Blogs / What Should a Weekly Marketing Report for a Client Include?

What Should a Weekly Marketing Report for a Client Include?

Nov 18, 202614 min read
Pulkit Khurana

Founder, SproutMe

A line drawing of a single cog gear, illustrating what a weekly marketing report template should include to run an account.

Every Friday your team exports numbers from Google Ads, Meta and Analytics into a deck. The client scrolls straight past it to one cell and asks why it moved. The deck cost hours and answered nothing, and the usual fix, adding more charts, only buries that cell deeper.

A weekly client report should be a short decision document, not a shrunken monthly report. It carries four things: pacing against budget, the one outcome metric the client is judged on against an agreed target, what changed and why, and what you need from them. Breakdowns and verdicts on performance belong in the monthly, because a week rarely holds enough data to judge them.

What job does a weekly report actually do?

A weekly report runs the account. It doesn't judge it. Those are two different jobs, and most bloated client reports come from forcing one document to do both.

A week is the right interval for operational calls that can't wait a month:

  • budget pacing
  • search terms that need excluding
  • a creative that has stopped pulling its weight
  • a form that quietly broke on Tuesday

A week is usually the wrong interval for deciding whether the marketing works. Most accounts don't log enough conversions in seven days to separate a real shift from noise.

So the answer to "weekly or monthly?" is nearly always both, at different weights. The weekly note covers pacing, changes, needs and decisions in a few lines a client can read on their phone. The monthly report judges results against the goal and recommends what happens next.

Spend, sales cycle and data volume decide the balance:

  • Modest spend. A local service business on a small budget needs a couple of lines a week.
  • High spend in peak season. A large e-commerce account in its biggest weeks can justify weekly calls and daily pacing alerts.
  • Long B2B sales cycle. The weight shifts to the monthly, where windows are long enough to compare absolute counts rather than percentages built on tiny samples.

The full case for setting that balance, and agreeing it with the client at onboarding, is in weekly or monthly reporting.

Here is a weekly marketing report template that holds across most accounts:

  • Pacing: [spend to date] of [monthly budget]. On course to land [on / under / over].
  • Headline: [outcome metric] at [figure] over a rolling 30 days, against a target of [figure].
  • What changed: [bid, budget, audience or creative change], because [reason].
  • What we need: [approval, asset, access, or a read on lead quality from sales].
  • Decisions: [what was agreed], and [what happens next, by when].

The last line does the most work. Without a running decision log, the weekly report becomes a record of activity. Activity is the least persuasive thing you can send a client.

Notice what the template leaves out: trend commentary, channel tabs and verdicts. If a number moved, say so in a line. Hold the interpretation until there's enough data to support it.

The headline line is where the cadence question meets the metrics question, and it has to be settled before the first note goes out. Which outcome metric, measured where, against what target? Those aren't reporting decisions. They're agreements with the client.

What must you agree before the first report?

Value isn't proven in a report. It's defined with the client before the campaign starts, and the report checks progress against that definition. This is where most agency client reporting fails, quietly.

You show a strong return on ad spend while the client's finance director judges marketing on new customers acquired. You're right by your measure and losing by theirs. No amount of chart design fixes a definition nobody agreed to.

Start from what your contact is held accountable for internally. That might be revenue, qualified pipeline, new customers or a lower acquisition cost. Work back from it to the marketing metrics that lead there.

Then ask who else reads the number. Finance may count revenue net of returns, and sales may count only the opportunities it accepted. If your "leads" figure doesn't match theirs, it gets discounted every time it reaches the boardroom.

Put the agreement on one page:

  • the business goal
  • the primary KPI, plus two or three supporting metrics
  • what counts as a conversion, in writing
  • which system is the source of truth for each number
  • the attribution view you'll report on
  • what good looks like, and by when

Two lines on that list carry extra weight.

The conversion definition matters beyond reporting. Automated bidding on Google and Meta optimizes toward whatever you've told it a conversion is, so a loose definition damages performance, not just the trend line.

The source of truth matters because the systems disagree. Google Ads, GA4 and the CRM will report different conversion counts every month. Decide at kickoff which one wins. Otherwise every review becomes an argument about whose data is right, rather than what it means.

Attribution belongs on the page for the same reason. Last-click credit flatters branded search and retargeting. It undercounts the prospecting, generic search and paid social that created the demand in the first place. If you need a second model to tell that story, agree it now. Switching to a kinder model halfway through reads as moving the goalposts.

