Your H1 Attribution Post-Mortem

Most marketing teams plan H2 with H1's plan still running.

They roll the existing channel mix forward and call it a budget update. That's how the dead channels stay funded. The mix gets inherited instead of audited.

H1 has 6 months of real performance data against a plan built before the data existed. The mid-year audit is the structural opportunity to recut the mix against what actually happened.

The audit framework is 3 questions. Each one removes a category of attribution error. The output is one spreadsheet that the H2 budget meeting signs off on, with the data underneath the call.

Run it the last week of June. Recut the mix the first week of July. Have the H2 budget meeting against a clean slate.

The H1 attribution post-mortem is the work that makes H2 planning honest.

Key Takeaways

• H1 audits run the last week of June, against the first 6 months. The output feeds H2 planning the first week of July.
• Question 1: Which channels are double-counting credit? Channel-claimed credit usually sums above 100% because per-channel dashboards report full credit on each touch.
• Question 2: Which channels can't survive losing last-touch credit? The ones whose numbers crater under first-touch were riding the funnel.
• Question 3: Which channels have zero win/loss data? A channel with no attributable conversions after 90 days is a sunk cost.
• The kill list is the output: three columns, three decisions per channel (keep, instrument, kill). The data is the recommendation.

The Mid-Year Audit

If you're still running the channel mix you planned in January, you're running blind into H2.

H1 has 6 months of real performance data against a plan that got built before the data existed. The mid-year audit is the structural opportunity to recut.

I've watched marketing leaders kick the attribution audit to Q3 because H1 numbers were on plan. By the time Q3 closes, the cut decisions for the back half of the year are already made. The team is committed to a mix that hasn't been audited since launch.

The mid-year audit's question is which channels actually contributed. The H1 number is the table stakes; the audit reveals which channels deserve to claim it.

The timing matters. Run the audit the last week of June, after H1 closes but before H2 budgets lock. The first week of July is the budget meeting. The audit feeds the meeting; the meeting signs off on the recut.

I've seen teams run this audit annually instead of semi-annually and lose a quarter of optimization time. The H1 view is fresher than the prior year's view. Six months of recent data beats 12 months of mixed data every time.

The audit is also faster than people think. The 3 questions take a week with a marketer and an analyst.

The hard part is acting on the output. Running the audit itself is mechanical.

Question 1: Double-Counting

The first question in an H1 attribution audit is which channels are double-counting credit.

Every multi-touch attribution model assigns fractional credit across the touches in a conversion path. Per-channel dashboards usually report each touch at 100%. Add the dashboards together and your total channel-claimed credit exceeds 100%.

I've audited reports where the sum of channel-claimed credit was 140% of actual revenue. The 40% over was double-counting. Two channels both claimed the same conversion at full credit because each one was the last touch from its own dashboard's perspective.

The fix is one rule: every conversion has exactly 100% of credit to distribute. If channels A and B both claim 60%, one of them is lying. The audit picks the model (last-touch, first-touch, linear, time-decay) and applies it consistently across all channels.

Pick one model. Stick with it for the audit. The argument about which model is best happens after the audit closes.

What the double-counting question reveals is which channels look bigger than they are. Once the 140% gets normalized down to 100%, the channels that were taking inflated credit shrink. The channels that were giving credit (the demand-generation channels) grow.

The shrinkage hurts the channels that look strong on the dashboard. It also feels uncomfortable for the team that runs those channels. Their performance numbers come down, and the budget allocation conversation gets harder.

That's the audit working. The pre-audit numbers were wrong, and the team was making spend decisions on wrong numbers.

Question 2: Last-Touch Removal

The second question is which channels can't survive losing last-touch credit.

Strip last-touch credit from your H1 report. Recount each channel's contribution. The ones whose numbers crater were riding the funnel. The ones whose numbers hold up were actually driving demand.

I've watched paid-search budgets stay flat for 4 quarters because the team gave paid search 80% of the credit on a last-touch model. Stripped to first-touch, the same channel showed 12%. The category being searched was created by the LinkedIn campaign 6 months earlier.

Last-touch is incomplete. It catches the final click. The demand it captures was created somewhere upstream.

Run the audit with last-touch and again with first-touch. The gap between the two is the demand-generation channel you're under-crediting.

