Data Analytics Consulting
The shortcut: Most data consultants get hired to build dashboards nobody opens — because they never asked which decision the dashboard was supposed to change. Sell decisions, not visualizations. Every engagement starts by naming the one weekly meeting where this answer gets used, and the dashboard is built backward from that.
Industry: Software & Tech | Investment level: Small — $2,000-$10,000 | Time to launch: 4-8 weeks (one tool stack picked + two free pilot dashboards + first paid scoping call gate the launch)
Best for: Someone who's already lived inside SQL, a BI tool (Tableau, Power BI, or Looker), and at least one operational system (Shopify, HubSpot, NetSuite, Stripe). You can read a P&L, push back on a vague question, and explain a number to a non-technical CFO without making them feel small. What you'll likely make: $1,500-$3,000 month 3, $5,000-$9,000 month 6, $9,000-$15,000 month 12 (one fixed-fee project plus a small monthly reporting retainer book). Math is in Section 4.
Market Opportunity
It's Tuesday morning. The CFO of a 40-person e-commerce company needs Q3 numbers for a board call on Thursday. Her finance person is exporting Shopify orders to a CSV, pasting them into a spreadsheet, and pivoting against ad spend from another tab. Last quarter the numbers didn't reconcile until 9pm Wednesday. The CFO knows there has to be a better way and doesn't know what to Google. She knows somebody who knows somebody who "does data." That somebody is you.
That moment — every SMB that grew past 20 people without a finance ops hire — is the entire business. The crowded version is "we build dashboards." The quiet version is the analytics translator: someone who sits with a non-technical owner, names the three decisions costing them sleep, and ships a Looker Studio or Power BI view that answers exactly those questions.
The other wedge is the monthly reporting retainer. Dashboards rot — schemas change, ad accounts get renamed, the CEO wants a new metric. SMB clients pay $500-$2,000/month to keep the numbers honest. Eight retainers at $1,000 is $8,000/month before any build lands. Demand is not the problem. Scope discipline is.
- Tableau Creator is $75/user/month; Power BI Pro is $10/user/month — the cost gap shapes every tooling recommendation Tableau | Power BI.
- dbt Cloud starts at $50/month per developer seat (free OSS Core tier exists); Snowflake and BigQuery clients increasingly expect dbt fluency dbt pricing.
- HIPAA-covered entities cannot use Google Analytics or Looker Studio without a BAA — Google offers neither — a real consulting gap for healthcare-adjacent SMBs HHS HIPAA.
- SMB project pricing reality: dashboard builds $2,000-$8,000; monthly retainers $500-$2,000/month; audit + roadmap for 5-50 person companies $3,000-$6,000 fixed benchmark — VERIFY.
Launch With AI
Pro section. AI doesn't replace the work — it cuts the parts that drained you (transcribing the CFO discovery call, writing the metric definition doc, drafting the dbt models, narrating the monthly retainer summary). Spend the saved time on what AI can't do: sitting with a non-technical owner watching them try to find the number, asking the VP of Ops why nobody opens the dashboard, and getting clear on the one decision that should change because of this build.
The trap most first-year analytics consultants fall into: they think AI is for shipping more dashboards faster. Backwards. The bottleneck isn't dashboard count — it's converting a 90-minute discovery call into a fixed-fee proposal the same afternoon. AI compresses the writing. The judgment about which decision the dashboard should change is yours and stays yours.
Important up-front: AI cannot interpret a P&L for you, reconcile a Shopify-to-QuickBooks variance, or take responsibility for a wrong number that ends up in a board deck. It will also confidently hallucinate dbt macros and Power BI DAX formulas — every model and every measure needs a human pass against the live data. The fee you're charging is the value of getting the answer right; that part doesn't outsource.
