What is the top of the sales funnel?
The top of the sales funnel is every person who fits your ideal customer profile and now knows you exist, whether you reached them or they found you. In B2B it is often shortened to TOFU, with MOFU and BOFU for the middle and bottom. At the top, nobody has agreed to a call yet. They have only become someone you can follow up with.
Two tests decide whether a name belongs there: the person fits (role, company type, size, location), and there has been a first touch in either direction. A list of names you bought or exported is a target list until you contact someone on it.
| Example | Top of funnel? | Why |
|---|---|---|
| A head of finance you sent a first email to on Monday | Yes | Fits, and you made the first touch |
| A founder who commented on your post and runs a 20-person agency you could sell to | Yes | Fits, and they made the first touch |
| A warm introduction from a customer | Yes, often further down | Fit and interest may both be known already |
| 5,000 exported names nobody has contacted | No | A target list, not yet a funnel stage |
| A student or a competitor who follows you | No | Aware, but not a buyer |
| Someone who replied asking for a price | No, middle | They have shown interest, so they have moved down |
The middle starts at the first sign of interest: a positive reply, a booked call, a form that names the problem. For the full set of stages and what each one converts at, see our guide to sales funnel stages and conversion rates.
Why is most of the top of the funnel not ready to buy?
Because most of your market isn't buying anything right now. LinkedIn's B2B Institute calls this the 95-5 rule: in an article by Ty Heath, Director, Market Engagement at LinkedIn, it says its research shows 95% of potential buyers aren't ready to buy today. The same page cites joint research with the Ehrenberg-Bass Institute, published as How B2B Brands Grow, a series of papers that includes one by John Dawes on the 95-5 rule. The 95-5 page quotes that joint research: 75% of companies buy computers once every 4 years, and 80% change banking services once every 5 years.
The arithmetic behind it is simple. If a company replaces a tool once every 5 years, and purchases are spread evenly, about 1 ÷ 5 = 20% of companies buy in a given year and 20% ÷ 4 = 5% in a given quarter. If your market is 2,000 companies that fit, that is about 100 buying in any quarter, and 1,900 that won't buy from anyone yet, however good your message is. Your category's cycle may be shorter or longer, so treat 5% as a planning figure and swap in your own replacement cycle.
LinkedIn sells advertising, and its page advises investing in lead generation for the 5% while investing far more heavily in brand advertising that reaches the entire category. A small team without an ad budget can still split its top-of-funnel time the same way:
- Find the few who are buying now. Look for a reason to write today: a job post, a new role, a public question about the problem. Our list of 15 free buying signals a small team can act on covers where to find them.
- Stay known by the rest. Regular posts, useful comments and a short note every quarter, so that when one of the 1,900 starts looking, your name is already familiar. The same series includes a paper by Jenni Romaniuk on why lack of brand awareness is a much bigger problem for a B2B brand than negative perception.
How do you fill the top of the sales funnel?
Pick one or two channels where your buyers already are, and learn each channel's own rules before you add volume. The table below compares six common channels for a small B2B team. In the first three rows the rule comes from the platform's or the regulator's own page, checked 5 Nov 2026; the last three rows are practical notes.
| Channel | Best when | Rule to know first |
|---|---|---|
| Cold email | You can find work emails and your offer reads well in five lines | The FTC's CAN-SPAM compliance guide says the law makes no exception for business-to-business email, and requires a valid physical postal address and an opt-out honoured within 10 business days |
| LinkedIn connection and message | Your buyers have current LinkedIn profiles and post or comment there | LinkedIn's User Agreement (effective 3 November 2025) bans bots and unauthorised automated methods for adding contacts, sending messages and engaging with posts |
| X replies and DMs | Your buyers argue about tools and problems in public on X | X's automation rules (updated April 2026) say unsolicited Direct Messages may not be sent in a bulk or automated manner |
| Posting and commenting | You have opinions or results worth sharing every week | Slow to start; it mostly serves the out-of-market majority |
| Referrals and warm introductions | You have a few happy customers or a strong network | Each one needs a personal ask, so volume stays low |
| Communities and events | Your buyers gather in a forum, Slack group or meetup | Each community sets its own rules on self-promotion; read them before posting |
Two of those rules contradict common growth advice. LinkedIn's User Agreement lists, among things you agree not to do: "Use bots or other unauthorized automated methods to access the Services, add or download contacts, send or redirect messages". LinkedIn's help page on prohibited software repeats that wording from section 8.2. X's rules give a concrete example of what isn't allowed: "sending automated replies to posts based on keyword searches alone is not permitted". Read our breakdown of what LinkedIn allows and what gets accounts restricted before you buy any tool for the top of your funnel.
The CAN-SPAM summary above is a short reading of the FTC's guide, not legal advice. The same guide says each separate email in violation can bring penalties of up to $53,088.
