The information yesterday that OpenAI acquired TBPN really demonstrates how worthwhile distribution is. Because it’s more and more extra worthwhile but in addition tougher to construct, many corporations and leaders have been inquiring extra about paid adverts. So, we teamed up Rex Gelb, Founder at Summit Chase and Head of Paid Media at Cursor, to convey you this version.
Rex spent over a decade main paid media at HubSpot. He began as a workforce of 1 managing a single Chrome extension on a $20k/month price range. By the point he left, he had constructed and led a world workforce of 20+ operators working campaigns throughout dozens of merchandise, 15+ nations, and 6 languages.
Over 12+ years, he has managed roughly $750M in spend throughout Google, Meta, LinkedIn, and different main platforms. Now, he’s main efficiency advertising at Cursor.
This information walks by the 0 to 1 playbook. When to start out, how a lot to spend, what to check first, and find out how to scale as soon as one thing lastly works. Allow us to get into it.
Gross sales at present nonetheless depends on guide work: stitching collectively “Frankenstacks” and toggling between a dozen tabs simply to log a single name.
Reevo is breaking the mildew. As a vertically built-in Income Working System, Reevo replaces the fragmented maze with a single, AI-native workspace. It captures 100% of your first-party information, from emails to assembly recordings, to assist reps discover high-intent patrons, automate administrative “rubbish,” and generate context-rich outreach that truly sounds human.
No extra losing time on information entry or disconnected instruments. It’s one unified engine for the whole income cycle. Go Stackless. Study extra at reevo.ai.
Step 1: Know When Not to Run Advertisements
The most important mistake founders make with paid media is beginning too early.
Advertisements should not the device that creates product-market match. They amplify it.
In the event you haven’t reached primary traction but, paid media often simply accelerates the educational that one thing isn’t working. Rex has seen this sample a whole lot of occasions: founders pour $30K-50K into adverts earlier than retention exists, then conclude paid media doesn’t work for them. The truth is the product was not prepared. The adverts simply made that seen quicker. With out product-market match, you might be paying to accumulate customers who churn instantly, and the platform learns to search out extra customers who appear like your churning customers. It turns into a destructive suggestions loop.
If retention is just not there but, you might be most likely too early. Paid media works finest when it pours gasoline on a hearth that already exists.
Earlier than investing in adverts, you must already see indicators that the product works:
-
Customers are sticking round. Not simply signing up, however coming again. In case your week-4 retention is under 20%, adverts will amplify a leaky bucket.
-
Clients are renewing. If clients should not staying, spending extra to accumulate new ones is a dropping method.
-
Phrase of mouth exists. Folks recommending your product with out being requested.
-
Early offers are taking place organically. Income coming in by inbound, referrals, or outreach that converts with out heavy discounting. In the event you can’t shut offers in any respect, paid media is not going to repair your gross sales movement.
Paid media works finest when it pours gasoline on a hearth that already exists.
The exception is early product testing. Some startups will run small paid campaigns to rapidly recruit beta customers or collect suggestions. In that case the purpose isn’t income – it’s studying. Small campaigns ($2K-5K) to recruit beta customers, check messaging, or validate positioning are superb. However large-scale pipeline technology ought to wait till traction exists.
Step 2: When You Have Traction, It’s Virtually All the time Value Testing
After you have product-market match and a few capital to deploy, paid media turns into a really uneven guess.
The draw back of testing is comparatively small. You spend $20K-50K over a couple of months and study what doesn’t work. The upside is a repeatable pipeline engine that compounds for years. That risk-reward profile is why most Collection A corporations ought to be testing paid.
Most B2B corporations begin testing paid media round Collection A, once they have each traction and price range.
Typical beginning ranges look one thing like this:
Seed / Pre-PMF ($0-20K/month)
Principally small exams and sign gathering. At this stage, paid media is a analysis device, not a progress engine. Rex began at this degree at HubSpot: one particular person, one product, $20K/month price range. The purpose was to determine whether or not paid may work in any respect.
Collection A ($20-50K/month, ramping to $100K+)
Aggressive groups could ramp towards $100K+ if exams work. That is the inflection level. Check 3-4 marketing campaign sorts concurrently; measure pipeline, not simply leads. Two to a few months is typical earlier than one thing works constantly. Resist the urge to give up after 4 weeks.
