Skip to content

How to Run Account-Based Marketing for SaaS

How ABM works for SaaS, from picking accounts on fit to reading trial and usage signals for timing and sending each one to an owner in sales.

Written by Sabih AhmedPublished 19 min read

Account-based marketing for SaaS has a data source sales-led ABM programs lack. Your product already records who at your target accounts is trying it. That data usually sits in an analytics tool your sales team never opens. So an ABM program can end up treating accounts already in a trial as strangers, showing them ads while nobody in sales knows they're trying the product.

The usual ABM advice was written for sales-led companies, so it says little about trials, self-serve sign-ups or who should act when usage jumps. This guide covers how account-based marketing for SaaS works with product and trial signals, which signals matter and who acts on them. It also shows how the work changes with budget and team size, and how to prove what it added to pipeline.

How does account-based marketing work for SaaS?

Account-based marketing works for SaaS the way it works anywhere, plus one source of timing other companies lack, which is what people at a target account do in your product. You choose accounts on fit, read trial and usage activity by account for timing, and send each signal to a named owner in sales.

ABM means account-based marketing, where sales and marketing work a named list of accounts together, with plays built for each account or group instead of a broad audience. In a sales-led company, the signs that an account has a reason to act now come from outside the product, such as a new sales leader or a hiring spree. A SaaS company with a free trial, a freemium plan or self-serve sign-up sees something closer to the decision. People at the account are already using it, and that tells you what other ABM data can't, which is whether the product works on their own data.

A sales-led SaaS company without a free trial sees product signals only from customers, so its new-business ABM runs on outside timing signals and product data drives expansion.

It's tempting to treat a good-fit list as a ready-to-buy list, but fit and timing are separate questions. Fit asks whether a company could buy from you and has the pain you solve, and timing asks whether it has a reason to act now. A list of good-fit accounts with no trigger goes quiet, however well the ads run. In SaaS, product activity is how you answer the timing question.

So a SaaS ABM strategy keeps 2 lists. The fit list holds every account that could buy, and the core list holds the fit accounts showing a reason to buy now. Product activity is the strongest way onto the core list, but it never puts an account on the fit list by itself. Between them, the 2 lists make your target account list, the named accounts a program works, chosen on fit, checked for timing and agreed with sales before anything launches.

SaaS teams merge the 2 lists in either direction, and both cost pipeline. One treats the fit list as ready to buy and spends a quarter's ad budget on accounts with no reason to change. The other treats every busy trial as a target account, so reps spend their week on companies too small for the plan they sell.

Plays run at different depths. 1:many means ads and email to the whole target list, and 1:few means one play for a group of 5 to 15 accounts that share a trigger. 1:1 means plays built for a single account, and our guide to one-to-one ABM explains all 3 tiers. In SaaS, the trigger is often the same product signal, such as a trial that stalled at the same setup step.

How to run account-based marketing for SaaS in 6 steps

To run account-based marketing for SaaS, take these 6 steps in order, and settle the first 5 before the first ad runs.

  1. Build the fit list from your closed-won deals and the pain your product solves, because size and industry alone only narrow the list.
  2. Set aside 1 in 5 fit accounts at random as a holdout, so you can later show what the program added.
  3. Tie every sign-up and product event to a company, so usage can be read by account rather than by person.
  4. Find the 2 or 3 product signals that came before past deals far more often than before trials that didn't buy, such as a second active user.
  5. Agree with the sales lead on the rule that moves a fit account onto the core list, and name an owner for every account.
  6. Run 1:many ads to the fit list, minus the holdout when the list is big enough, and from about $10k a month, 1:few plays for core accounts.

Is ABM worth it for SaaS?

ABM is worth it for SaaS when a deal is big enough to pay for a rep's time on one account and several people at the buyer decide together. It isn't worth it when buyers sign up, pay by card and never talk to anyone.

The first test is annual contract value against the cost of a rep's attention. A 1:few play takes hours of selling time and some content for each group, so it has to win deals whose contract value covers that time. When a typical customer pays a small monthly fee and one person decides, the product, the onboarding emails and the pricing page do the selling.

