Background: I'm not a developer. I work in finance. I decided
to challenge myself to build and ship 20 products in 20 weeks
as a side income project — one product per week.
This is Week 1.
The problem I found: Freelance social media managers spend
2–3 hours manually building Instagram audit reports in Canva
or Google Docs before sending to clients. They charge
$300–500 per audit. The report building is the painful part.
What I built: Enter any public Instagram username → get a
downloadable PDF audit report in 60 seconds. Covers
engagement rate vs benchmark, follower quality score, content
mix (Reels/Images/Carousels), overall grade (A–D), and
up to 5 recommendations.
https://auditgram.onrender.com/
Tech stack: Python, Flask, ReportLab, Instagram Looter2 API,
deployed on Render.
What the last 10 days actually looked like:
→ Day 1: Wrong idea. Started with generic Instagram analyzer
→ Day 3: Pivoted after Fiverr research showed the real gap
→ Day 5: Hit 4 broken APIs before finding one that worked
→ Day 7: First working PDF generated
→ Day 10: Live on the internet
Honest assessment of myself: 7-day plan took 10 days. Lost
3 days in a Reddit API rabbit hole that went nowhere. Still
no paying customer. But the product works and it's live —
which is more than I had 10 days ago.
Questions I genuinely need help with:
Would love brutal honest feedback from anyone who's been here
before.
Hi @bninad
First point - minor HTML issues with the LP right now, need fixing.
I'm an expert on PDF generation and we've even generated PDFs just like this for clients. I'd say the PDF is OK and it really depends on who the clients are and where this PDF is going. If it's just internal use for a weekly update, that's fine. If it's an agency sending it to clients (and that's the most likely) then I'd work on it more. Pro tip: get an AI agent to generate a few distinct designs and copy the good bits. It'll only use HTML but I think you could still get a quick improvement from there. Get some inspiration from Behance and Dribbble, too.
Big problem is the page breaks, especially with the variable-length recommendations section. This is exactly the kind of thing we solve at Papermill.io - you can grab our free tier and MCP and get Claude Code to put something together for you.
We actually launched a similar product a while back. A company came to us with a similar problem and we solved it for them. Then we tried to sell it elsewhere. For a month I rang performance marketing agencies, went to their meetups, spoke to more than I can remember. Over 90% of them told me they used Looker Studio, which was free but a bit crap. It doesn't integrate with Meta natively but they bought plug-ins that did. What I found was that Looker Studio owned the market. Also found they were pretty tight on cash generally, but I think your pricing is probably too cheap.
I think your best bet is to really specialise on Instagram in particular, get as much analysis and feedback in there as possible (probably a longer report, include things like comparable accounts and strategy) and then hope pricing convinces them to adopt. Consider some sort of partnership where SMM helps you design an audit together for free and they offer them to their clients, as well as letting you go on to sell them elsewhere.
The good news is most marketers want similar things, so focusing on making it amazing for a handful of users should generalise.
Good luck!
@david_papermill Looker Studio intel is the most useful competitive context I've gotten so far — had no idea it was that entrenched. The page breaks on variable-length recommendation sections are a known issue I've been avoiding; fixing that this week. The partnership idea (co-design with one SMM, let them offer it to clients) is actually how I'd want to find my first real user anyway, so that framing is helpful.
One thing to keep in mind: we've taken investment, so we couldn't follow a small market or consider cheaper ticket prices. It just wasn't a big enough opportunity for us at the time. Good luck!
10 days to shipping is legit. Props for that.
Real talk though, the PDF probably looks good enough, but SMMs won't care about the design. They care if it actually saves them time. If the recommendations are just generic stuff like "post more Reels," they'll ignore it. But if you're telling them specific things like "your engagement tanked when you stopped posting 3x/week," they'll actually use it.
$9 though? Honestly probably too cheap. Not because it's expensive, but because SMMs who charge $300-500 per audit are gonna be skeptical. They'll want to test it free first before changing their workflow. So it's less about price and more about proving it works.
If I had to pick one thing that'd actually move the needle, show them how they compare to benchmarks in their niche. Like "you're at 2.3% engagement, industry average is 4.1%." That's the thing SMMs can immediately show clients and be like "see? you need my help."
But real question: have you actually talked to any SMMs yet? Not asking if they'd pay, just if they'd use it for free. That's usually where things break.