Collection B+ (usually $200K/month and up)
That is widespread relying on progress targets. At this stage you might be scaling what works: new channels, geos, viewers segments. Key: CPC has risen for 87% of industries (WordStream). If a channel produces ROI at present, scale aggressively now, not subsequent quarter.
These should not guidelines, they’re simply patterns.
Step 3: Anticipate It To not Work at First
The primary few months of paid media are often messy. Rex shared at HubSpot’s INBOUND convention that roughly 90% of Fb advert campaigns fail once they launch. The bar is just not “get it proper the primary time.” Slightly, it’s to “construct a system that finds what works quicker than opponents do.”
The purpose early is just not effectivity. The purpose is discovering sign.
Campaigns miss targets, messaging feels off, concentrating on is mistaken. That is regular. Keep in mind, the purpose is discovering sign and studying.
Successful advert packages come from testing mixtures throughout 5 variables. Consider it as a five-dimensional matrix. Every mixture you check narrows the house of what works.
The 5 variables to check:
-
Platforms. Which channels does your viewers truly reply on? LinkedIn may appear apparent for B2B, however Rex has seen Google competitor campaigns outperform LinkedIn for sure merchandise. Don’t assume. Check.
-
Audiences. Job titles, seniority ranges, firm sizes, industries. Rex constantly finds that the customer you assume will convert finest is just not at all times the one who truly does.
-
Inventive. Lengthy copy vs. quick copy. Video vs. static. In 2026, genuine native-style content material more and more outperforms polished manufacturing, particularly on Meta.
-
Gives. Demo request vs. free trial vs. gated content material. A demo request usually converts to pipeline at 3-5x the speed of a whitepaper obtain.
-
Touchdown pages. On LinkedIn particularly, Lead Gen Types improve completion charges dramatically by auto-filling person information.
Ultimately one thing begins to click on. You then double down.
The Core Three: Platform-by-Platform Playbook
Under is a sensible beginning framework for the three platforms most B2B corporations ought to check first. These cowl the three major shopping for modalities: skilled concentrating on (LinkedIn), energetic search intent (Google), and broad attain at scale (Meta). Most B2B corporations ought to grasp these three earlier than increasing to YouTube, Reddit, Twitter/X, or CTV.
LinkedIn Advertisements
LinkedIn is often the primary paid channel B2B groups check. It presents the cleanest concentrating on for skilled audiences. You possibly can goal by job title, seniority, firm measurement, business, and even particular corporations. No different platform comes near this degree {of professional} precision.
However the largest mistake corporations make on LinkedIn is including pointless friction. Driving site visitors to a touchdown web page sounds logical, but it surely usually kills conversion charges. The person has to go away LinkedIn, watch for the web page to load, manually fill out a kind, and submit. Every step loses individuals.
As a substitute, begin with LinkedIn Lead Gen Types. These preserve the person inside LinkedIn and auto-fill their data (title, electronic mail, firm, title) from their LinkedIn profile, which dramatically will increase completion charges.
What to check on LinkedIn:
Thought management adverts. Quick insights from founders, operators, or material specialists constantly outperform polished advertising copy. LinkedIn now helps Thought Chief Advertisements that run from private profiles somewhat than firm pages. Early information exhibits CPCs of $5-$15 with meaningfully larger engagement charges.
Doc Advertisements. These are an underused format that deserves extra consideration. You add a PDF (whitepaper, information, information report) and customers can preview it instantly of their feed with out leaving LinkedIn. The obtain interplay pre-qualifies the lead. Doc Advertisements ship the bottom common CPL on the platform: $256, in comparison with $317 for single picture adverts.
In-feed adverts. Easy posts that seem naturally within the feed. These work properly for demo requests, product bulletins, and occasion promotions.
Message or dialog adverts. Direct outreach model adverts that mimic a gross sales dialog.
Content material presents or demo requests. Check each gated content material and demo CTAs.
Additionally check a number of personas. generally the customer you assume converts finest is just not the one who truly does. Run the identical provide to 3-4 totally different persona segments and let the info let you know who your actual purchaser is. You would possibly uncover that mid-level managers convert higher than VPs, or {that a} particular business phase responds at 3x the speed of your assumed ICP.
