SaaS advertising strategy

Two SaaS companies run near-identical Google Ads accounts. Same $40,000 monthly spend, same 3.2 percent click-through rate, same $310 cost per lead, same tidy account structure. One of those accounts is the best-performing thing in the business and the other should be shut off this afternoon. The difference is that one company sells a $2,400 annual subscription and the other sells a $140,000 platform deal.

That is the whole argument. A paid media plan is not a plan until you know the contract value and the motion it has to serve, so what follows is four separate playbooks rather than one flexible one. Find your band, run that playbook, and ignore the other three.

Why the same ad account gets graded pass or fail on ACV alone

Because allowable CAC is an arithmetic output, not a preference. Take annual contract value, multiply by gross margin, multiply by the payback period in months you can actually finance, divide by twelve. That is the most you can spend to acquire a customer, and every channel with a click price above what that number supports is off the table before you open Campaign Manager.

At $2,400 ACV with 80 percent margin and a twelve-month payback ceiling, you can spend $1,920 per customer. If your paid-traffic-to-paying-customer rate is 1.5 percent, your maximum cost per click is about $29 and your maximum cost per lead is a few hundred dollars. At $140,000 ACV with the same margin and a twenty-month ceiling, allowable CAC is over $180,000, and suddenly a $700 cost per lead on LinkedIn looks conservative.

Notice what changes across those rows. Not the channels, though those change too. The conversion event changes, the acceptable waiting period changes, and the evidence you are allowed to call proof changes. A team running the enterprise playbook and reporting weekly cost per lead is measuring itself with the wrong instrument, and a team running the self-serve playbook and asking for two quarters before judgement is stalling.

Playbook one: under $5,000 ACV, self-serve

Buy intent that is already in motion, optimise to revenue events inside the ad platform, and stay away from anything sold on a rate card. Speed of feedback is your advantage here, so use it.

Channel mix at a $15,000 monthly budget: Google Search 50 percent, Microsoft Ads 8 percent, Reddit 15 percent, Meta 17 percent, retargeting 10 percent. The search half concentrates on competitor alternatives queries and problem-shaped queries, because category head terms at $18 a click rarely clear the math at this price point. Reddit ads for SaaS deserve a real slot rather than a token one, since sub-$3 clicks from a subreddit full of your exact user is the cheapest qualified traffic available to a self-serve product.

The conversion event is first payment where volume allows, activated trial where it does not. Import it properly rather than firing a page-view tag on the thank-you screen, because smart bidding is only as good as the event you feed it. The offer is the product: free trial, free tier, instant access. Gated PDFs at this ACV attract people who collect PDFs.

Avoid entirely: LinkedIn, programmatic ABM, intent data vendors, trade publication sponsorships, anything with an annual contract. Payback ceiling is twelve months and honestly six is safer, because a self-serve product with monthly churn between 3 and 6 percent does not have a twenty-four month customer to pay back against.

Playbook two: $5,000 to $25,000 ACV, PLG-assisted

This is the band where paid works best and gets run worst. You have enough contract value to afford real click prices and enough self-serve mechanics to get fast feedback, which is a rare combination.

Mix at $40,000 a month: Google Search 40 percent, LinkedIn 20 percent, review marketplaces 12 percent, Reddit and Meta 13 percent, YouTube and Demand Gen 10 percent, Microsoft Ads 5 percent. The reason LinkedIn earns a fifth of the budget here and none in the band below is that a $14,000 contract can absorb a $450 cost per qualified lead. The reason it does not earn more is that search is still cheaper per accepted lead in almost every category, which the side-by-side in Google Ads vs LinkedIn Ads for B2B SaaS works through with real numbers.

Bid on activated signup, defined as a work-email signup that completed the one action correlated with retention in your product. For Figma that was a second file. For a data tool it is usually a connected source. Pick yours from your own cohort data, not from a blog post.

The offer splits by intent. High-intent search gets a trial or a demo choice on the same page. Cold social gets a tool, a calculator or a benchmark, never a whitepaper. Payback ceiling is eighteen months, measured on signup-to-paid cohorts by source rather than on platform-reported conversions, which will double-count enthusiastically if you let them.

Avoid: named-account ABM platforms, enterprise field events, and any agency proposing a six-month brand awareness phase. You cannot afford the measurement lag and you do not need it.

Playbook three: $25,000 to $100,000 ACV, inside sales

Paid stops being an acquisition channel and becomes a pipeline channel. The person who clicks is now one of four to six people who have to agree, and the job of the ad account shifts from producing signups to producing meetings that a rep is glad to take.

Mix at $80,000 a month: Google Search 32 percent, LinkedIn 28 percent, review marketplaces 12 percent, retargeting and Demand Gen 12 percent, newsletter and podcast sponsorships 10 percent, Microsoft Ads 6 percent. LinkedIn Ads for SaaS finally earns its cost here, because you can target the exact seniority and function that signs, and because a $600 cost per lead against a $45,000 contract is a rounding error rather than a crisis.

The critical mechanic in this band is offline conversion import. Send sales-accepted leads and opportunities back into Google and LinkedIn so the bidding algorithms learn which clicks became pipeline, not which became forms. Teams that do this typically see cost per opportunity fall 20 to 40 percent over two quarters with no change in creative or targeting, which is the single highest-return piece of plumbing in SaaS paid media.

