estimate what it would change and what that change is worth over a year. Here’s the rough math for Fieldline.

Paid Search

Non-brand clicks in this category cost around $18, so an extra $60,000 buys about 3,300 clicks. Run those through the funnel above and you get roughly 30 demo requests, 15 accepted leads, seven or eight opportunities, and one or two deals. Call it $36,000 in new ARR for $60,000 of spend. That also assumes the next 3,300 clicks convert as well as the first ones, which they rarely do once you push past the efficient keywords.

Conversion Rate Optimization

Moving the demo request rate from 0.9% to 1.1% on existing traffic adds about 80 demo requests a month. Through the same funnel, that’s four more deals a month, or roughly $1.1 million in new ARR over a year, for about $15,000 of testing and design work. That’s the best ratio on the list, if the tests hit. A 20% lift in conversion rate is a real outcome, but it’s not a given, so it gets funded at cost and judged on results.

Retention

Cutting first-year churn from 22% to 16% keeps about $480,000 a year. It takes product and lifecycle work, so it shows up slower, but it compounds.

AI Search Visibility

Say Fieldline’s five main buyer prompts, run across ChatGPT, Gemini, Perplexity, and Google’s AI Mode, return Fieldline in five of 20 answers. Its two closest competitors show up in 15 and 12. That shows up in the win rate, not in traffic. A buyer who builds a shortlist in ChatGPT and doesn’t see you on it walks into your demo treating you as the alternative, and that’s a hard place to win from. At Fieldline’s volume, every point of win rate is worth about $260,000 a year.

More Content

Fieldline already publishes eight posts a month. Another $40,000 of volume is unlikely to touch win rate or churn.

Customer Research

The cheapest option on the list, and the one that tells you which of the other estimates to believe.

Put the numbers side-by-side, and the opportunity cost is hard to ignore. Every dollar that goes into the next 3,300 clicks is a dollar that didn’t go toward a retention fix worth $480,000 a year, or a win rate problem worth more than that.

Step 3: Spend The First $10,000 Finding Out

Before the big money moves, I spend about 10% of it on research. For Fieldline, that looks like:

• Fifteen interviews with customers who signed in the last six months.
• Ten interviews with customers who churned in their first 90 days.
• A review of 30 recorded sales calls in Gong, tagged for where the buyer first heard of Fieldline and who else they evaluated.
• An AI assistant option on the “How did you hear about us?” field of the demo form. I walked through that setup in my article on the AI search metrics I track.

Twenty-five interviews with a $150 incentive each, plus a researcher’s time to run and synthesize them, comes in around $10,000.

In the Fieldline version, the research says two things. Churned customers mostly left before their technicians ever used the mobile app in the field. Onboarding taught office managers the scheduler and never got crews set up. And won customers brought up ChatGPT without being asked. Nine of 15 said they used an assistant to build their shortlist, and the deals Fieldline lost went to the two competitors that keep showing up in those answers.

That turns a seven-way guess into two constraints with evidence behind them.

Step 4: Fund The Constraint, Stage The Money, Set Kill Lines

Here’s where Fieldline’s $100,000 goes:

Paid search gets nothing extra, and it doesn’t get cut either. It’s still the best read on buying intent, and it’s still producing deals at a cost the business can live with. The next dollar there is worth less than the next dollar somewhere else. The base budget is a different question, and I covered it in how to restructure your marketing team and budget for AI search.

The reserve is there because I’ll be wrong about something. Holding 15% back until day 45 means the money follows the evidence instead of my forecast. The order is the same one I use in my 90-day growth audit and the experimentation framework I wrote about earlier this year: measure, then test, then scale.

Each line also gets one number that decides whether it earns more money. If technician activation hasn’t moved by day 45, that $30,000 goes back into the pool. No line gets renewed just because it was in last quarter’s plan.

The onboarding line isn’t marketing, at least not on the org chart. It sits with product and customer success. That’s normal. The next $100,000 of growth often belongs to a team that doesn’t own the marketing budget, which is why someone has to look across the whole funnel before the money gets split.

How To Run This At Your Company

You need a spreadsheet and a couple of honest conversations with finance and sales.

1. Build the funnel from first visit through month-12 retention, using real numbers from your CRM and billing system.
2. Mark the stage that converts worst against what you’d expect for your price point and sales motion. If two stages are close, pick the one nearer to revenue.
3. For each spending option, write what it changes, what that’s worth per year, and when you’d see a signal.
4. Spend about 10% on research before you commit the rest.
5. Fund the constraint, hold back 15%, and give each line one kill metric.

If the constraint turns out to be the shortlist stage, here’s how I’d build an AI search strategy today. If it turns out to be a hire rather than a program, here’s the order I’d hire a growth team in.

The step most teams skip is the first one, specifically everything after closed-won. Marketing owns the top of the spreadsheet, finance owns the bottom, and nobody reads both on the same page. That gap is usually where the next $100,000 belongs.

Back to Fieldline’s VP of Marketing. Their read on paid search was right. It was working. It just wasn’t the thing holding the company back.

Before your next budget meeting, put retention on the same page as traffic and see which number you’d fund first.

More Resources:

• How To Get The Perfect Budget Mix For SEO And PPC
• The Business Case For SEO Is Changing; Budget Justifications Should Too
• How Much Of Your Paid Media Budget Should Be Allocated To Upper Funnel?

Featured Image: mayam_studio/Shutterstock