LinkedIn success comes from systematic testing. 2026 benchmarks: Avg B2B CPC $5.50-$8.50. CTR 0.44-0.65%. B2B ROAS 4.1-8.3x. LinkedIn now captures ~39% of all B2B paid media budgets globally.
Google Advertisements
Google is commonly the highest-intent paid channel. However many corporations launch Google campaigns incorrectly. All the time begin with Search adverts, not show or YouTube.
There are three core marketing campaign sorts to check:
1. Model Campaigns
If opponents are bidding in your firm title, they’ll seize demand you created. Model campaigns shield that site visitors.
It might really feel unusual paying for clicks by yourself title. In spite of everything, your natural itemizing is true there. However in observe, competitor adverts push your natural outcome additional down the web page, and even a small proportion of clicks going to opponents represents actual pipeline leakage. Model campaigns are often very low-cost (low CPC, excessive CTR, excessive conversion fee) as a result of the person already has intent to search out you.
2. Competitor Campaigns
Competitor phrases are sometimes one of many highest ROI exams in early Google packages.
The conversion charges on competitor campaigns are typically decrease than model campaigns (they’re on the lookout for another person, in any case), however the lead high quality is commonly distinctive as a result of these are patrons deep in an energetic analysis. The secret is having a compelling cause for them to contemplate you: a transparent differentiator, a comparability web page, or a case examine from a buyer who switched.
3. Non-Model Search
These are class key phrases like “HR compliance software program,” “B2B lead technology platform,” or “advertising attribution device.”
These might be aggressive and costly, however they show you how to seize new demand. Different Google codecs like Efficiency Max, Show, and YouTube can work properly later. However for 0 to 1, Search ought to at all times come first.
Meta (Fb and Instagram)
Meta adverts behave very otherwise than LinkedIn and Google.
The platform depends (much more) closely on machine studying. The advert system learns from huge volumes of behavioral information and makes concentrating on selections that always outperform guide viewers picks. Making an attempt to micromanage campaigns or over-segment audiences often makes efficiency worse, not higher.
As a substitute, take a consolidated method. This implies working a small variety of campaigns with broad concentrating on and letting the algorithm optimize. Particularly:
-
Don’t separate Fb from Instagram. Let Meta determine the place your advert performs finest for every person. The platform has extra information about person conduct throughout its properties than you do.
-
Don’t break up cellular from desktop. Once more, the algorithm handles this higher than guide segmentation.
-
Don’t create tiny viewers slices. A marketing campaign concentrating on 5,000 individuals provides the algorithm nothing to work with. Broader concentrating on (50K-500K+) lets the machine studying system discover your patrons inside a bigger pool.
-
Keep away from utilizing lead types except you embrace robust qualification questions. Meta lead types usually produce excessive quantity however low high quality as a result of the auto-fill makes it too straightforward. Driving site visitors to your web site with a transparent conversion motion (demo request kind, free trial signup) often produces better-qualified leads.
The 2026 inventive shift: Meta’s algorithm more and more favors genuine, native-style content material over polished manufacturing. Quick-form video with CPVs as little as $0.01-$0.02 is turning into the default. A founder recording a fast tackle their telephone could outperform a two-week company manufacturing.
All the time measure pipeline contribution, not simply top-of-funnel quantity.
After the Core Three, Broaden Fastidiously
After you have information from LinkedIn, Google, and Meta, extra channels could also be price testing.
Widespread subsequent experiments embrace:
-
YouTube for video training and retargeting
-
Reddit for area of interest technical communities
-
Twitter/X for startup and developer audiences
-
CTV for model consciousness at scale
However most B2B corporations ought to give attention to the core three first.
Spreading price range throughout too many channels early slows studying. The maths: $30K/month throughout three platforms provides you $10K per platform. $30K throughout six provides you $5K every – usually under the edge for statistically significant outcomes on any of them.
The Most Essential Step: Set Up Your Monitoring First
Earlier than launching any adverts, you have to arrange correct conversion monitoring.
With out it, the platforms can’t study who to focus on. That is the part Rex feels most strongly about – it’s arguably the one largest determinant of whether or not your paid media program succeeds or fails. In case your monitoring solely captures 40% of precise conversions (blocked by advert blockers, iOS privateness adjustments, or damaged pixels), the algorithm is working with a distorted image of actuality.
