Short answer
These are not rival philosophies, they are two positions in a sequence, and the order is set by two facts about your business: annual contract value and whether search demand for your category already exists. Above roughly $10,000 a year per customer, start with outbound, because it is the only channel that produces a first conversation inside a quarter, and if no search demand exists it is the only channel at all. Below that line, with demand already being expressed, start with inbound, because outbound economics rarely survive a small deal size. Below that line with nobody searching, start with neither: the constraint is pricing, packaging or the buying motion, and no channel purchase fixes it. Most companies above the line should eventually run both. The argument worth having is about which one you fund this quarter, not which one is morally superior.
Written by WeeGrow, who run the outbound half and not the inbound half, so read the recommendations with that in mind. Two of the four quadrants below carry a line in words saying we are the wrong call there. Benchmark figures were checked in August 2026, and each carries its source, its denominator and its sample size, or says that the source does not publish one.
The matrix, in four lines
Outbound can produce a first conversation in weeks. Inbound usually produces nothing at all for months, then produces leads for years.
Channel mechanics, not a measured statistic. Section 03 sets out what has to be true before either one delivers a conversation.
The definitions take one paragraph. Inbound lead generation means the buyer starts the conversation: they search, read, watch, ask a peer, and then fill in a form or book a call. Outbound means you start it, by reaching a named list of people who were not looking for you today, through LinkedIn, email or the phone. Direction of first contact is the whole difference. Everything else people argue about, intent, cost, trust, scale, is downstream of that one fact.
Because the definitions are settled, the useful question is which one you fund first. Two variables decide it. The first is annual contract value, because a conversation costs roughly the same to produce whether you sell a $3,000 tool or a $300,000 platform, so the deal has to be big enough to carry it. The second is whether search demand already exists, because inbound can only capture intent that is already being expressed. If nobody types the phrase, publishing against it produces nothing, however good the writing is.
You publish against questions your buyers already ask, earn the position, and wait. Intent is high because the buyer arrived under their own steam, and cost per lead falls over time as the asset keeps working. What you cannot do is control who shows up, or make it happen faster by spending more this month. Inbound rewards patience and punishes emergencies.
You choose the accounts, the people and the message, and go and find them. Fit is controlled at the input, volume is a dial you can turn, and the first replies arrive in the same weeks the sequence sends. What you cannot do is stop: nothing compounds, so the month you pause is the month the conversations stop. Outbound rewards consistency and punishes a bad list.
A programme that connects the two, so that outreach, messaging and conversion are one system rather than two departments, is what B2B demand generation is supposed to mean. But the strategy layer does not change the order of operations, and the order is what the next section settles.
Plot yourself on two axes. Vertically, annual contract value above or below roughly $10,000. Horizontally, whether people already search for what you sell. The $10,000 line is a judgement about when a single closed deal can carry a quarter of outbound, and it is worth seeing the sum rather than taking the number, because it does not hold across the whole market range.
Where the $10,000 line comes from
A quarter is three months. The market range for a done-for-you agency published on our B2B lead generation cost page is $1,500 to $5,000+ per month, so a quarter costs $4,500 at the bottom of that range and $15,000 or more at the top. Running it the other way, a $10,000 contract covers three months of retainer only up to about $3,300 a month, which reaches roughly the midpoint of the published range and no further. At the top of the range a $10,000 deal does not carry a quarter at all.
That sum is also on revenue rather than on gross profit. At a 70% gross margin a $10,000 contract contributes $7,000, which carries a quarter only up to about $2,300 a month, in the bottom quarter of the same range. So read $10,000 as a line derived from a retainer in the lower part of the market range, on revenue, in a business that closes what it starts. Move it up if your retainer is nearer $5,000, if your margin is thin, or if your close rate means several deals have to start for one to land.
Inbound first
People are already typing the phrase, so the cheapest lead in the market is the one that arrives at a page you already own. At a small deal size, outbound has to produce conversations at a cost per meeting that market retainers rarely reach, so it is the second move, aimed only at the segments your inbound data proves are worth chasing.
WeeGrow is the wrong first call here
Neither, yet
Nobody is searching, and the deal is too small to carry a per-conversation cost. Buying either channel here burns budget on a pricing or packaging problem. The honest moves are self-serve or product-led distribution, partnerships and communities where your buyers already gather, or moving the offer upmarket until a deal can pay for a conversation.
