Reporting
How to report on a B2B lead generation campaign week by week
Weekly reporting only earns its keep if it changes what happens the following week.
8 min read
Most weekly reports on a B2B lead generation campaign are status updates dressed up as analysis. They tell you what happened. They rarely tell you what to do differently, and if a report doesn't change next week's plan, it's not reporting — it's a newsletter.
The difference is structure. A report built to drive action looks the same every week, covers a fixed small set of numbers, and ends with a decision, not a summary.
Why weekly, not monthly
Monthly reporting on outbound feels more mature, but it hides the exact information you need most: which week something broke. A domain that starts landing in spam, a list segment that stops replying, a new rep who isn't following up on handoffs — all of these show up as a dip in a specific week. Roll that into a monthly average and the dip disappears into noise, and you find out about the problem a month after it started costing you pipeline.
Weekly reporting is more work. It's worth it because the cost of a broken campaign compounds daily, and the fix is usually cheap if caught in week one and expensive if caught in week four.
The core loop
The loop only has value if step F actually happens. A lot of teams do A through E competently and then send the same messaging, the same list, and the same cadence the following week regardless of what the diagnosis said.
What belongs in the weekly view
Keep it to numbers you can trace to a cause. For a typical b2b lead generation process running multi-channel outbound, that's:
| Metric | What it tells you | Acceptable weekly swing |
|---|---|---|
| Contacts touched | Whether volume matches plan | Low — should be stable |
| Bounce rate | Deliverability health | Should trend down or flat |
| Positive reply rate | Message and targeting fit | Some noise expected |
| Qualified conversation rate | Whether positives clear real criteria | Watch trend over 3+ weeks |
| Meetings booked | Near-term pipeline signal | High noise at low volume |
| Handoff acceptance | Sales' view of lead quality | The number that catches self-grading |
Six rows. If your report has thirty, it's not being read closely by anyone, including whoever built it.
Reading a single week correctly
A single week of data is mostly noise if you're running an effective b2b lead generation program at normal SMB or mid-market volume. One large account replying, one rep on vacation, one list segment that happened to skew toward a slow-moving vertical that week — any of these can move a percentage point by itself. Don't restructure a campaign off a single bad week. Do write it down, because three bad weeks in a row that each looked explainable in isolation is usually one real problem that got explained away three times.
The way to tell noise from signal is the 4-week rolling average next to the weekly figure. If the weekly number moves but the average doesn't, it's probably noise. If both move in the same direction two weeks running, look closer.
Diagnosing where the funnel broke
Every b2b sales lead generation funnel narrows in the same order — touched, delivered, replied, qualified, accepted — and a drop at any stage has a different fix than a drop at the next one:
- Fewer delivered relative to touched: deliverability problem, not a messaging problem. Check bounce rate and sending domain reputation before touching copy.
- Fewer replies relative to delivered: targeting or message problem. The list may be wrong, or the offer isn't landing.
- Replies up but qualified conversations flat: you're getting attention from the wrong accounts. Tighten the ICP before writing more copy.
- Qualified conversations up but handoff acceptance down: a definition mismatch between what marketing calls "qualified" and what sales is willing to work. This is a conversation, not a metric problem.
Teams lose the most time treating the wrong stage. Rewriting subject lines when the actual issue is a stale list, or expanding volume when the actual issue is deliverability, are the two most common versions of this mistake.
What to bring to the weekly review, besides numbers
Bring five replies — read verbatim, not summarized. Numbers tell you that qualified conversations dropped; they don't tell you that three prospects in a row said "we just signed with someone else" or "not now, check back in Q3." That kind of pattern shows up in language weeks before it shows up as a clean trend line, and it changes what you do next far more than the trend line does.
Putting it together
A weekly report that works has three parts, in this order: the table of six numbers with a 4-week average column, a one-paragraph diagnosis of the one thing that moved most, and a single stated change for next week. Everything else — the charts, the channel breakdowns, the historical comparisons — is useful as backup material, not as the thing people read on a Monday.
If you're running this cadence in-house and it's eating a disproportionate amount of a marketer's week, that's usually a sign the campaign itself needs restructuring, not just the reporting. LeadsLogik runs outbound as a managed program with this reporting cadence built in from week one — see /services, or check /outbound-fit if you're deciding whether managed outbound makes sense for where you are right now.
See how this applies to your pipeline.
Take the short outbound fit assessment and get a straight read on your setup.