AI Tools You'll Use
| Tool |
Price |
What it does |
| ChatGPT Plus |
$20/mo |
Discovery → SOW, metric definition docs, board-deck narratives, retainer summaries |
| Claude Pro |
$20/mo |
Long-context schema review, dbt model writing, multi-source reconciliation logic |
| Cursor |
$20/mo |
SQL + dbt model dev, Power BI/Looker Studio config, version-controlled engagement repo |
| Granola |
$18/mo |
CFO discovery call transcripts + auto-summaries (revenue/EBITDA-heavy calls) |
| Loom AI |
free |
Dashboard walkthroughs, monthly retainer recordings, audit report deliveries |
The Workflow
Discovery call → fixed-fee SOW (Granola + ChatGPT, ~30 min/prospect). The CFO call decides the engagement. Granola records and auto-summarizes; ChatGPT turns it into a proposal before the prospect's afternoon ends. Paste:
"Below is a Granola summary of a 60-min discovery call with the [CFO/Head of Ops] at a [vertical, headcount, revenue] company. Write a fixed-fee SOW: (a) 1-paragraph problem restatement using the 3 specific decisions they named, (b) scope as either a $3K-$6K data audit OR a $5K-$8K dashboard build (whichever fits their data maturity), (c) the 3 source systems we'll integrate (Shopify/QuickBooks/HubSpot/etc.) named explicitly, (d) the metric definition doc deliverable, (e) explicit out-of-scope (machine learning, predictive modeling, custom warehousing — until phase 2), (f) optional $500-$2,000/month retainer add-on with 3-5 hour SLA. 50% deposit + remainder on go-live."
Send within 4 hours of the call. The CFO's memory of the conversation is your closing window.
Metric definition doc (Claude long-context, ~60 min/engagement). Every dashboard ships with a 1-page metric definition doc or you'll be in disputes about what "revenue" means by month two. Paste each source system's schema into Claude, then:
"Below are the schema exports from [client]'s Shopify, QuickBooks, Stripe, and Google Ads. Build me a metric definition doc covering: (a) Revenue (recognized vs cash, refund handling, multi-currency), (b) Gross Margin (which COGS lines, freight handling, transaction fees), (c) Customer Acquisition Cost (which spend channels, attribution window), (d) Customer LTV (cohort vs blended, which time horizon). For each metric: the formula in plain English, the SQL/DAX equivalent grounded in their actual table names, and the 3 most common ways teams get this number wrong. Output as a 1-page Notion-ready doc."
Walk the CFO through it line-by-line in the kickoff call. The doc gets signed before any dashboard builds. That's your liability cap on the bad-number scenario.
dbt models + SQL queries (Cursor + Claude, ~2-3 hrs/engagement). Open Cursor with Claude 3.5 Sonnet selected. The reusable starter has chart-of-accounts mapping, refund handling, multi-currency normalization. Paste:
"I'm building dbt models for [client] on [Snowflake/BigQuery]. Sources: Fivetran-synced Shopify, QuickBooks, Stripe. Generate: (a) the staging models for orders, refunds, customers, line items with consistent timestamp normalization, (b) intermediate models that reconcile Shopify orders → Stripe payments → QuickBooks invoices (flag the variance row when amounts don't match within $1), (c) a marts layer with daily_revenue, daily_gross_margin, customer_ltv_cohort matching the metric definition doc. Use dbt project conventions (snake_case, source/staging/intermediate/marts, refs not hardcodes). Code only."
Build the first set in 3 hours. The reusable framework you keep (per the IP carve-out in Section 6) means engagement 4 takes 90 minutes.
Dashboard walkthrough Loom (Loom AI, ~20 min/engagement). Final delivery is a 5-min Loom narrating the dashboard against the original 3 questions. Loom AI auto-titles, auto-summarizes, generates a written transcript. Three uses: (a) the kickoff for the retainer conversation ("here's what'll need to drift in 90 days"), (b) attached to the invoice as proof of delivery, (c) becomes the case study with permission. Same recording, triple use.
Monthly retainer summary (ChatGPT, ~30 min/client/month). The retainer is the business — eight at $1,000/mo is $96K/year of floor income. Paste the dashboard's automated stats into ChatGPT:
"I run the data retainer for [client]. Below is this month's dashboard refresh stats and the queries I ran ad-hoc. Write a 1-page client report: (a) the 3 numbers from the metric doc with this month's value and 3-month trend, (b) one variance the data is flagging that the team should look at (a margin compression, a CAC spike, an LTV cohort underperforming), (c) what I built this month (model fixes, new measures, source updates), (d) one recommendation for next month (a number to add, a process to automate, a source to integrate). Tone: senior operator who has the CFO's ear, not vendor."