Fashionable paid media depends closely on machine studying. Google’s Good Bidding, Meta’s Benefit+, and LinkedIn’s advert supply programs all optimize primarily based on the alerts you ship again to the platform. When a person clicks your advert after which converts (fills out a kind, begins a trial, books a demo), that conversion occasion tells the platform: “that is the type of particular person I need extra of.” The platform then makes use of that sign to search out extra customers who look related.
The fashionable monitoring stack:
Server-side monitoring with Conversions API (CAPI). Browser-side pixels are more and more blocked by advert blockers, iOS privateness adjustments, and cookie restrictions. Server-side monitoring bypasses all of those limitations by sending conversion information instantly out of your server to the advert platform’s server. That is now not non-compulsory. It’s the basis of efficient paid media in 2026. Google calls it Enhanced Conversions. Meta calls it Conversions API. LinkedIn calls it the Conversions API as properly. The implementation is technical, however the affect is dramatic: corporations that implement server-side monitoring usually see a 15-30% enchancment in attributed conversions, which in flip provides the algorithms considerably extra information to optimize with.
Superior Matching. This sends hashed buyer information (electronic mail addresses, telephone numbers) to the advert platforms to enhance attribution accuracy. When a person converts, the platform can match that conversion again to the precise advert click on extra reliably, even throughout units and periods. That is particularly vital for B2B the place shopping for cycles span weeks or months: the one who clicked your LinkedIn advert on their telephone two weeks in the past would possibly convert on their desktop at present. With out Superior Matching, that conversion is invisible to the platform.
Consent Mode. With GDPR, CCPA, and evolving privateness rules, you want monitoring that works inside consent frameworks. Google’s Consent Mode adjusts information assortment primarily based on person consent preferences, permitting the algorithms to nonetheless study from aggregated, anonymized alerts even when customers decline full monitoring. That is particularly vital if you’re working campaigns in Europe or concentrating on privacy-sensitive enterprise patrons.
Consider it like coaching a machine studying mannequin. The higher the info you present, the higher the outcomes. With out clear monitoring, the algorithm is actually flying blind.
Another crucial level: by 2026, platforms like Google (Efficiency Max, AI Max for Search) and Meta (Benefit+) now not provide automation as an non-compulsory function. They assume it. The AI handles bidding, viewers concentrating on, inventive meeting, and placement. The function of the human operator has shifted from managing marketing campaign settings to making sure the standard of inputs. Your monitoring information, inventive belongings, product feed, and touchdown web page construction are actually the first levers you management. The whole lot else is delegated to the machine.
Consider it like coaching a machine studying mannequin. The higher the info you present, the higher the outcomes. With out clear monitoring, the algorithm is actually flying blind.
Step 4: When One thing Works, Hit the Gasoline
Ultimately you can find a marketing campaign that produces actual pipeline. When that occurs, many corporations usually make one other mistake.
They scale too cautiously. The danger of scaling too slowly is commonly bigger than the danger of scaling too quick. CPC has risen for 87% of industries over the previous 12 months (WordStream). B2B SaaS CPCs particularly rose ~9% YoY. The window to scale a profitable channel at favorable economics is at all times narrowing.
If a channel produces constructive ROI, the very best time to speculate is now.
Advert stock hardly ever will get cheaper over time. Competitors will increase, costs rise. Successful corporations transfer rapidly as soon as a channel proves itself.
Paid media works finest while you:
-
Check aggressively
-
Determine winners
-
Scale rapidly
Velocity issues. Observe this rhythm: check aggressively in months 1-3, determine profitable mixtures, then shift to scaling mode the place 70-80% of price range goes to confirmed winners and 20-30% stays in testing to search out the following breakthrough. By no means cease testing completely.
The 2026 Shift: AI Is Rewriting the Paid Media Playbook
Another dimension price addressing, as a result of it adjustments how each founder ought to take into consideration this channel going ahead. The function of AI in paid media has shifted from “nice-to-have optimization” to “the whole working system.” Understanding this shift is crucial for anybody constructing a paid media program in 2026.
Inventive is now the first lever
When the platform controls bidding and concentrating on (which it more and more does, by default), the variable that separates a powerful marketing campaign from a weak one is the standard of what you feed the machine.
