How to build a one-page monthly marketing report that gets budget decisions made on the call.

The owner of a home services company used to receive a 14-page monthly marketing report, with charts for traffic, rankings, email opens and ad clicks, and it took most of a day to build. On every monthly call the owner skipped all of it and asked, "Did it bring in work?" The answer was on page 11, under a chart of leads by source.

I cut it to one page in four parts, with the detail further down. The calls got shorter, and budget decisions that used to wait a month got made on the call. Make the change ahead of the year-end review, which judges every month at once.

Why a monthly marketing report gets skipped, and what gets one read.

A report usually gets skipped when page one asks for no decision. Gartner surveyed 377 users of marketing analytics in May and June 2022 and found analytics influenced 53% of marketing decisions, and about a quarter said decision makers don't review what the analytics team provides. (Gartner sells research to the leaders it surveys.)

Count the decisions (a budget move, a paused campaign, a page test) from the last three monthly meetings. If most ended with no change, the report describes activity.

In Scientific Advertising (1923), Claude Hopkins wrote that "our final conclusions are always based on cost per customer or cost per dollar of sale" (chapter 1). A lead count belongs on page one only beside how many qualified and what revenue they became.

Step 1. Put every first-page number through one question.

For each number, ask "What would we do differently if this moved?" Picture it doubling, then halving. If neither would change a budget, campaign, page or email, it moves off page one.

Avinash Kaushik makes the same cut in a 2010 post drawn from his first book, Web Analytics: An Hour A Day (2007): “If at the third ‘so what’ you don’t get a recommendation for an action you should take, you have the wrong metric.”

Numbers that usually pass:

1. Cost per qualified lead (spend ÷ qualified leads) decides where the budget goes.
2. Landing page conversion rate (form submissions ÷ visits to that page) decides when the page gets a new test.
3. Reply rate on outreach (replies ÷ people who received that email) decides when an email gets rewritten.
4. Pipeline from marketing (open deal value from marketing leads) shows if the plan is working before revenue arrives.

Numbers that usually don't, on their own: impressions, which say little about who wanted anything; total sessions, which count a job seeker the same as a prospect; followers; and email open rate. Apple's Mail Privacy Protection "prevents senders from seeing if you've opened the email message they sent you."

Litmus counted over 1.5 billion opens in September 2026, and 76.56% reported as Apple, a group that includes Mail Privacy Protection opens, which Litmus doesn't consider reliable. Litmus sells email analytics and pools all its customers' sends, consumer brands included. A B2B list can differ, and your platform's client report shows your own mix.

These go below the line. A short watch list covers what warns you before something breaks: forms reaching the CRM, uptime and deliverability (Google's sender guidelines require every sender to stay under a 0.3% spam rate in Postmaster Tools and recommend below 0.10%).

At a larger company, agree on the first-page numbers with finance and sales once a year, and print the qualified-lead definition on the page.

Step 2. Build the first page in four parts.

1. New leads, and how many qualified, under one written definition, with cost per qualified lead for the last three months.
2. Revenue from deals that came from marketing leads, including leads from earlier months. Revenue from existing clients gets its own line.
3. What changed, in two or three lines, including tracking changes, which can move a number with no change in results.
4. One decision for next month, written as a proposal before the meeting, with the amount, the owner and the month it gets checked.

The decision isn't always about marketing. Eliyahu Goldratt's The Goal (1984) rests on the point that a system moves only as fast as its slowest step. If leads pile up without becoming deals, the decision may be getting proposals out within two days of a good call (the yes-to-signature guide), and more leads would only lengthen the line.

At a larger company, keep a decision log (date, owner, expected effect, result).

Step 3. Credit revenue to the month leads came in.

A revenue report shows when deals close, not which month's marketing produced them. Gartner describes B2B purchases as "looping" through six buying jobs, and a form fill can come months before the signature.

Grouping leads by the month they came in fixes it. In HubSpot:

1. Mark each contact's month with First conversion date (the date the contact first submitted a form), or Create date for contacts with no form, like imports.
2. Each month, pull Lifecycle stage, the Date entered properties for your qualified and customer stages, Original Traffic Source and Total revenue (associated closed-won deals). The custom report builder needs Professional or Enterprise, and every plan can export a saved contact view.
3. In Google Sheets or Excel, put the months in down the rows and the months after across the columns, and count clients in each cell, with qualified leads and revenue to date at the end. The original source splits any month by channel.

In Salesforce, build the grid from lead records, converted ones included, with each lead's Created Date and Lead Source, which carries over to the contact and opportunity at conversion. With Mailchimp, Constant Contact or no CRM, keep one sheet row per lead with first date, source, status and invoiced revenue.