The second test is the buying group, the people at one account who decide a purchase together. Typically that's the person who'll use the product, the one who owns the budget and the one who could block it.

If the user also owns the budget, the trial does the job ABM would do. ABM earns its cost when other people at the account have to agree before anyone pays.

A SaaS company with a self-serve plan and a sales-assisted plan for larger teams usually needs ABM only for the second. Small accounts keep buying on their own. ABM picks out the larger accounts where a rep, a security review or a procurement step stands between the trial and the contract.

ABM also needs a list you can name. If your buyers are thousands of small firms you can't tell apart before they sign up, put the budget into the product and the self-serve funnel instead.

Which product signals matter for ABM?

The product signals that matter for ABM show an evaluation spreading through an account, moving toward a purchase or stalling. A single sign-up shows one person's interest, while a second active user, a connected integration or a plan limit shows a team making the product part of its work.

Product data has to be read by company, and tools such as Amplitude and PostHog record people by default. Amplitude's documentation on account-level reporting (opens in a new tab) says it reports on individual users by default, and analysis by company, team or account comes with its Accounts add-on. PostHog's group analytics (opens in a new tab), which aggregates events by companies instead of individual users, is a paid add-on. Without an account add-on, matching sign-up email domains against the fit list still shows which fit accounts have started a trial.

Sign-ups from personal email addresses can't be matched to a company by domain, so ask for a work email or the company name at sign-up.

SignalWhat it tells youWho acts
A work-email sign-up from a fit accountSomeone at an account you want has started an evaluationMarketing, by keeping the account in the 1:many ads and the trial's onboarding email
2 or more people from one account active in the same weekThe evaluation has spread past one personThe account's owner in sales
The trial reaches the setup step your past paid accounts had in commonThey've seen the product work on their own dataThe account's owner in sales
An invite to someone in another team or roleThe buying group is formingThe account's owner, with content for the new role
A plan limit reached, such as seats or a paid featureThe current plan no longer fits how they use itSales for free-plan accounts, the account manager or customer success for customers
An active user visits pricing or security pagesThey're asking buying questions, not usage questionsThe account's owner in sales
A busy trial goes quiet for several daysThe evaluation stalled or moved to another productThe account's owner, with a short note
Usage falls in a paying accountThe renewal is at riskThe account manager or customer success

Every signal in the table is timing. Product data can also confirm fit, because the integration a trial connects shows the stack the account runs. Still, how much an account uses the product says when to talk, not whether the account is worth talking to.

Logins, page views inside the app and one person clicking around are activity rather than signals. They belong on the account record as context and never trigger a rep on their own. Some product-led teams call an account that meets their signal rule a product-qualified account. The name matters less than the rule behind it, which should come from your own past deals.

How to pick accounts from sign-ups, usage and trials

To pick accounts from sign-ups, usage and trials, check every active account against the fit list first, then rank the fit accounts by how far their usage has gone. Fit decides who is eligible, and usage decides the order.

The usage that matters comes from comparing your own trials. Pull the accounts that became paying customers or opened a deal in the last 2 quarters, and look at what they did in the trial before that happened. Then look at the fit trials from the same quarters that didn't buy.

The step paid accounts took far more often than those trials, such as connecting a data source or adding a second user, is your activation signal. A step nearly every trial takes, such as the first login, tells you nothing. With only a handful of paid accounts, use every one you have and treat the signal as a guess to check each quarter.

Keep the fit score and the usage score apart rather than adding them into one figure. A combined score lets heavy usage at a poor-fit company outrank a quiet trial at your best account, which is merging the 2 lists in a spreadsheet. Our lead scoring skill applies the same split to your won and lost deals, building a fit score and a behavior score from CRM fields and keeping them apart.

  1. Ranking on activity: Fit and usage added into one score

    Fit first, then usage: Fit decides who is eligible, and usage decides the order

  2. Ranking on activity: A step nearly every trial takes, such as the first login, read as a signal

    Fit first, then usage: The step paid accounts took far more often than fit trials that didn't buy, such as adding a second user

  3. Ranking on activity: Heavy usage outside the fit list, treated as a reason to call

    Fit first, then usage: Heavy usage outside the fit list, treated as a reason to check its fit

Our view: keep fit and usage apart when you rank trial accounts.

