ABM vs Demand Generation: Which One Fits Your Deal Size

Try to upload your list of 60 target accounts to LinkedIn and it will not take. A company targeting list needs at least 300 rows, and 60 rows is not 300. In our work that error message tends to arrive before any of the strategy does. It is the ABM vs demand generation decision showing up early, in a form that has nothing to do with strategy.

So this is not a piece about which model is better. They answer different constraints. The constraints are arithmetic. What follows is the fit test we run: what the ad platforms will let you target, what one deal is worth, and how many accounts your sales team can genuinely work. For the execution side, once you have decided, we wrote a separate account-based lead generation playbook.

Two models, two different constraints

Strip the labels back. The real difference is what your money is attached to: a click, or a company.

Demand generation buys attention across a market and sorts it afterwards. The cost attaches to a unit of interest: an impression, a click, a download, a booked call. You do not know which company will show up. So you build a machine that handles whoever does. ABM buys attention at a list of companies you named in advance, and the cost attaches to the account, whether or not it ever raises a hand.

That one difference sets everything else. In our work, the cost of a unit of interest is something you can push down: better creative, tighter targeting, a landing page that converts. The cost of an account is mostly labour, and labour barely moves. The research, the tailored content and the rep’s follow-up happen once per account, however many you run. Two hundred accounts cost roughly ten times what twenty do.

Your account list has a floor, and LinkedIn sets it at 300

This check is mechanical, and it comes before any framework does. Named-account targeting is not free-form. The platforms publish minimums, and they may not be where you expect.

LinkedIn draws two separate lines, and they count different things. A company targeting list needs at least 300 rows to upload, and it can hold up to 300,000 companies. Then the audience that list matches has to reach 300 member accounts before an ad set will deliver. Match fewer and Campaign Manager shows < 300 members and refuses to run it. You also have to pick a location, and that filter cuts the match again, so a segment that looked fine can drop back under the line.

Data table

Measure	Value
Member accounts: the minimum audience size for an ad set	300
Rows: the minimum for a company targeting list upload	300
Companies: the list size LinkedIn recommends	1,000

Now read LinkedIn’s own recommendation next to that floor. They suggest 1,000 or more companies for a company targeting list. For Sponsored Content and Sponsored Messaging they suggest a minimum audience of 300,000 members. Those are suggestions, not gates, and the distance between them and the 300 floor tells you what the product was built for. A 60-account list is not a small version of it. It is a different activity that happens to use the same upload screen.

Google Ads sets its own line, and it is lower. A Customer Match list uploaded or refreshed after 1 February 2024 needs at least 100 active users to serve, and Search, Display and YouTube each need 100 active users in the last 30 days. Membership has a maximum duration of 540 days, and a list needs at least 100 members added or updated inside that window to stay eligible. So a list nobody touches can stop qualifying while it is still sitting in the account. Do not read LinkedIn’s floor onto Google or the other way round. Each platform counts its own thing, and you have to clear each one separately.

None of this says ABM is wrong at 60 accounts. It says a 60-row list is not a LinkedIn company audience, so the touches have to come from somewhere with no audience minimum at all: sales outreach, events, a partner, direct mail, a rep who actually knows the company. If your plan for reaching 60 named accounts was a LinkedIn company list, the plan has a hole in it. The hole is visible on day one.

What one deal is worth sets your budget per account

Now the deal-size half. Two numbers you already have decide it.

Data table

Average contract value	Budget per account per year
$5,000	$125
$20,000	$500
$50,000	$1,250
$150,000	$3,750

The first is what you will spend to win a deal. Take your average contract value and the share of first-year revenue you accept spending on acquisition. The second is your account-to-deal rate. Out of the named accounts you work in a year, how many sign?

Budget per deal is contract value times the acquisition share. Budget per account per year is that number divided by how many accounts you work per deal won. Neither input is a benchmark, so put your own in. To show the shape of it, take a 25% acquisition share and an account-to-deal rate of one in ten.

A $5,000 contract value gives you $1,250 to win a deal, which leaves $125 per account per year. Move to $20,000 and it is $5,000 per deal, $500 per account. A $50,000 contract value produces $12,500 per deal and $1,250 per account. The $150,000 row lands at $37,500 per deal and $3,750 per account. Same two inputs every time. The budget per account moves by a factor of 30.

Start at the $5,000 end and the answer is already written. One hundred and twenty five dollars a year does not buy account research, a tailored asset, a sequence and a rep’s attention for one company. It does not buy any one of them on its own. At that contract value you are not choosing ABM over demand generation. You are choosing between demand generation and nothing.