Send the first business day of every month. Retainer renewals stop being conversations once the client gets a real summary.
Time Saved Per Week
Roughly 7-10 hours/week once Granola, your metric template, and your dbt starter are built:
- Discovery → proposal: 4 hours per prospect → 30 min (Granola + ChatGPT)
- Metric definition doc: 3 hours/engagement → 60 min (Claude long context)
- dbt model dev: 6-8 hours/engagement → 2-3 hours (Cursor + reusable framework)
- Final delivery walkthrough: 90 min hand-recorded → 20 min (Loom AI)
- Monthly retainer reports: 90 min/client → 30 min (ChatGPT template)
Trade that time for: the second vertical you'll add, the Snowflake or BigQuery cert that gates senior-tier work, and the case study you haven't published yet. Those compound; admin doesn't.
Total AI Stack Cost
- Budget tier ($20/mo): ChatGPT Plus only. Cursor's free tier is fine for the first 3 engagements; Claude is reachable through ChatGPT Plus's GPT-4o for most prompts; Otter.ai's free tier (300 min/month) covers your first 5 discovery calls. Right for the first 60 days.
- Full tier ($78/mo): ChatGPT Plus + Claude Pro + Cursor + Granola + Loom AI free. Worth it once you cross 3 paying clients — Claude's 200K context handles a full client schema export in one paste, which Cursor + ChatGPT can't.
- Compare: A junior data analyst doing your transcripts, metric docs, and monthly reports runs $1,500-$3,500/month. The full AI stack is one-thirtieth that cost and the operator voice your CFO clients trust stays yours.
Cancel anything you don't open in a 7-day window. The trap on the analytics side is buying 3 different BI tools — pick the one your client already pays for and stay there.
Your First Win
30 minutes from now your metric definition template is built. Open ChatGPT (free tier works for this one). Paste:
"I'm a freelance data analytics consultant. Build me a reusable metric definition template I'll send every new SMB e-commerce client. Cover: (a) Revenue — recognized vs cash, refund timing, multi-currency, taxes/shipping treatment, (b) Gross Margin — which COGS lines, freight, transaction fees, returns, (c) CAC — paid vs organic split, attribution window choice, internal-team labor exclusion, (d) LTV — cohort vs blended, time horizon, churn handling, (e) AOV vs ARPU vs ARPPU. For each metric: the formula in plain English, the 3 most common ways SMBs get it wrong, and a 1-line 'how I'd compute this in your stack' placeholder. Output as a 1-page Notion-ready doc with a 'CLIENT-SPECIFIC NOTES' section at the bottom."
This template is your liability cap on every engagement. It's also the cleanest 'free 20-min audit' lead magnet — the prospect leaves the call holding a $5K-quality deliverable, and your conversion to paid scoping calls jumps from 15% to 35%+. Build it once today; reuse it across every client for the next 3 years.
Product / Service Offering
You are selling one of three things. Pick which one is on the proposal before the first call ends.
- Data audit + roadmap — Two-week fixed-fee engagement. Inventory data sources (Shopify, QuickBooks, HubSpot, Stripe, ad platforms), interview leadership about the three decisions that hurt, deliver a 6-page roadmap with tooling and build order. $3,000-$6,000. Best entry offer — low risk for the client, high information for you.
- Dashboard build — One BI dashboard answering 3-5 questions, grounded in cleaned data, with a one-pager defining every metric. $2,000-$8,000, 1-3 weeks. The most common engagement.
- Monthly reporting retainer — You own the numbers. Refresh dashboards, send a 1-page written summary monthly, hop on a 30-minute call. Includes 2-4 hours of ad-hoc "I have a question" time. $500-$2,000/month. Bread-and-butter long-term.
Standard stack — pick per client: Looker Studio (free, default for Shopify + Google Ads on Google Workspace), Power BI (default for Office 365 SMBs), Tableau (only when the client already pays $75/seat), Snowflake or BigQuery for warehousing once data lives in 3+ systems, Fivetran for ELT pipelines (many SMBs already pay for it but never turned it on), dbt for transformations once a warehouse exists.
Do NOT pitch machine learning or predictive modeling on engagement one. SMB data volumes rarely justify it, results are inconclusive, and it erodes the trust you need for retainer #1.