Take a hypothetical case. A company sells reporting software to finance teams on a 14-day free trial, with a monthly self-serve plan for small teams and annual plans sold by 2 account executives. Its fit list holds 400 companies with 50 to 500 employees that run an accounting system it connects to and still close their books in spreadsheets.

Its paid accounts from the last 2 quarters share a pattern. Nearly all connected an accounting system and had a second person active in the first week of the trial, while few of the fit trials that didn't buy did both.

In October, 120 trials start. Matching the 120 trials' email domains against the fit list finds 26 fit accounts outside the holdout. The rest are holdout accounts, smaller firms, personal addresses and companies outside the list.

Among the 26, 8 accounts have connected an accounting system and have a second active user. Each of the 8 joins the core list on the day it shows both, and goes to its named account executive with the signal attached.

The other 18 stay in the trial's onboarding emails and the 1:many ads until they show both. A weekly review with sales catches any that showed both signals but never reached their account executive.

One company outside the fit list has 5 active users and a connected accounting system. That usage is a reason to check its fit, not to call. It has 20 employees, too few for the annual plan to make sense, so it stays on the self-serve path. If several accounts like it buy the annual plan on their own, the 50-employee floor is too high, and that's a finding for next quarter's list.

When should sales act on a product signal?

Sales should act on a product signal when it comes from a fit account and shows the evaluation spreading, moving toward a purchase or stalling. The account's named owner acts, within a time limit agreed with the sales lead before anything launches.

Every account has a named owner for its next step, and that owner is usually in sales. The common setup has marketing run ABM and pass warm accounts to whoever picks them up, and product signals make that worse. An alert posted to a shared channel belongs to nobody, so it often gets a reaction and no reply. Marketing can warm an account, but a rep turns interest into a meeting.

So the rule comes before launch, with the sales lead in the room. Launching before sales agrees to the list is the common failure, because reps keep prospecting their own territories and the signals reach accounts nobody calls. The table below adds what the owner does and how fast, as a starting rule to adjust once your own deals show which signals came first. In the hypothetical case, the finance reporting company's rule asks for a connected accounting system and a second active user together, because that's what set its paid accounts apart.

When this happensWho actsWhat they doHow fast
A fit account has 2 or more active users in a weekThe account's owner in salesOffers help with the next setup step and asks who else will use itWithin 2 working days
A fit trial reaches the activation step and someone views pricingThe account's owner in salesOffers a call about the plan and the rolloutThe same day
A busy fit trial goes quiet for 5 daysThe account's owner in salesSends a short note asking what got in the wayWithin 2 working days
An account outside the fit list shows heavy usageMarketing opsChecks fit, then adds it to the fit list or leaves it self-serveAt the weekly review
A customer reaches a plan limitThe account manager or customer successTalks about the plan that fits how they now use itWithin the week

The first touch on a trial is help with the next step, which is why the call in the table waits for the activation step. A call request before the product has worked for them asks for time they haven't decided to give. Once the account reaches the activation step, a conversation about the plan and the rollout has something to stand on.

Speed matters more in SaaS than in sales-led ABM, because a trial has an end date. A rep who first reaches a 14-day trial in its second week has likely missed the days when the team decided whether the product works.

Once product fields reach the CRM, the rule becomes a workflow that creates a task for the account owner when the account crosses the line. Until then, a daily list from product analytics does the same job. Whichever route you take, the signal travels with the task, so the rep knows what changed. A rep who gets more alerts than they can work starts ignoring all of them, so the rule should send fewer accounts than the team can follow up.

A 30-minute review with sales each week covers the accounts that moved onto or off the core list. It's where the rule gets adjusted, and where an account nobody has worked gets a new owner or goes back to the fit list.

How does ABM work with product-led growth?

ABM works with product-led growth by adding a rep and targeted plays to the self-serve path only where a fit account needs them. The product still does the first part of selling, and ABM covers the parts it can't reach.

The biggest part it can't reach is the people in the buying group who never log in, often the budget owner and the person who could block the purchase. ABM's job in a product-led motion is reaching them, with the business case for the budget owner and the security and procurement answers for the person who could block it.