Which channel fills it best? A worked example
Compare channels by founder hours per positive reply, not by replies per 100 people contacted. Here is a worked example for a sample company, Northwind Ltd, with 6 hours a week for the top of the funnel. Every count and minute below is invented for the example; replace them with your own after a week.
| Channel (example) | People reached | Time | Positive replies | Per 100 reached | Hours per positive reply |
|---|---|---|---|---|---|
| LinkedIn messages to people with a signal | 30 | 30 × 6 min = 180 min = 3 h | 4 | 4 ÷ 30 × 100 = 13.3 | 3 ÷ 4 = 0.75 |
| Cold email to a researched list | 100 | 100 × 1.5 min = 150 min = 2.5 h | 2 | 2 ÷ 100 × 100 = 2 | 2.5 ÷ 2 = 1.25 |
| Two posts and 10 comments | Unknown | 0.5 h | 1 (an inbound DM) | Can't compute | 0.5 ÷ 1 = 0.5 |
| Total | 130+ | 6 h | 7 | 6 ÷ 7 = 0.86 |
Three things to read from it. First, cold email looks worst per 100 people but costs only 0.5 hours more per reply than LinkedIn, so it isn't automatically the channel to drop. Second, posting has the lowest cost per reply in this example and no measurable reach, which is normal: you can't count the people who saw a post and said nothing. Third, 1, 2 or 4 replies are tiny numbers. One extra reply would change the cold email row from 1.25 to 0.83 hours (2.5 ÷ 3). Run the same split for four weeks before moving hours between channels.
How do you measure the top of the funnel?
Measure the top of the funnel with four weekly counts and three ratios, all of which you can pull from your inbox, calendar and CRM.
| Number | How to count it | What it tells you |
|---|---|---|
| New fit people added | Names added to your CRM this week that pass your fit test | Whether research is keeping up |
| Share with a written reason | Added with a signal noted ÷ all added | If you're writing to the 5% or guessing |
| People contacted | First messages sent, by channel | Your real activity, not your plans |
| Positive replies | Replies that show interest, by channel | The handover to the middle of the funnel |
| Inbound first touches | Form fills, DMs and comments from people who fit | Whether staying known is working |
| Positive reply rate | Positive replies ÷ people contacted | Message and targeting quality |
| Hours per positive reply | Top-of-funnel hours ÷ positive replies | Which channel earns your time |
The "written reason" column is the easiest to skip, and the first fix below depends on it. If you'd rather not do the research by hand, Siela's Lead Finder looks for a reason to reach out now, scores every lead hot, warm or cold with a one-line reason, and saves it to the built-in CRM in its own group for each run, so each lead arrives with its reason already written. Once replies arrive, the work moves to the middle of the funnel; our guide to booking more sales meetings from replies picks up there.
How do you fix a weak top of funnel?
Most top-of-funnel problems show up in the weekly sheet as one of four patterns. Use the one that matches your numbers.
Many contacted, almost no positive replies
Check the reason share first. Cut volume by half for two weeks and write only to people with a written reason.
Replies, but from people who can't buy
Your fit test is too loose. Add the role, company size or location that the last five bad replies had in common to your exclusions.
Good replies, but too few people added each week
Research is the bottleneck. Time how long one name takes; if it is over 5 minutes, narrow to one signal type you can find quickly.
No inbound first touches at all
Nobody outside your messages has heard of you. Book two short posting slots a week and keep them for a quarter, since the payoff comes from people who aren't buying yet.
Frequently asked questions
What does top of funnel mean in sales?
It means the first stage of the buying process, where people who fit your customer profile become aware of you, either because you contacted them or because they found you. Nobody at this stage has shown clear interest yet. They move to the middle once they reply positively, book a call or ask about price.
What do TOFU, MOFU and BOFU stand for?
They stand for top, middle and bottom of the funnel. TOFU is awareness and first touches, MOFU is interest and qualification, such as replies and discovery calls, and BOFU is the decision, such as proposals and contracts. The labels come from marketing, but a small team doing its own sales can use them for the same stages.
Is top of funnel a marketing job or a sales job?
In larger companies marketing usually owns inbound awareness and sales development owns outbound first touches. In a small B2B team the same people often do both, so the useful split is by activity: finding fit people, making first contact and staying known. Track all three on one weekly sheet whoever does them.
What is a good top of funnel conversion rate?
There is no single figure that holds across channels and offers, because each source counts replies and leads differently. Measure positive replies divided by people contacted for each channel, then compare that with your own previous four weeks. A rising rate matters more than matching someone else's benchmark.
How many leads should be at the top of the funnel?
Work backwards from the meetings you need and your own reply and booking rates, then check that number against your market size. If only a few hundred companies fit, far fewer are buying in any quarter, so the realistic goal is reaching the buying few now and staying known by the rest.
Sources
- LinkedIn B2B Institute: 95-5 Rule (Ty Heath) (checked 5 Nov 2026)
- LinkedIn B2B Institute: How B2B Brands Grow (checked 5 Nov 2026)
- FTC: CAN-SPAM Act, A Compliance Guide for Business (checked 5 Nov 2026)
- LinkedIn User Agreement (effective 3 November 2025) (checked 5 Nov 2026)
- X Help Center: Automation rules (updated April 2026) (checked 5 Nov 2026)
- LinkedIn Help: Prohibited software and extensions (checked 5 Nov 2026)