Platforms have gotten extra opaque
Google’s Efficiency Max and Meta’s Benefit+ give advertisers much less management over particular person marketing campaign settings. You can’t manually set bids for particular key phrases in Efficiency Max. You can’t management which placements your Benefit+ adverts seem on. The tradeoff is commonly higher combination efficiency at the price of visibility.
The operator’s function is shifting
Much less hands-on-keyboard marketing campaign administration. Extra strategic oversight, inventive route, first-party information administration, and measurement structure. The ability set that issues in 2026 is just not bid administration. It’s understanding what alerts to feed the system, find out how to construct a inventive testing pipeline, find out how to personal first-party information technique, and find out how to interpret why the AI made the choices it did.
For extra on how AI is reshaping software program broadly, see our version on “The “SAASpocalpse.”
Last Thought
Paid media is just not magic. Nevertheless it is without doubt one of the strongest progress engines out there to B2B corporations when the timing is true and the execution is systematic.
The chance remains to be huge. The most effective time to construct your paid media muscle is earlier than your opponents determine it out. Begin now. Begin small. However begin.
Tag @GTMnow so we are able to see your takeaways and assist amplify them.
Trimble is buying Doc Crunch, bringing AI-powered contract threat evaluation into its development undertaking administration ecosystem. Development is without doubt one of the final industries to modernize round doc compliance, and purpose-built AI that catches fee disputes and notification failures earlier than they escalate is an actual wedge. Vertical AI discovering its manner into incumbent platforms is the sample to look at.
Granola raises $125M at a $1.5B valuation, up from $250M lower than a 12 months in the past. Assembly notes are a commodity at this level, and Granola is aware of it — the actual guess is on turning into the enterprise context layer, with new APIs that permit groups pull assembly intelligence into AI workflows. The valuation bounce says traders imagine whoever owns the context wins.
VC: Investing at Inception within the Age of AI Brokers with Ed Sim
For the complete factor, pay attention on Apple, Spotify, YouTube or wherever you get your podcasts by looking “The GTMnow Podcast.”
Armada – named amongst Quick Firm’s Most Progressive Corporations in computing, alongside NVIDIA and Google. They’re constructing distributed AI infrastructure for the environments most platforms ignore. Value watching as enterprises and governments push AI into the sector, not simply the cloud. To study extra it’s also possible to try the GTMnow episode with CEO Dan Wright.
Examen – raised $4.3M in whole funding and launched. Examen is constructing an autonomous analyst for industrial actual property.
-
Enterprise Account Govt, US & Canada (Dallas, TX)
-
Buyer Success Supervisor at Esper (Austin, TX)
-
Buyer Success Supervisor at gaiia (Distant – Canada)
-
Affiliate Account Govt (Sports activities Playing cards) at Alt (Distant – US/Canada)
-
AVP, buyer success (West) at Author (Hybrid – San Francisco, CA)
-
Development Advertising Supervisor (PALs/Client) at Tavus (San Francisco, CA)
See extra high GTM jobs on the GTMfund Job Board.
Upcoming occasions you gained’t need to miss:
-
MicroConf 2026: April 12–14, 2026 (Portland, OR)
-
GTMfund Dinner: April 14, 2026 (Austin, TX)
-
HockeyStack’s High Secret Launch Celebration: April 16, 2026 (San Francisco, CA)
-
SaaStock USA: April 15–16, 2026 (Austin, TX)
-
Forrester B2B Summit: April 26–29, 2026 (Phoenix, AZ)
-
SaaStr Annual: Might 12–14, 2026 (San Mateo, CA)
-
GTMfund Dinner: Might 14, 2026 (San Francisco, CA)
-
GTMfund Dinner: June 9, 2026 (London, UK)
-
Dreamforce 2026: September 15–17, 2026 (San Francisco, CA)
-
INBOUND: September 16–18, 2026 (Boston, MA)
-
Pavilion GTM2026: September 28–October 1, 2026 (NYC, NY)
-
CVC Week by Counterpart Ventures: September 29, 2026 (San Francisco, CA)
-
Buyer Success Week: October 5-9, 2026 (NYC, NY)
-
TechCrunch DISRUPT: October 13–15, 2026 (San Francisco, CA)
Some GTMnow group (founder, operator, investor) love to shut it out – we recognize you.