A free trial has a deadline, and a freemium plan doesn't. A freemium account can stay on the free plan for months, so its timing comes from limits, new teams joining and the buying questions people ask.

Ads play a small part inside a trial. LinkedIn's help page on company list targeting (opens in a new tab) says an uploaded list needs at least 300 rows, and the audience can take up to 48 hours to build. A week's core list of trial accounts is usually far shorter than 300 companies, and waiting up to 2 days for an audience eats into a 14-day trial. So the 1:many ads run on the fit list, where they keep you known before anyone signs up, and core accounts hear from their owner instead.

300 rows

the minimum for an uploaded company list

LinkedIn Help, Company list targeting

Up to 48 hours

how long the audience can take to build after a list is uploaded

LinkedIn Help, Company list targeting

How do you run ABM for expansion revenue?

To run ABM for expansion revenue, treat your customers as a target account list of their own, with fit set by the room to grow and timing read from usage. The owner is usually the account manager or customer success manager, because they already hold the relationship.

Room to grow could mean other teams, regions or business units that could use the product, or a seat count well below the company's size.

The timing comes from the product and from people. An invite to someone in a new team is a reason to talk now, and so are a plan limit reached and usage spreading to a new department. A new executive who used your product elsewhere is another.

Customers who reach the same limit share a trigger, so they make a natural 1:few group, with one play built for that limit. The step above 1:few is 1:1, plays built for a single account. In expansion, that's the largest accounts, where a new business unit would change the size of the contract.

Falling usage is never an expansion signal. An account whose usage drops before renewal needs its account manager or customer success to find out why. An upsell play at that moment tells the customer you weren't watching.

Every expansion account needs a named owner as well. When customer success owns usage and sales owns new business, a signal from a new business unit falls between them unless the rule says who acts.

Ads matter less here, because the account manager already holds the relationship and can reach the buying group directly. In-product messages carry the rest of the play.

How does ABM for SaaS change with budget and team size?

Budget sets the shape of the program, from 1:many on one list at $5k a month to all 3 tiers at $100k and up. It also sets how product signals reach sales, and team size sets how many core accounts you can follow up.

A program at $5k a month and one at $100k a month are different programs, not one program scaled down. The table shows the program each budget can carry, how product signals reach sales at that level, and who usually runs it. The monthly budget is program spend on media, content, tools and mail, and the people who run it are on top. These tiers are our starting point, so adjust them to your deal size and sales cycle.

Monthly budgetProgram shapeHow product signals reach salesWho runs it
$5k1:many on one tightly chosen listA daily list of fit accounts that crossed the rule, from product analytics or sign-up domains, checked manuallyOne marketer, with an hour of the sales lead's time each week
$10k1:many, plus 1:few plays for the best-fit groupA few account fields synced into the CRM, so each owner sees usage on the account record2 marketers, or 1 with sales running the 1:few outreach
$20kTiered, with 1:few for the top group and 1:many for the restFit and usage scores kept apart on the account record, with a task for the owner when an account crosses the line2 or 3 marketers, with content or ops support
$50k1:1 for a small top tierProduct data synced from the warehouse into the CRM and the ad platforms on a scheduleA small team with an ops owner
$100k and upAll 3 tiers, with a dedicated ownerThe same sync, with a platform that joins product, site and ad signalsA dedicated ABM team with reps assigned to accounts

Below $5k a month, ABM isn't smaller, it's narrower. It runs on data you already own, such as sign-ups, trial activity and site visits, so the budget goes into acting on those signals rather than buying them.

Take the finance reporting company with a 14-day trial and 2 account executives. At $5k, one marketer pulls the fit accounts that crossed the rule each morning and sends each to its account executive, and the weekly review covers what moved. The 1:few plays wait until $10k.

If the same company grows into a $50k budget, product data reaches the CRM and the ad audiences without anyone exporting a file. Hightouch's documentation (opens in a new tab) calls that reverse ETL, moving warehouse data into the tools where teams act, such as CRMs and ad platforms.

The mistake at every budget is buying the platform before the process. A tool that scores product-qualified accounts automates a rule, so it can't help until you know which signal came before your deals and who acts on it. Start with the CRM and the analytics you have, and buy a platform once moving lists manually costs more than it does.