Once this page exists, the weekly report has a spine. Every number in it should trace back to a line on the page. Every movement gets a sentence of plain commentary covering four things:

  • what changed
  • why
  • what it means for the business
  • what you're doing about it

Report the test that failed alongside the one that worked, and end with the decision you need. That commentary, more than any chart, is what the client is paying for.

The longer treatment, including how to show upper-funnel value that attribution misses, is in showing clients the value of your marketing work.

Which KPIs belong in a weekly client report?

Most lists of marketing KPIs for clients are far too long for a weekly note. What works is a short stack, layered by job.

The headline: three numbers. Spend, conversions and one outcome ratio answer whether the budget bought what the client is paying for. The ratio is cost per acquisition or cost per lead for lead generation. For e-commerce accounts that pass order values back to the platforms, it's return on ad spend. Pick it by what the account sells, not by which one flatters the month, and never show both as equal headlines.

ROAS carries a caveat clients misread constantly: it measures revenue, not profit. A product with a 40% gross margin needs 2.5x ROAS just to break even. A 3x return that looks healthy may barely cover costs. Show the break-even line next to the figure, and the ratio becomes something the client can actually judge.

The diagnostics: shown only to explain a change. Below the headline sit click-through rate, cost per click and conversion rate. On Meta, CPM is the more useful cost signal than cost per click. These aren't KPIs in their own right, and they're most useful read together:

  • Strong CTR, weak conversion rate: the landing page isn't delivering what the ad promised.
  • Rising CPC, steady conversion rate: auction pressure is pushing up costs. The account isn't broken.

In a weekly note, diagnostics usually collapse into a single line. They move up only when they explain a change in the headline.

One platform-specific metric each. On Google, that's search impression share, split by cause. Share lost to budget is a client decision. Share lost to rank is an account problem. On Meta, it's frequency. Rising frequency alongside rising CPM and falling CTR is the signature of creative fatigue.

What stays out of the headline. Reach, impressions, likes, video views and Quality Score describe activity, and they can rise while sales fall. The exception is a genuine awareness campaign, where reach is the outcome. Say so explicitly, so the client doesn't read it as a consolation prize.

The rule that matters most across channels: never add Google's and Meta's conversions together. Each platform counts conversions under its own attribution model and lookback window, and both will claim the same sale.

Report each platform's figures as that platform's view. Then put one blended figure above both, taken from the store or CRM:

  • E-commerce: total revenue divided by total marketing spend, often called MER.
  • Lead generation: CRM leads divided by total spend.

That's the number that should reconcile with the client's own books. It's also the practical answer to the source-of-truth line on the one-page agreement.

For comparison, use the account's own history before reaching for published benchmarks. Most benchmarks are compiled from agency and vendor client bases, and they rarely define a conversion the way your account does.

The platform-by-platform breakdown is in which KPIs to report for Google and Facebook Ads.

How much Google Ads detail should it carry?

Less than a dedicated Google Ads report, and the difference is worth being deliberate about. A full Google Ads report has five parts:

  1. a one-page summary of what changed and what's next
  2. outcome metrics against target
  3. a campaign breakdown
  4. the segments that explain any movement
  5. a tracking note whenever measurement changed

That structure suits a monthly deliverable. A weekly note borrows only some of it.

What carries over every week is the tracking note. Any change to how conversions are counted breaks the comparison with the period before. That covers:

  • a new primary conversion action
  • a switch from counting "one" conversion to "every"
  • a new attribution model
  • a broken tag

A client who isn't told reads the break as a change in performance. That line belongs in the weekly note the week it happens, not in next month's appendix.

Actions carry over too. If you added negative keywords after finding queries that spent and never converted, that's a "what changed" line.

What doesn't carry over is the judgment. Automated bidding needs conversion volume to settle. A seven-day view swings on a handful of conversions, and reacting to it usually makes a campaign worse. That's why the template frames the headline against a rolling 30 days.

Conversion lag pushes the same way. Google Ads records conversions against the date of the ad click, so the last few days of any period keep filling in after the report goes out. Unless you say so, the most recent week will almost always look like the weakest. Where the business is seasonal, add a same-period-last-year comparison, or every post-peak week looks like a collapse.

There's a genuine split on how much attribution to show. One view says a second model belongs in the report wherever it changes the story. The other wants one consistent model for headline numbers, with comparisons only when they change a budget decision. In practice the two meet in the same place: one agreed model for the headline, and a second as context only when it moves money.