Payback ceiling is twenty months. Grade on cost per opportunity created and on pipeline coverage against the quarter’s target, not on cost per lead. If your sales cycle is 75 days, accept that a campaign launched in March cannot be judged before June, and set the review date when you launch so nobody relitigates it in week three. The allocation logic across tiers is worked through in more detail in SaaS PPC budget allocation.

Playbook four: $100,000 plus ACV, enterprise

Stop optimising to cost per lead. At this contract value your total addressable buyer list is often under 3,000 accounts, and the goal of media is to be present and credible inside those accounts while sales does the actual work.

Mix at $150,000 a month: LinkedIn account-list campaigns 35 percent, programmatic ABM 20 percent, industry publication and event sponsorships 15 percent, Google Search on brand and competitor terms only 12 percent, YouTube 10 percent, retargeting 8 percent. Note how small search gets. Enterprise buyers do search, but they search your name after a colleague mentioned you, which is why brand defence stays and category head terms mostly go.

There is no reliable conversion event to bid on. Twelve form fills a month cannot train an algorithm, so run reach and frequency buys against the account list, cap frequency at four to six a week, and resist every platform recommendation to broaden targeting. The offer is a named-executive roundtable, a benchmark report your buyer’s board would find useful, or an analyst-style teardown of their category.

Payback ceiling is thirty months and the review cadence is quarterly. Expect a two-quarter lag before anything looks like proof.

Why PLG paid economics break above roughly $25,000 ACV

Because the buyer stops being one person. That is the entire mechanism, and it shows up in the numbers as a self-serve conversion rate collapse rather than as rising ad costs.

Below $25,000, one practitioner can sign up, get value, and put the subscription on a corporate card or a manager’s approval. Above it, three things enter the process: a security review, a procurement team with a standard contract, and a budget owner who was not the person who clicked your ad. Your paid traffic-to-paid-customer rate does not decline gently across that boundary. It falls off, often by half, because the path the ad promised no longer exists.

The data on motion is more encouraging than that makes it sound. Aggregated PLG benchmark reporting through 2025 put median annual growth for PLG companies near 35 percent against 26 percent for non-PLG, with PLG companies spending roughly 39 percent less on sales and marketing to get comparable growth. The interesting number is the third one: about 67 percent of hybrid companies running PLG alongside a sales motion hit their net revenue retention targets, against 58 percent of pure-PLG companies. Hybrid wins on retention, which is exactly what you would expect when someone is paid to notice that a team of 40 users has no executive sponsor.

The practical read for an advertising plan: as you cross $25,000, keep the self-serve entry point running for the practitioner, and run a second, parallel campaign set aimed at the budget owner with a completely different offer. Snowflake, Datadog and Atlassian all run some version of this split, and it is two campaigns because it is two jobs.

The four failure modes that show up in every band

They repeat with enough regularity that you can diagnose most struggling SaaS ad accounts from a list of four questions.

The most expensive of those is premature ABM, because it comes with a twelve-month contract and an internal champion who now needs it to work. I have watched a $28,000 ACV company sign a $72,000 platform deal, spend nine months building account lists, and produce fewer meetings than the LinkedIn campaign they paused to pay for it. The platforms are not bad products. 6sense and Demandbase do what they say. They are priced for companies with a named account list, a sales team to work it, and contracts large enough that a $6,000 monthly software line is not a strategic decision.

Channel monogamy is the quiet one. An account that gets 90 percent of its leads from Google Search looks healthy on every dashboard right up until the category gets three well-funded new entrants and CPCs go up 60 percent in a quarter. The fuller catalogue of what else exists sits in digital advertising for SaaS companies, and the point of reading it is not to run everything but to know what you would switch on if your one channel broke.

What to do if you sit between two bands

Pick the lower band and run its playbook. This is the opinionated part, and it is deliberate: the cost of running the enterprise playbook on mid-market economics is a wasted year, while the cost of running the mid-market playbook at enterprise ACV is leaving some efficiency on the table.

Two situations complicate this honestly. A company with a bimodal contract distribution, say a $6,000 self-serve tier and a $90,000 enterprise tier with nothing in between, should run playbooks one and four as separate budgets with separate owners and separate reporting. Blending them produces an average ACV that describes no actual customer. The second situation is a company moving upmarket deliberately. Run the current band’s playbook with 15 percent of the budget testing the band above, and move the whole plan only when the new band’s deals exceed 40 percent of new bookings.

What to do in the next two weeks

Calculate your allowable CAC on paper: ACV times gross margin times tolerated payback months divided by twelve. Write the number on a slide and show it to your head of sales, because the argument about lead quality usually turns out to be an argument about this number that nobody has stated.

Then check three things in the ad account. What conversion event is the bidding strategy actually targeting, and is it the one your band’s playbook specifies. What percentage of spend sits in your single largest channel. And what date you agreed to judge the current campaigns, which should match your sales cycle rather than your reporting calendar. Compare the resulting numbers against SaaS PPC benchmarks with the band attached, because an unsegmented benchmark is worse than none.

If the account is already running and already disappointing, start with the fixes rather than the strategy. Most struggling SaaS ad accounts have three or four of the problems catalogued in SaaS PPC mistakes that waste budget, and fixing those usually improves the numbers more than any reallocation. New creative for whichever band you landed in can come from the SaaS ad copy templates once the plumbing is right. The rest of the operating detail, from bid math to platform selection, sits in the SaaS PPC and paid ads hub.