One mistake to avoid is the most recent conversion date. A past client who fills out a new form shows up as a new lead in the current month, which flatters recent months. Total revenue also sits on each contact, which counts a deal twice when two contacts share it. When that's common, build revenue from a deals export.

Compare months at the same age. At a larger company, group by quarter and opportunity created date in a report named like "2026-q4-leads-by-quarter-in", with expansion revenue apart.

Step 4. Settle attribution with two sources that check each other.

The two sources are the CRM's first touch (Original Traffic Source in HubSpot, Lead Source in Salesforce) and the answer to "How did you hear about us?", asked on the form or the first call.

Refine Labs, a B2B demand generation agency that runs its own marketing on social media and a podcast, put a required free-text version of the question on its strategy-call form for its 2023 analysis and compared 620 conversions with HubSpot's attribution.

The gap sits in web search, the direct and organic visits that often follow a conversation elsewhere. Prospects named social media, podcasts, word of mouth and community, and the software didn't measure podcasts or community at all. A large "direct" or "organic" number points to the answers, not to a channel to fund.

Ruler Analytics, which sells attribution software, reported on 350+ demo requests in 2022 and found 72% of answers vague or missing, and 47% of one client's leads who picked the first option were attributed wrong.

Use a short list with an "Other" box and "Select one" at the top, and ask again on the first call when the answer is blank or says "Google." Refine Labs prefers free text with no options, and a list is faster to count.

When both name the same channel, campaign or partner, move the budget, and when they disagree, look closer first.

Larger companies with long sales cycles and big budgets get more from detailed attribution. In HubSpot, contact create attribution comes with Marketing Hub Professional, and deal create and deal revenue attribution need Enterprise.

Randall Lewis and Justin Rao analyzed 25 consumer field experiments at large US retailers and brokerages and found the median confidence interval on return on investment over 100 percentage points wide (Quarterly Journal of Economics, 2015). The paper still calls tests progress, because they aren't fooled by who the ads were aimed at. Before a large move, pause a channel that brings in a big share of leads, or hold out one region, and compare leads with the weeks before and the same weeks last year. A change that size usually shows in the count, even when the return can't be measured.

A worked example.

Example numbers: a services company spends $7,500 a month, $5,000 on search ads and $2,500 on LinkedIn ads. By close month, September shows $11,000, the weakest of the fall. By month in, as of early December:

September had the fewest leads and the most qualified leads and revenue, at $7,500 ÷ 16 = $469 per qualified lead against $536 in October. November is too young to judge until month three.

For September's 16 qualified leads, first touch says 9 non-brand search ads, 1 LinkedIn ad, 4 direct and 2 organic. The answers say 8 search, 1 LinkedIn, 5 a podcast interview from August and 2 a referral. Both agree search carried the month and LinkedIn ads produced one qualified lead for $2,500. Next month's decision moves $1,000 of LinkedIn budget to search, total unchanged, checked two reports later, and a larger move waits until three months in a row agree (next year's budget starts here). "Direct" points to the podcast, which gets a closer look first.

Two templates to copy.

H  Date came in:      =IF(B2="",C2,B2)
I  Month came in:     =IF(J2="check","",YEAR(H2)*100+MONTH(H2))
J  Months to client:  =IF(F2="","",IFERROR(DATEDIF(EOMONTH(H2,-1)+1,EOMONTH(F2,-1)+1,"m"),"check"))
K  Qualified (1/0):   =IF(OR(E2="Sales qualified lead",E2="Opportunity",E2="Customer"),1,0)

B2 Clients by month N:  =COUNTIFS(leads!$I:$I,$A2,leads!$J:$J,"<="&B$1)
F2 Qualified:           =SUMIFS(leads!$K:$K,leads!$I:$I,$A2)
G2 Revenue to date:     =SUMIFS(leads!$G:$G,leads!$I:$I,$A2)

Common mistakes, and the pushback you'll hear.

• Changing the definition of qualified mid-year without restating earlier months, which makes a rule change look like a trend.
• Judging the newest month too early. Set a minimum age, like three months for a short cycle.
• Counting existing-client revenue as new. It belongs on its own line (past clients and your network).

"First touch credits the first tracked visit, often a search for your name." That's often true, which is why agreement between two sources is the signal, and why search ads get judged on non-brand terms.

A test to run on your next two reports.

1. Baseline. Note the length and decisions of the last two monthly meetings.
2. Build (one afternoon). Rebuild page one in four parts, with the grid by month in.
3. Run (two meetings). Send the page two days ahead with the decision written as a proposal.
4. Measure. Decisions made on the call, days from report to decision, meeting length.
5. Result. At least one budget, campaign or page decision made on the call in both meetings, and more than the baseline. If not, give the decision line an amount and an owner.