WeeGrow is the wrong call here, and so is every other channel purchase
Both, outbound first
One closed deal pays for a meaningful stretch of outbound at market rates, so outbound pays back fast and fills the calendar now. Search demand means content is a real asset rather than a hope, so start publishing in parallel at low intensity. In twelve months the mix should have shifted. In the first quarter, outbound is what produces pipeline.
This is the half WeeGrow runs
Outbound, clearly
Either the category is new or your buyers do not know the problem has a name, so content aimed at a keyword nobody types produces nothing no matter how good it is. Outbound is the only channel that reaches a buyer who is not looking, and it doubles as market research: the replies tell you the language to publish against once demand starts to form.
This is the half WeeGrow runs
| Annual contract value | Existing search demand | Fund this first | Why, compressed | Where WeeGrow fits |
|---|---|---|---|---|
| Under $10k ACV | Search demand exists | Inbound first | The cheapest lead is one already searching for you, and a small deal rarely carries a cost per conversation. | WeeGrow is the wrong first call here |
| Under $10k ACV | No search demand | Neither, yet | Nobody is searching and the deal is too small to pay for a conversation, so the constraint is the offer, not the channel. | WeeGrow is the wrong call here, and so is every other channel purchase |
| Over $10k ACV | Search demand exists | Both, outbound first | One deal pays for a meaningful stretch of outbound, and real search demand makes content an asset rather than a hope. | This is the half WeeGrow runs |
| Over $10k ACV | No search demand | Outbound, clearly | Content aimed at a keyword nobody types produces nothing, and outbound is the only channel that reaches a buyer who is not looking. | This is the half WeeGrow runs |
How to settle the demand axis in ten minutes: write down the phrase a buyer would type if they had your problem and did not know your company existed. Check whether it has any search volume at all, and whether anyone ranks for it with a real page. If the only meaningful searches are for your brand name, you are in the no-demand column, whatever your category page says.
Outbound can produce a first conversation in weeks. The sequence sends on day one, acceptances land within days, and a reply is an event inside the same send cycle rather than the end of a ranking cycle. What gates the speed is send volume and list quality, both of which you control.
Inbound has a queue of prerequisites before anything converts. The page has to exist, be crawled and indexed, earn enough authority to displace whoever holds the position today, and then accumulate enough traffic for a small conversion rate to produce a countable number of leads. That is months of work before the first lead, and the early part of it is unmeasurable by design. None of this is a criticism of inbound. It is the shape of the channel, and it is the same shape that makes inbound the cheaper source of leads three years later.
| Stage of the comparison | Outbound | Inbound |
|---|---|---|
| What has to exist before day one | A defined list, a profile or domain in good standing, and a sequence. Days of work, not months. | A page or post that answers a question people actually search, plus a site with enough authority to rank against whoever holds the position today. |
| What happens in the first fortnight | Requests and emails land. Acceptances and replies come back inside the same send cycle, because they are events in the sequence. | The page is crawled and indexed. Nothing converts yet, and that is the design of the channel, not a failure of it. |
| Where the first conversation comes from | A reply from someone on your list. You chose who they are, so fit is controlled at the input. | Whoever happens to search and convert. Intent is higher, fit is whatever the query brings you. |
| What actually gates the speed | Send volume and list quality. LinkedIn activity limits cap one profile, so more volume means more seats or more channels. | Indexing, competition for the query and accumulated authority. None of those respond to spending more this month. |
| What happens if you stop | Conversations stop within weeks. Nothing compounds. | Rankings and traffic persist for a long time. Everything compounds. |
If you have a number to hit this quarter, that asymmetry decides the order for you, and it is not close. Inbound is an excellent investment and a poor emergency plan. Funding it as your only channel when the board wants pipeline in ninety days is the single most common way B2B teams end a year short. The converse is also true: funding outbound forever, with nothing compounding behind it, means your cost per conversation in year three is the same as it was in month one.
Published outbound rates are useless without the denominator underneath them. A 10.4% reply rate measured against the accepted connections that received a message is a 3.0% reply rate measured against the requests sent to get them, because only 28.5% of requests are accepted in the first place. Same campaign, same replies, and it looks roughly three and a half times better on the narrower denominator. Most figures in circulation never say which was used. Every rate below carries its source, its denominator and its sample size, or says that the source does not publish one.