Building is the easy part. Getting people to actually test it and give feedback is where most products die.
@workwise Honest answer to your last question: no, not yet. I posted here before reaching out to any real SMMs, which is probably backwards. Fixing that this week — going to DM 5 freelance SMMs on Reddit/LinkedIn and offer free audits in exchange for feedback. The "engagement dropped when you stopped posting 3x/week" level of specificity is exactly what I need to build toward.
Shipping a working app in 10 days while working a day job in finance is a massive win, congrats! To answer your question about whether an SMM would pay $9: absolutely, but only if the final PDF looks incredibly polished. If they are charging clients $300+ for an audit, they won't hand over a report that looks like a basic, rigid Python script output. ReportLab can look a bit clinical out of the box. If you can make the styling and typography look as clean as a premium Canva template, $9 to save two hours of manual labor becomes an absolute no-brainer for them.
@eva_nomadOS This is the feedback I needed. ReportLab out of the box does look clinical — going to spend a day on typography and layout before adding any new features. You're right that a freelancer charging $300 won't send something that looks like a script output.
So glad it helped! While you're tweaking the layout, see if you can make it easy to inject a custom accent color and a logo. SMMs are incredibly protective of their agency's branding, and even a super basic "white-label" look makes this tool ten times more valuable to them. Once they see their own brand colors on a gorgeous ReportLab output, you'll have no trouble turning them into paying users.
Is that a 9 dollar payment for one audit? No recurrent payments since its a one time action, right?
Right now it's per audit — pay once. No subscription yet. That's partly why the pricing feels off; the model doesn't match how SMMs actually work. Moving toward a monthly plan for unlimited audits, which changes the math entirely.
Solid Week 1, man. Respect for jumping from finance into shipping something real in 10 days.
The idea hits a clear pain point — those manual audit reports sound painful. The core value (60-second PDF with engagement, content mix, grade + recommendations) feels right for freelance SMMs.
On pricing: $9 seems reasonable as an impulse buy if the PDF looks professional. For the one killer feature, I’d say niche-specific benchmarks — comparing the account to similar ones in the same industry/size would turn it from “nice report” into something they can actually use in client proposals.
@orvi2024 thanks - "You're at 2.3% engagement, industry average is 4.1%" is exactly the line an SMM can drop in a client call to justify their retainer. Niche benchmarks are going on the roadmap.
Which model are you using for audit?
"Impressive work for a first launch in 10 days! Moving from finance into development is a huge challenge.
To answer your question on value: the time saved (2-3 hours) easily justifies $9, provided the PDF looks professional at first glance. Regarding the feature that would make this a 'must-have,' I believe automating the report delivery (via a unique link or direct email) would be a key step to remove all remaining friction.
Have you considered allowing PDF branding customization? For an SMM consultant charging $500, being able to add their own logo to the audit report is often what turns a simple tool into an essential part of their workflow."
@NoorHouai White-label is now on the build list after multiple people flagging it. The automated email delivery angle is one I hadn't considered — removing even the step of downloading and attaching the PDF reduces friction meaningfully. Adding that too.
Love that you're iterating this fast. The automated email delivery is a game-changer for SMMs who manage multiple clients — they can literally send audits while sleeping.
Quick question: have you thought about a 'recurring audit' feature? Monthly automated reports for retainer clients. That's where the real LTV is.
Also — I'm building ProfitPulse (AI business auditor for freelancers) with a similar $0 stack (n8n + Groq + Sheets). Would love to exchange notes on pricing psychology for solo tools.
I run a social media SaaS with 15,000+ users and freelance SMMs are exactly our buyers: they won't pay $9 per audit, they'll pay $29 to $49 a month for unlimited white-labeled reports, because audits are how they win retainer clients. Price the workflow, not the file. And one caution on 20 products in 20 weeks: if week 1 finds a paying niche, staying beats the next 19 ideas.
@GregoryScottHenson The pricing logic is clear - price the workflow, not the file. $29-49/month for unlimited white-labeled reports is how SMMs think about tools. On the 20-in-20 question: I'm going to set a rule for myself - if I can get one paying customer in the next 7 days, I'll reconsider the schedule entirely. If I can't, I'll treat that as signal and move on. Your point about staying vs. the next 19 ideas is the most honest challenge to the format I've heard.