The $150,000 end reads the other way. Three thousand seven hundred and fifty dollars per account per year is a real programme. Proper research, something written for that company, a coordinated set of touches, a rep who has been briefed. And the volume model starts to strain up there, because a deal that size usually needs a buying group of six or more people who all have to be reached.

Sales capacity is the input nobody writes down

The third input is the one we most often find missing from a plan. It quietly voids the other two.

ABM only works if a person follows the marketing. Someone reads the account research. Someone notices the engagement, then writes the mail that references what the company said on its earnings call. That is a per-account cost measured in hours, and it lands on a sales team that already has a number to hit.

So count it before you size the list. How many accounts can one rep genuinely work in a quarter, on top of the pipeline they already carry? Ten. Twenty. Rarely more, if the work is the kind that makes ABM different from a mail merge. Multiply by your reps and you have the real ceiling, and it usually sits far below the number the workshop produced.

A list longer than that ceiling is not an ABM programme. It is a demand-generation audience wearing an account-based label. It will produce demand-generation results at ABM prices. When we pick up an account-based programme that is not working, this is usually what we find.

The fit test in one table

Three inputs, one reading. Nothing here is a verdict. Plenty of companies sit in the middle and should run both.

Read the rows, not the totals. One row can decide it on its own. If your reps have no capacity, the contract value row stops mattering, because the programme has nobody to land it.

Most teams need both, in a specific order

The comparison framing hides the thing that works. Run properly, these are not competing budgets. They are two stages of one machine, and the order matters.

Demand generation tells you which accounts to name. Before that, you are guessing, and a guessed list is the fastest way to burn an account-based budget. Run the volume layer. Watch which companies keep appearing in your analytics, your CRM and your pipeline generation reporting. Let the list assemble itself out of behaviour you observed.

Then ABM does the part volume cannot. It reaches the five other people at an account where one person already showed interest. That is the split we run at COSEOM®. The volume layer finds the account. The account layer converts the buying group. Neither does the other’s job, and running only the second means naming accounts off a wish list.

One practical marker for the handover. When a named account starts appearing in more than one channel without you paying to put it there, it has earned a place on the list. Until then it is an opinion.

Frequently Asked Questions

Is ABM better than demand generation?

Neither is better. They are priced differently and they solve different problems. Demand generation attaches cost to a unit of interest, and that is a cost you have levers on: creative, targeting, the page people land on. ABM attaches cost to a named account, and that cost is mostly labour. Which one fits comes down to your contract value, whether you can name the accounts in advance, and whether your reps have room to work them.

What is the minimum account list size for ABM on LinkedIn?

Two numbers, counting two different things. A company targeting list needs at least 300 rows to upload. The audience it matches then needs at least 300 member accounts, and a list matching fewer shows as < 300 members and cannot run. LinkedIn’s own recommended company list size is 1,000 or more companies, which is a suggestion rather than a gate.

How do I work out the budget per account for an ABM programme?

Multiply your average contract value by the share of first-year revenue you accept spending on acquisition. Then divide by how many named accounts you work per deal won. With a 25% share and a one-in-ten account-to-deal rate, a $50,000 contract value gives you $1,250 per account per year, and a $5,000 contract value gives you $125. Those are arithmetic on your own inputs, not industry figures.

Can you run ABM without paid media?

Yes. When your list misses the floor on the platform you wanted, that is exactly what you are doing. Sales outreach, events, partners, direct mail and a rep who knows the company are all account-level channels with no audience floor. LinkedIn joins in once the matched audience is big enough to serve there. Google’s Customer Match floor is a separate one, and clearing it tells you nothing about clearing LinkedIn’s.

At what deal size does ABM stop making sense?

There is no fixed line, but the arithmetic gets uncomfortable fast. On a 25% acquisition share and a one-in-ten account rate, a $20,000 contract leaves $500 per account per year and a $5,000 contract leaves $125. Once that will not cover research, one tailored asset and a rep’s follow-up, the programme is account-based in name only.

Should demand generation come first?

In most cases yes. It tells you which accounts are worth naming. A target list built from observed behaviour beats one built in a workshop, and the volume layer is what produces that behaviour. The exception is a market small enough to list every possible buyer, where there is nothing left to discover.

Do Google Ads have a minimum list size like LinkedIn?

Yes, though a lower one. A Customer Match list uploaded or refreshed after 1 February 2024 needs at least 100 active users, and Search, Display and YouTube each need 100 active users in the last 30 days for ads to serve. Membership also carries a maximum duration of 540 days, and the list needs at least 100 members added or updated inside that window to stay eligible.