Revenue Model
Unit economics for a solo consultant working from a laptop, no employees, billing through Stripe:
| Service |
Price |
Variable cost (tool licenses + Stripe 2.9% + $0.30) |
Build time |
Take-home |
| Data audit + roadmap |
$5,000 |
$0 + $145.30 |
10-15 hrs over 2 wks |
~$4,855 |
| Dashboard build (Power BI / Looker Studio) |
$5,000 |
$50 (client's tool already) + $145.30 |
1-2 wks |
~$4,805 |
| Monthly reporting retainer |
$1,000/mo |
$50 (dbt Cloud or warehouse share) + $29.30 |
3-5 hrs/mo |
~$920/mo |
| Snowflake/dbt warehouse build |
$8,000 |
$200 (warehouse credits during dev) + $232.30 |
3-4 wks |
~$7,565 |
Your first $1K month = one data audit at $3,000 with a 50% deposit = $1,500 in the door. The first close realistically takes 4-6 weeks of outreach.
Your first $3K month = one dashboard build at $5,000 (50% deposit) + one $500 retainer = $3,000 in the door. Roughly 1.5 weeks of work plus 3 hours of retainer maintenance.
The retainer is what makes this business stack up. Build clients churn once the dashboard ships. Retainer clients pay you to be the data person. Quote the retainer in the same proposal as the build, never three months later. By month 12, eight retainers at $1,000 is your floor and any new build is upside.
Startup Costs
Realistic all-in: $2,000 if you defer insurance 30 days, skip dbt Cloud, and use free tiers; $10,000 if you bind insurance year one, run an attorney MSA review, sit a senior cert, and pay a dbt Cloud seat.
Legal & Formation
Business entity. Single-member LLC the moment the first paying client signs. A dashboard that fed wrong revenue into a board deck is a much smaller problem when the lawsuit names your LLC, not your house. Filing fee is $35-$500 LLC University 50-state table. EIN is free at IRS EIN Online — never pay a third party. Once net profit clears $80K-$100K/year, run the math on an S-corp election via IRS Form 2553; for eight $1K/month retainers plus project work, that's a year-2 conversation.
Licenses & sales tax. No state professional license required for data analytics consulting. Custom analytics work billed as professional services is generally not taxable. But if you host dashboards on your own infrastructure and bill recurring SaaS-style access, about 25 states treat that as taxable SaaS state SaaS tax tracker — VERIFY. Cross $100K in sales or 200 transactions in any state and you have economic nexus there. Stripe Tax handles this for $0.50 per transaction.
Industry-specific risk. Three things will bite this specific business model and they trip people up in this exact order.
First, healthcare clients and the BAA gate. If a prospect is a healthcare provider, dental group, telehealth platform, or anyone touching PHI (protected health information), standard analytics tooling is off the table. Google offers no BAA for Google Analytics or Looker Studio. Build on a HIPAA-eligible warehouse (BigQuery with a Google BAA, Redshift with an AWS BAA), scope strictly outside PHI, or walk away. Do not promise HIPAA compliance over Looker Studio — that sentence ends your business HHS HIPAA.
Second, IP assignment vs framework retention. The client owns dashboards, queries, and dbt models written specifically for them. You retain the underlying methodology — your audit template, SMB chart-of-accounts mapping, Shopify ELT starter, reusable dbt macros. Spell it out in the SOW: client-specific deliverables assigned, reusable frameworks licensed (not sold). Without this, every engagement is a clean-slate rebuild 17 USC §101.
Third, liability cap and the bad-number scenario. A wrong number in a board deck is a bigger commercial risk than most analytics consultants think. Cap total liability in the MSA at fees paid in the trailing 12 months. Add a clause requiring the client to validate the underlying data source before relying on dashboards for material decisions. Bind a $1M-$2M E&O policy via Hiscox at $800-$2,000/year — less than one billable day at full rate.
