The project in 30 seconds
- Client
- CollisionCam
- Industry
- Crash-lead referral network
- Location
- United States
- When
- February 2026
- Engagement
- Fiverr
- My role
- Zoho Analytics developer
Where CollisionCam started
CollisionCam captures crash leads and passes each one to a partner: an attorney, or a physical-therapy / medical clinic. Every lead is a record in Zoho CRM.
The data existed, but it couldn't answer simple questions. How many leads came in this month? Which counties are growing? Which partner received how many, and how many people in those vehicles signed or started treatment?
What had to be solved
The reporting had to solve three things:
How I tackled it, step by step
- 1
Fix the data model
Standard picklists for lead source, county, destination type and partner, plus a clear lead-status path.
- 2
Define the metrics
Lead counts and people counts (sum of people in the car), never substituted for each other.
- 3
Build four dashboards
Macro, location, attorney performance and PT/clinic performance, with filters that update every chart.
- 4
Refine with the client
Three rounds of feedback, worked through line by line.
Inside the system
9 real screenshots from the build. Click any image to enlarge.
01The macro view leadership asked for
Year-to-date, 30-day, 7-day and 24-hour lead totals, leads by month, leads won and lost, a pipeline funnel with conversion rate, and the split of leads across affiliate partners.
02Where the leads come from
Leads by county and city, attorney-vs-clinic comparison tables by city, state and county, and a treemap of every city. Date, county and state filters update everything at once.
03How each partner is performing
The attorney dashboard shows leads sent, leads signed and people involved, by county and month. The PT/clinic dashboard mirrors it with treating instead of signed. Both can be filtered to a single partner.
What changed for CollisionCam
The calls that made it work
Good systems come from good decisions. Here's what I chose, and why.
Keep lead counts and people counts separate
A car with four people is one lead but four potential clients. Partners are paid on people, so every people metric sums "people in car" and never reuses lead counts.
Standardise counties as a picklist
Free-text counties split one place into several. A picklist keeps totals accurate.
Filter-first design
Partners and leadership look at the same dashboards through different filters, so there's one source of truth.
"Very talented and will use again!"
A similar system could help ifโฆ
- Your data is in Zoho CRM but reporting means exports and spreadsheets.
- You report to partners, franchisees or locations and each needs their own view.
- You need numbers you can defend, with clear definitions.
Dashboards show aggregate numbers only. Every screenshot on this page comes from the real project; nothing is a mock-up.