| Stage | Result | Rate, source, sample and denominator |
|---|---|---|
| Connection requests sent | 1,000 | Your input, not a benchmark. One profile in a month, chosen here only to make the arithmetic readable. It is not a volume LinkedIn is known to permit: LinkedIn polices activity and does not publish a number, so treat 1,000 as a unit of arithmetic rather than a target. |
| Accepted (Expandi reading) | 285 | 28.5% of requests sent (Expandi, 13,218,869 requests). Denominator: requests sent. |
| Accepted (Belkins reading) | 260 | 26% of requests sent (Belkins, which does not publish a sample size for this figure). Denominator: requests sent. |
| Variation inside acceptance | 27.6% vs 25.3% | Requests sent without a note vs with a note (Belkins, 11.5M requests). Separately, acceptance runs 17.5% to 40.1% across industries (Expandi, which does not publish a sample size for that breakdown). That is variation between industries, not between individual senders, so it tells you nothing about one campaign. |
| Replies (Expandi reading) | 30 | 10.4% of messages sent (Expandi, 6,730,447 messages), applied to the 285 accepted connections from the Expandi row above, one message each. 285 x 10.4% = 29.64. |
| Replies (Belkins reading) | 19 | 7.2% of messages sent (Belkins, 15.1M contacts), applied to the 260 accepted connections from the Belkins row above, one message each. 260 x 7.2% = 18.72. |
| Positive replies | Not published | No vendor publishes a positive-reply rate. Anyone quoting one is estimating, and that includes anyone who calls all 19 to 30 replies conversations. |
| Meetings booked | Not published | No vendor publishes a reply-to-meeting rate either. Instrument your own from the first month. |
Check the arithmetic yourself, and keep each reading inside one source. Expandi: 1,000 × 28.5% = 285 accepted, then 285 × 10.4% = 29.64, which rounds to 30 replies. Belkins: 1,000 × 26% = 260 accepted, then 260 × 7.2% = 18.72, which rounds to 19 replies. So one profile sending 1,000 requests in a month is working towards roughly 19 to 30 replies, an unknown fraction of which are positive, and the first of them arrive in the same weeks the requests go out. The gap between 19 and 30 is the gap between two vendors reading the same channel, not a forecast range. The full stage-by-stage version, including where the published data runs out, is on our LinkedIn lead generation funnel page.
Two honest caveats. First, nobody publishes a positive-reply rate or a reply-to-meeting rate, so any page that walks you from requests to booked meetings is estimating the last two steps. Instrument those yourself from month one and replace the assumptions as soon as you have thirty of your own data points. Second, a month of published content, started the same week, has produced zero leads at this point in the comparison and will not produce its first for a long while yet. That is not an argument against publishing. It is an argument about which channel you can rely on for the number you owe someone in ninety days. This is how we run the channel mix in practice: LinkedIn lead generation opens the conversation and cold email outreach carries the follow up.
If you want to test the $10,000 line against your own numbers rather than take ours, put your deal size, close rate and target into the ROI calculator and see how many meetings outbound has to produce before it pays for itself. Pricing on our side is a scoped quote after a free audit, not a tier you pick off a page.
For most companies above the $10,000 line the right answer is both, sequenced rather than split. Outbound at full intensity for the pipeline you need now. Inbound started in parallel at a cadence you can sustain, so that it is compounding by the time outbound reaches its ceiling. The failure mode is starting both at half strength: an outbound programme too small to produce meaningful volume, and a content programme too thin to rank against anyone.
WeeGrow runs the outbound half of that sequence, done with you rather than done for you: targeting, LinkedIn, cold email, calling, reply handling and the CRM underneath it. We do not write your blog, build backlinks, run your paid media or fix your technical SEO. If your quadrant says inbound first, your first hire is not us, and we would rather say that here than on a call after you have spent an hour on it.
Two of the four quadrants say your first move is not us, and both are marked in words on the matrix above. The first is a small deal size with search demand already being expressed: capture what people are already typing before you pay anyone to manufacture a conversation, and come back to outbound once your own data says which segments are worth chasing. The second, and the one we are most often wrong for, is a small deal size with no search demand. There the channel is not the constraint: pricing, packaging or the buying motion is, and buying outbound will produce expensive conversations with people who were never going to pay enough. Founder to founder, we will tell you that on the audit call. Asfand has spent 8+ years leading B2B sales, most recently as Head of Corporate Sales at InvoZone, with $30M+ in cumulative closed revenue across SaaS, technology, fintech, healthcare and telecom, and that experience is mostly useful here for saying when outbound is not the answer.
Inbound means the buyer starts the conversation: they search, read, ask a peer, then fill in a form or book a call. Outbound means you start it, by reaching a named list of people who were not looking for you today, through LinkedIn, email or the phone. That is the entire definitional difference. The decision worth spending time on is not which one is philosophically better, it is which one you fund first, and that depends on your deal size and whether anyone is already searching for what you sell.