Marketing & First Customers
Your first three clients come from people who already know you can read a P&L and write SQL. After that:
- Direct outreach to 30 SMB CFOs / heads of ops in one vertical (DTC e-commerce, B2B SaaS, marketing agencies, professional services). Sales Navigator filter: 20-200 employees, $5M-$50M revenue, "Director of Finance / Head of Ops / VP Operations" titles. Three-line LinkedIn DM with a 60-second Loom of a sample dashboard, the price band, and a 15-min call link. Target 10-15 DMs/day. Reply rate 3-7%, call-to-paid 15-25%.
- Free "data audit" lead magnet. A 90-minute call where you screen-share the prospect's Shopify, ad platforms, and QuickBooks and walk them through the three numbers that would change a decision. Converts at 20-30% to a paid $3,000-$5,000 audit. Cap at 4 free audits/month — they're expensive in your time.
- Vertical communities, not generic ones. Indie Hackers for SaaS clients, e-commerce-focused Slack/Circle groups (Operators Guild, RevOps Co-Op), local CFO meetups. One thoughtful answer per week to "how do I track X" builds inbound over 6 months.
- Upwork for the first 1-2 builds only — 10% flat fee, real inbound volume. Filter to "Power BI" / "Looker Studio" / "Snowflake" jobs over $2,000. Per Upwork's terms, move them onto a direct retainer after.
Disclose your AI tool stack upfront. If you use Cursor, Claude, or ChatGPT to write SQL or dbt models, say so on the first call — some regulated clients require disclosure or prohibit it. GitHub Copilot for Business includes commercial IP indemnification.
First 90 Days
- Week 1. File LLC, get free EIN, pick your vertical (DTC e-commerce, B2B SaaS, agencies — exactly one). List the 5 questions every CFO/COO in that vertical asks every month.
- Week 1-2. Build two free portfolio dashboards: one in Power BI on synthetic Shopify data, one in Looker Studio on Google Ads + GA4. Each answers 3 questions and ships with a 1-page metric definition doc.
- Week 2-3. Stand up a personal Snowflake or BigQuery sandbox + a dbt project with 5-10 models. This is your "I can do warehouse work too" portfolio.
- Week 3-4. Attorney reviews MSA + SOW with explicit IP carve-out for your reusable framework. Bind E&O. Set up Stripe with deposits enabled.
- Week 4-6. Send 30 cold DMs/emails to your vertical. Target: 4-6 free audit calls, 1 paid pilot at $2,000-$3,000 in exchange for a written case study.
- Week 6-8. Ship the pilot. Record a 2-minute Loom of the dashboard answering the client's questions. Post to LinkedIn + one vertical community.
- Week 8-10. Raise audit price to $3,000-$5,000. Quote your first $5,000-$8,000 dashboard build. Convert the pilot to a $500-$1,000/month retainer.
- Week 10-12. Land 1-2 more engagements. Target: $5,000-$9,000 in 90-day revenue, 1-2 retainers signed, written case studies from two clients.
Common Pitfalls
- Building before asking which decision the dashboard changes. Most consultants get hired to build "a sales dashboard" and ship 12 charts — none of which the CEO opens twice. Every engagement starts with a 60-minute discovery call where the client names the meeting this answer gets used in and the one decision hinging on it. If they can't, the engagement isn't a build — it's a $3,000-$5,000 audit. Skipping this step costs the $12,000 retainer that would have followed a clearly-scoped build.
- Quoting hourly instead of fixed-fee. Hourly keeps analytics consultants stuck at $3,000-$5,000/month forever. The fee is the value of the answer, not the time it took. Fixed-fee per project. Fixed-fee per month for retainers. Time-and-materials only for emergency work on an existing retainer.
- Skipping the maintenance retainer conversation. Every dashboard rots. Quote the $500-$2,000/month retainer in the same proposal as the build. Clients who hear "$5,000 build" without "$800/month maintenance" assume the dashboard maintains itself, then disappear when it drifts. Eight retainers at $1,000 is $96,000/year of floor income before any new build lands.
- Saying yes to a healthcare client before checking the BAA gate. A dental group asking for "patient analytics" is a contract you walk into thinking is a $6,000 build and walk out of with a six-figure HIPAA enforcement action. Route them to a HIPAA-eligible warehouse (BigQuery with a signed Google BAA), scope strictly outside PHI, or pass. Five minutes of due diligence prevents a settlement that erases three years of revenue.
Get your full launch plan — take the free 60-second quiz.