This question is the hardest one for paid-search and SEO teams to hear. Both channels live at the end of the funnel and look great under last-touch. Under first-touch, they often shrink dramatically.

The right read is that paid search closes demand that other channels create. The dependency was hidden by the last-touch model. Once the audit exposes the dependency, the team can decide whether to fund the demand-creation channel that's been carrying paid search invisibly.

The mid-funnel channels (display, content, organic social) usually win the most credit under the recut. They've been doing the demand-creation work without getting credit for it because the last-touch model gave the credit to whoever closed the conversion.

The audit's job is exposing what the model hides so the team can budget against the full picture. The attribution model in production can stay the same.

Question 3: No-Data Channels

The third audit question is the one most teams skip: which channels have zero win/loss data?

The dashboards show spend. They show no conversion events tied back to the channel.

I've audited channel mixes where 2 of 6 line items had been running for 3 quarters with no measurable conversion event tied back. Just spend, going out. The channel kept its line in the budget because no one could prove it should be cut.

A channel with zero attributable data after 90 days is a sunk cost. The line in the budget says marketing; the data says expense.

Either instrument it before H2 starts, or kill it. Standing still is the choice that costs the most.

No-data channels usually fall into a few categories. The legacy channel that's been running since before measurement was set up. The experiment that never got proper UTM tagging. The brand-budget line that lives outside performance accountability.

Each one is recoverable. UTM-tag everything inbound. Add conversion tracking on the destination pages. Set up the analytics events that didn't exist when the channel launched. Most no-data channels can be instrumented in a week.

The instrument-or-kill decision forces honesty. A channel worth instrumenting commits to a 90-day window with proper measurement. A channel without that justification comes out of the budget.

The audit gives every no-data channel the choice. Instrument by Q3 start, or come out of the budget. No third option.

The Kill List

The output of the H1 attribution audit is one spreadsheet. The kill list.

Three columns: channel name, H1 attributable revenue, H2 decision. The decision is one of three: keep, instrument, kill.

Keep: channels with attributable revenue above your CAC threshold. Instrument: channels with spend but no data, given one quarter to prove themselves. Kill: channels that failed the instrument-or-kill test from last year's audit.

I've watched marketing leaders take this list into the H2 budget meeting and walk out 30 minutes early. The conversation moves from "what should we cut?" to "here's what we're cutting, here's the data, sign off."

The kill list is a decision-ready artifact. The data underneath is the recommendation. The meeting signs off on it.

The list also creates a paper trail. The reasons each channel got its decision are written down. Future audits start from the prior audit's decisions, which compounds the discipline.

A typical first-time audit will move 4-8 channels into different statuses. A team that runs audits semi-annually for two cycles ends up with a stable mix where most channels have a clean decision and only the experimental ones need fresh reviews.

The kill list also surfaces channels that don't show up in the regular budget conversation. The dark-pool channels (vendor co-ops, affiliate spend, partner programs) sometimes have no clear line items and show up in the audit for the first time as no-data channels. That's the audit working.

The output is the decision, made.

What this changes about your H2 budget meeting

H2 budget meetings without the audit run the same script every year. The team defends the existing mix, the CFO asks where the cuts are, marketing names a small percentage that doesn't change the underlying mix, and the meeting ends.

With the audit, the meeting opens with the kill list. The cuts are pre-decided. The conversation moves to where the freed budget goes, which is the real conversation.

For now, the H1 attribution post-mortem fits in 3 questions and one spreadsheet. Run it the last week of June. Use it the first week of July. Plan H2 against the recut, not the inherited mix.

Further Reading

On Professor Leads

• Your Attribution Model Is Lying to You is the upstream piece on how attribution dashboards get the channel-health question wrong.
• Incrementality Without a PhD is the measurement framework that gives the audit's recommendations causal weight.
• When to Kill a Channel covers the kill-decision discipline in more depth, including the wind-down plan for channels that exit the mix.

On Forbes (by William DeCourcy)

• Why Chasing Metrics Is Killing Your ROI (And How To Fix It) is the bigger argument about which metrics actually drive business outcomes.
• Your Marketing Isn't Broken; Your Lead Generation Strategy Is covers the diagnostic framing that the audit framework draws from.