TOFU, MOFU, BOFU explained

If you have spent more than a week in marketing, you have seen the acronyms — TOFU, MOFU and BOFU — usually thrown around as if everyone already agreed on what they mean. They are not jargon for jargon's sake. They are a simple, durable map of how a stranger becomes a customer, and once the map is clear, almost every content and budget decision gets easier. This is TOFU MOFU BOFU explained the way an operator would explain it: what each stage is, what to publish into it, what to measure, and how to stop treating the three as separate worlds.

The short version: the funnel narrows because buyer intent changes. At the top, lots of people are vaguely aware of a problem. In the middle, fewer people are actively comparing solutions. At the bottom, a small group is ready to decide. Same human, three different mindsets — and your job is to meet each one with the right thing. Here is how the three marketing funnel stages break down.

What TOFU, MOFU and BOFU actually mean

The acronyms expand cleanly: TOFU = top of funnel, MOFU = middle of funnel, BOFU = bottom of funnel. The classic funnel shape maps those three to a journey of shrinking volume and rising intent — from awareness, through consideration, to conversion. The three stages do not count equally toward overall funnel health: in dolv funnel intelligence, TOFU is weighted 0.25, MOFU 0.40 and BOFU 0.35. That is a deliberate, opinionated default, and we will come back to why MOFU carries the most weight. First, let us walk each stage on its own.

TOFU: top of funnel (awareness)

Top of funnel is where strangers first realize a problem — and a category of solutions — exists. The person is not shopping. They are not comparing vendors. They typed a question, watched a video, or stumbled onto a post. Your entire job at TOFU is to be useful enough to earn attention and broad enough to reach net-new audience. Pitch the product here and you lose them; teach them something and you have started a relationship.

• Content that fits: educational blog posts, SEO landing pages, short video, thought leadership, social posts, podcasts.
• What to measure: impressions, sessions, new visitors, audience growth and branded search — pure reach signals.
• The trap: chasing raw traffic. A million unqualified visitors who never move to MOFU is a vanity number, not awareness.

MOFU: middle of funnel (consideration)

Middle of funnel is the consideration stage. The buyer now knows they have a problem and is actively evaluating approaches — reading comparisons, attending webinars, building a shortlist. This is the stage where an anonymous reader becomes a known, engaged lead with real intent. It is also, not coincidentally, where most funnels quietly leak: plenty of TOFU reach, a strong BOFU offer, and a soft middle that fails to convert interest into pipeline.

• Content that fits: comparison guides, case studies, webinars, gated templates and tools, and email nurture sequences.
• What to measure: form fills, content engagement, lead score, MQLs and pipeline created.
• The lever: intent. The dolv intent-signal lead scoring separates a casual reader from a buyer in evaluation, so MOFU effort goes to the leads worth nurturing.

BOFU: bottom of funnel (conversion)

Bottom of funnel is the decision stage. The buyer is down to you and maybe one or two alternatives, and they need two things: proof that you can deliver, and a reason to act now. BOFU content is unapologetically about your product — because that is exactly what the buyer is researching at this point.

• Content that fits: demos, free trials, pricing pages, ROI calculators, and "vs" / alternatives pages like dolv vs HubSpot.
• What to measure: opportunities, demo requests, close rate, revenue and customers won.
• The trap: last-click thinking. BOFU closes deals, but TOFU and MOFU created the demand — credit them with attribution or you will starve your own pipeline.

Map your tools to the stages, or you're guessing

Knowing the three marketing funnel stages is step one. Step two is connecting each stage to live data, because a stage you cannot see is a stage you cannot improve. The reason teams struggle to "do full-funnel" is mechanical: reach lives in GA4, leads live in the CRM, spend lives in three ad managers, revenue lives in a spreadsheet. dolv ships 20 read-and-write integrations and maps each one to TOFU, MOFU or BOFU automatically.

With Gmail, Calendar, Drive, Sheets, Docs, Outlook, Teams, OneDrive, Excel, GA4, Search Console, Ahrefs, YouTube, WordPress, Product Hunt, the Google/Meta/LinkedIn ad platforms and LinkedIn all connected, every stage of the funnel reflects real activity — not the slice that happened to be easy to export. Search and content tools feed TOFU; ad platforms feed paid acquisition across TOFU and MOFU; the built-in CRM and intent-signal lead scoring feed MOFU and BOFU. A read-and-write connection matters here: it does not just report a number, it lets the system act on it later. For the difference between rules that fire and agents that reason, see AI agents vs marketing automation.