Team size sets a limit of its own. One marketer can run 1:many on one list and the weekly review. 1:few plays for trial accounts also need an owner in sales with time to follow up within days. If sales can't take more than a few new accounts a week, tighten the rule rather than sending accounts nobody will call.

How do you measure ABM in a SaaS business?

Measure ABM in a SaaS business by comparing the fit accounts in the program with a random holdout from the same fit list. Compare opportunities opened, trials converted to paid and contract value, because influenced pipeline can't tell you what would have happened anyway.

ABM proves itself by incremental pipeline against a holdout group, not by influenced pipeline. Influenced pipeline counts every deal from an account that saw a play. In a self-serve motion, some trials convert to paid without any help from sales, so an influenced count credits ABM with revenue it didn't create. Finance discounts the figure for that reason.

Before the core list exists, set aside 1 in 5 fit accounts at random. They keep whatever your trial process did before ABM, such as onboarding emails and replies to inbound questions. They get no ABM plays or alerts to an owner, and no ABM ads when the list is big enough to spare them (see below). A holdout account that starts a trial stays in the holdout, even when it shows a signal.

Taking the holdout out of the 1:many ad audience shrinks it. With 1 in 5 held back, a fit list needs about 375 companies for the uploaded list to reach LinkedIn's 300-row minimum. A shorter fit list can't do both, so leave the holdout in the 1:many ads and hold back only the plays and alerts. The test then measures what those added.

Compare every fit account outside the holdout, core accounts included, with the holdout. The core list alone can't be compared with a random group, because it was chosen for its signals. Accounts added to the fit list after launch go through the same random draw, or stay out of the comparison until next quarter's list is drawn.

Divide each group's opportunities, paid conversions and contract value by its number of accounts. Multiply the gap between the program's rate and the holdout's by the program's accounts. That's your incremental pipeline, and it ties ABM to the one pipeline figure every channel answers to.

5 equal columns: 4 for the program's fit accounts and 1, set apart, for the random holdout of 1 in 5. A thin line carries the holdout's level across all 5. The band above that line in the 4 program columns is lime: the incremental pipeline. It is a schematic, not measured data.

The holdout has a cost, because some of its trials would have converted with a rep's help. That cost buys a figure finance can check, because the comparison leaves out deals that would have opened anyway.

With a small fit list, the holdout opens too few opportunities in a quarter for its rate to mean much. With 1 in 5 held back, a 400-account fit list leaves a holdout of 80. Keep the holdout and read both rates over 2 or more quarters together, because a longer window gives each rate more deals to rest on. A comparison with the quarter before launch can support the result, but it's weaker, because it also counts changes in trial volume and season.

Where to start

Fit tells you who, product signals tell you when, and a named owner in sales turns the 2 into a conversation. If you'd rather not build the workflows that route those signals yourself, here's how we build CRM workflows. Whichever route you take, start by pulling the accounts that paid or opened a deal in the last 2 quarters, and the fit trials that didn't. Find the trial step the first group took far more often, then agree with your sales lead who acts when a fit account reaches it.

Sources

Every figure in this post links to where it comes from. These are the pages we checked.

  1. 01Amplitude's documentation on account-level reporting · amplitude.com
  2. 02PostHog's group analytics · posthog.com
  3. 03LinkedIn's help page on company list targeting · linkedin.com
  4. 04Hightouch's documentation · hightouch.com

About the author

Sabih Ahmed

Founder

Sabih founded SHAKEUP to run strategy, story and channels as one blend for B2B SaaS teams, with pipeline as the measure. He has sat on every side of the table, building campaigns, managing multi-million euro budgets and owning revenue targets. That is why his advice starts with the number a marketing leader answers for. Today he leads our strategy and messaging work, and wrote our B2B Google Ads playbook.

LinkedIn

Newsletter

One note a month on what's working.

One email a month from the work: what we tested, what moved pipeline and what we'd do differently.

  • One email a month, never more
  • Written by the people running the campaigns
  • Leave in one click

Sign-up opens with the first issue.

Prefer tools to reading? See our skills