Branded search needs similar care. Branded conversions often come from people who were already looking for the client, and they can flatter the whole account. Pair in-platform ROAS with total revenue against total ad spend.

Campaign tables, device and location splits, and Quality Score and ad strength belong in the monthly, or in a live dashboard the client can open. The test for each table is whether it changes a decision this week.

The full structure, from the first page to the tracking note, is in what a Google Ads report should include.

What if this week's numbers are bad?

Then the weekly report is the wrong place for the client to find out. A drop they discover in a scheduled document arrives as two problems: the drop, and the fact that you didn't mention it.

When you see a material change, send a short note or make a call. Give the plain number and say when you'll know more. The report then confirms what you already said.

"Material" needs a definition, because two good pieces of advice pull against each other here. Telling the client early is right. So is the warning against turning every weekly wobble into a crisis call.

The way through is the trigger agreed alongside the KPIs, such as cost per acquisition sitting outside an agreed band for two consecutive weeks:

  • Inside the band: the movement is a line in the weekly note.
  • Outside the band, or a confirmed fault like a broken tag: the client hears about it before the report arrives.

Before you say anything, work out whether fewer people inquired or fewer inquiries were recorded. Put platform-reported conversions next to the leads in the client's CRM for the same period.

If the CRM held while the platforms fell, the problem is measurement. Common causes include a republished tag container, a reworded consent banner, Safari's tracking prevention and ad blockers.

If both fell together, demand or delivery changed. Look for a budget cap hit early or a bid strategy reset into learning. Then check the client's side for a price change or a sales team that stopped answering the phone.

Don't cut spend reflexively while you check. If it turns out to be a tracking gap, a cut starves a channel that is still producing leads you simply can't see.

When you tell the client, lead with the number in one plain sentence. A reason given before the result sounds like a defense being prepared. Then sort what you know into three groups and say all three:

  • Confirmed: what you know for certain, including any cause that's yours.
  • Likely, still testing: what you suspect and how you're checking it.
  • Ruled out: what you've already eliminated.

The ruled-out list shows the client the ground you've already covered. If the cause is yours, put it in the confirmed group without hedging. Close with a next step specific enough to check, with an owner and a date.

Telling early depends on seeing early. Manual reporting tends to surface a drop only while someone is compiling Friday's deck. Companion's daily brief flags what changed in spend, performance and pipeline across connected Google Ads, Meta, Analytics and HubSpot data. A decline shows up the day it starts, not the day the report is due.

The full script, from diagnosis to follow-up, is in explaining a drop in leads to a client.

Conclusion

A weekly client report earns its place by being the document that runs the account. It carries five things:

  • pacing against budget
  • one agreed outcome metric on a rolling window against target
  • what changed and why
  • what you need from the client
  • a running log of decisions

It rests on definitions settled before week one: the conversion, the source of truth and the attribution view. It reports platform figures as each platform's view, under one blended number from the client's own systems.

Bad news travels ahead of the report, not inside it. The monthly report does the judging the week can't support.

Build it that way and the client stops scrolling for the one cell that moved, because the note already told them why.

See how Companion builds cross-channel performance reports on demand from live Google Ads, Meta, HubSpot and Shopify data, so the weekly note starts from current numbers.

Frequently Asked Questions

Keep the structure fixed and let the metrics vary. A client reporting template with pacing, the headline outcome against target, what changed, what you need and decisions works for almost any account. A layout that stays identical week to week teaches the client where to look. The outcome metric itself should follow the goal: cost per lead for lead generation, ROAS for e-commerce.

Report leading indicators, and label them as leading indicators. These include rankings for commercially valuable terms, organic traffic to pages that convert, click-through rate from search results and the first organic leads. SEO results lag effort, so a single week rarely shows revenue movement. At the outset, give the client a realistic window for when organic leads should start to appear.

Automate the data collection and keep the interpretation human. Most of the hours in manual reporting go into exporting and reconciling numbers across platforms, which is work a machine does well. Automated reports with no commentary tend to go unread, though. The weekly note earns its place through a plain-language explanation of what changed and the decision you need.

Grow smarter with AI marketing tips

Join our newsletter to get practical insights, automation ideas, and performance tips straight to your inbox.

Get a complimentary audit to uncover AI opportunities hidden in your data.

Put these strategies to work