Roll TOFU, MOFU and BOFU into one number

Here is where most explanations stop and dolv keeps going. Three stages with three sets of funnel metrics is still three things to interpret. The point of the funnel model is to compress all of it into a single, comparable score, then drill down only when that score moves. dolv does this with a weighted composite health score: each stage gets a normalized 0–100 score based on how its current metrics compare to a rolling 30-day baseline — your own recent normal — and the three are blended by weight.

The math is intentionally boring: 0.25·TOFU + 0.40·MOFU + 0.35·BOFU. The value is in the inputs — real metrics from real integrations, compared to a baseline that updates every day. A composite of 73 that is up 3 versus baseline tells you the funnel is improving even if a single tile looks red. Context beats raw counts. If you want the full mechanics of scoring and baselining, our companion guide on how to measure full-funnel performance goes deeper than we can here.

Why MOFU carries the heaviest weight

Back to the weighting. The three stages are scored 0.25 / 0.40 / 0.35 because middle of funnel is where intent either becomes pipeline or evaporates. You can grow impressions all quarter and still have a flat business if MOFU is broken — the demand is arriving and leaking straight back out. BOFU sits close behind because it is the last yard before revenue. TOFU matters, but high reach with a broken middle is the most expensive kind of funnel: you are paying to fill a bucket with a hole in it.

This is exactly the kind of judgment the funnel model is meant to encode. Rather than staring at twenty charts and arguing about priorities, the weighted score makes the trade-off explicit: improving MOFU by ten points moves overall health more than improving TOFU by the same amount. That is the difference between a funnel diagram and a funnel you can operate.

Find the leak, then fix it where you found it

Once TOFU, MOFU and BOFU are scored against a baseline, the next question is causal: why did the weak stage dip? Eyeballing line charts for coincidences is exactly the work software should do. The dolv cross-metric correlation engine continuously compares metrics across stages and surfaces relationships you would otherwise miss — "MOFU lead score fell the week a gated asset changed," or "BOFU close rate rises when demo-booked emails go out within an hour." It turns a wall of funnel metrics into a short list of hypotheses worth testing, validated with real ICE-scored experiments and a genuine z-test so a win is statistically a win, not a lucky week.

And because dolv is a command center, not a dashboard, the diagnosis flows straight into action. Spot the MOFU leak, and the grounded AI can prepare the fix — a comparison guide, a nurture sequence, a multi-agent campaign run by a Director — and queue it for your sign-off. Agents do this with guardrails: each has a defined role, a $250/month budget cap and a full run history, and anything that publishes externally is held in the human-in-the-loop Approvals inbox — prepared → approved → executing → done. Internal, reversible work runs immediately; anything public waits for your approval.

The result lands back in the funnel score, so the next baseline reflects the change — that is the execute-and-measure loop. It is why our take on whether AI can run marketing campaigns on its own is "mostly yes — with a human gate," and because every draft is grounded in your company context, the output sounds like you, not a generic model. More on that in how to keep AI on-brand.

Funnel vs pipeline: a quick clarification

One last point that trips people up. The TOFU/MOFU/BOFU funnel describes buyer intent across the whole journey; a sales pipeline describes deal stages once a lead is qualified. They overlap at BOFU but are not the same thing — most of TOFU and MOFU happens before sales is involved. The funnel maps awareness, consideration and conversion; the pipeline picks up at the opportunity and runs through proposal to closed-won in the CRM.

Because dolv includes a full CRM alongside funnel intelligence, you do not stitch these two views together by hand. TOFU/MOFU/BOFU and the revenue they produce share one source of truth, and multi-touch attribution across five models ties closed-won deals back to the TOFU and MOFU touches that started them. If you want crisp one-line definitions to share with the team, the dolv glossary defines each stage in plain English.

TOFU, MOFU and BOFU are not buzzwords — they are a map of intent. Learn the three stages, feed each with real data, and score them as one number, and you stop guessing where the funnel leaks. Map every tool to a stage, let correlations point at the leak, and close the loop with a human on the gate. That is the operator's version of the funnel. dolv it.