I watched a startup burn $40K in three weeks chasing "optimized" campaigns.
Their dashboards showed beautiful charts: CPCs dropping, impressions up, conversion rates "trending positive."
They had a dashboard for every channel and a metric for everything. But they had zero conviction about what to do next.
So we did something borderline old-school: close the dashboards, open a spreadsheet, pull the data, and find the story the data wants to tell.
A few hours later, the real story emerged. Two campaigns were subsidizing eight others that looked great on the surface.
Their best-performing creative was buried, and they were optimizing for clicks when they needed revenue.
That's why I am writing this. You don't need another tool to tell you what's happened. You need a way to see what's actually happening and decide what to do next.
Let's first address the elephant in the room.
In an age of AI-powered analytics and real-time dashboards, why are we talking about spreadsheets?
Flexibility is king. Unlike rigid dashboard templates, spreadsheets adapt to your unique business model, campaign structure, and KPIs. You're not forced into someone else's idea of what metrics matter.
Cost-effectiveness speaks volumes. While premium analytics tools can cost $500-2000+ monthly, Excel or Google Sheets costs virtually nothing and handles 90% of what most businesses need.
Complete data control. You decide what data to include, how to slice it, and what calculations matter most. No black-box algorithms making decisions for you.
Dashboards are great for monitoring. Spreadsheets are great for deciding.
Dashboards celebrate rising bars and green arrows. They rarely ask if those arrows point to pipeline.
A spreadsheet forces uncomfortable truth because you trace the money, line by line:
Spend ---> Click ---> Lead ---> Opportunity ---> Revenue
Suddenly that campaign with the amazing CTR? You see that It's generating leads that never convert. That "expensive" channel? It's your only source of high-intent buyers.
Here is what matters: can you explain, in two sentences, what worked and what didn't? If not, you're optimizing blind.
Forget vanity metrics. Use a simple schema you can maintain weekly.
Campaign basics
The money trail
Truth tellers
Context (where insights hide)
Now add one calculated field: Cost per Real Outcome (CPQL or CAC). This single number will destroy most of your assumptions about "what's working."
You can drop this into a preset sheet: [coming soon].
1) Ignore rising bars and green arrows. Find the trail.
Sort by Revenue/Spend, not CTR. Circle the top three. Those are your real winners. Yes, even if their CPL looks ugly.
Do this: rank by Revenue/Spend ratio. Circle the top 3 or label top 3 = Scale.
2) Follow the drop-offs
Where do prospects disappear? From click to lead? Lead to qualified? Qualified to opportunity?
Each leak implies a different fix.
Do this: Calculate conversion rates (CVRs) between each stage.
If clicks ---> leads is strong but leads ---> qualified is weak, your targeting is right but your qualifying/offer is wrong.
3) Pattern-hunt across winners and losers
Your best campaigns share DNA. Maybe it's audience type. Maybe it's message angle. Maybe it's timing. Find the pattern, then double down. Your job is to name it and replicate it.
Do this: Group your top 5 and bottom 5 campaigns.
What do the winners have in common that the losers don't? (Hook, type, angle, LP promise, audience intent.)
4) The 2-minute story test
If you can’t tell your CEO—in two minutes—what happened, why, and what’s changing, keep digging.
Campaign A: Google Search - "Marketing Analytics Software"
Spend: $8,500 → Clicks: 2,100 → Leads: 105 → Qualified: 8 → Revenue: $24,000
Cost per qualified lead: $1,063
Campaign B: LinkedIn - "Stop Drowning in Marketing Data"
Spend: $4,200 → Clicks: 320 → Leads: 28 → Qualified: 12 → Revenue: $48,000
Cost per qualified lead: $350
The dashboard loved Google's volume and the $80.95 ($8,500/105) CPL.
The spreadsheet revealed LinkedIn's message-market fit generated qualified leads at 1/3 the CPQL and 2x revenue per dollar.
Decision: Kill the generic Google keywords. Triple LinkedIn budget. Test the "drowning in data" message angle across other channels.
Result: 40% increase in qualified pipeline with 20% less spend.
1) Sample-size myopia. Don't make budget shifts based on 10 conversions. Wait until you have statistical significance, or at least enough data to trust the pattern. Set thresholds or confidence bounds.
2) Attribution blindness. Your "best" campaign might be riding the coattails of brand awareness built by your "worst" campaign. Look at the bigger picture before cutting anything.
3) Metric tunnel vision. Optimizing cost per lead? Great. But if those leads never buy, you're optimizing toward bankruptcy.
4) Perfect-data paralysis. Your data will never be perfect. Missing attribution happens. Tracking breaks. Make decisions with 80% certainty rather than waiting for 100% clarity that never comes.
I love spreadsheets, but they are not magic.
Consider graduating to proper analytics when:
Most companies make the mistake of graduating too early from spreadsheets. So master spreadsheet thinking first.
If you can't complete this audit, your data collection needs work.
If you can complete it but the story isn't clear, your strategy needs work.
It's better decisions.
One client found their "worst" source (organic social) was actually their best revenue driver. First-click made it look weak, multi-touch told the truth. They shifted budget toward social content and saw +60% pipeline.
Another discovered trade-show leads with 10x LTV vs "cheap" paid leads. They doubled their event budget and improved their retention rate.
The spreadsheet didn't create these results. The thinking did.
You already have the data. You already have the data and the tools. Ask better questions, follow the money, and make the calls you can defend.
Ready to see what your data is really telling you?
Upload your GTM spreadsheet to get Signals.
We'll show you the patterns you're missing and the decisions waiting to be made.
Because the best marketing insights aren't hiding in a new dashboard. They're hiding in the data you already have.