Huge congrats on shipping your first product in just 10 days as a complete beginner — that’s seriously impressive! 🔥
The idea is spot on. Freelance SMMs waste hours making these audits manually, and saving them that time (while helping them look more professional) is a real painkiller.
I checked out https://auditgram.onrender.com/ — the speed is great and the PDF looks clean. For your questions:
The PDF is definitely good enough to send to clients (especially after a quick polish).
$9 feels low. Many freelancers would probably pay $19–29 if it saves them 2–3 hours and helps close bigger deals.
One killer feature could be customizable branding (their logo + colors) so the report looks like it came directly from them.
Keep crushing it with the 20-in-20 challenge — Week 1 is already a win. Would love to see how it goes!
@NikolaGluhinic Thanks for actually visiting — that's more than most. Glad the speed felt right. White-label is getting built next, and moving price toward $19-29 after seeing this feedback consistently across comments.
Congrats on actually shipping in 10 days from a finance background. That's the part most people never get past.
On your three questions, I'd push on something nobody here has mentioned yet: the data underneath it. You hit 4 broken APIs, and that's not a one-time hurdle, it's the whole game for this category. A follower-quality score and engagement-vs-benchmark are only as trustworthy as the source feeding them, and unofficial IG endpoints (like Looter2) break, rate-limit, and quietly return stale or partial numbers.
That matters because the freelancer isn't really selling a PDF, they're putting their own credibility in front of a client. If your "follower quality" or engagement number is off because the API undercounted, the SMM looks bad and never comes back. So I'd spend your next few days on data accuracy and defensible benchmarks (cite the source and date on the report) before adding any features. That's what actually makes it "good enough to send to a client."
For what it's worth, I'm building in the same space (creator vetting for founders), and the data layer is where these tools live or die, way more than the UI. Happy to compare notes on providers if that's useful.
This is the most useful technical feedback I've received. You're right - I've already hit 403s on certain accounts and the data inconsistency is a real risk.
Would genuinely love to compare notes on providers. Building in public here so happy to connect - what are you using for your creator vetting tool?
Appreciate that, glad it was useful.
For Scruta I'm on EnsembleData. It's a paid API (you buy a daily unit quota) but it handles the proxy rotation and account access for you, so I'm not the one fighting 403s and rate limits, they are. That's the main reason to move off a raw scraper like Looter2: you're paying someone else to own the part that keeps breaking.
The catch for your model is cost. A managed provider eats into margin on every lookup, so at a $9 price it only really works if you cap usage or the price climbs (which lines up with what others said about positioning it higher). Roughly the landscape: unofficial scrapers (cheap, fragile), managed scraping APIs like EnsembleData (reliable, cost per call), and official connected-account data like Phyllo if you ever go the "creator links their own account" route (most accurate, more setup).
Happy to keep comparing notes, I'm building in public too. What's your rough cost-per-audit tolerance? That basically decides which tier makes sense for you.
@DOCL EnsembleData looks like the right move once I have paying users. Current cost tolerance is close to zero — so staying on Looter2 until someone actually pays, then the math changes. Would you be willing to share rough per-call costs so I can model the break-even at a $29/month price point? That would help me decide when to make the switch.
Happy to. Rough numbers, and check their current pricing since it moves, but on the entry plan it's around $100/month for a daily unit quota, and an Instagram profile plus recent posts runs something like 10 units, which works out to a couple cents per audit. Per call gets cheaper as you move up tiers.
But here's the thing that actually decides your break-even, and it isn't the per-call cost: it's the fixed monthly floor. Pennies per audit is noise. The $100/month minimum is the hurdle. At $29/month you need roughly 4 paying customers just to cover the data floor before you make a dollar, almost regardless of how many audits each one runs.
So your instinct is right: stay on the free scraper until people actually pay. The switch only makes sense once you've got 4 or 5 payers, and by then the per-call cost is rounding error. Some providers have pay-as-you-go or cheaper entry tiers that lower that fixed floor while you validate, worth a look, but free-until-first-payers is the correct call here.
Congrats on shipping in 10 days, that's honestly the hard part. I've been building something adjacent (a media kit generator for creators) and ran into similar questions, so here's my honest take:
On whether the PDF is "enough" for a real client, I think it depends on whether it replaces judgment or just replaces formatting. Freelancers don't need the audit to think for them, they need the 2-3 hours of copy-pasting screenshots gone. If yours reliably does that, it's already valuable even with fairly basic insights.
On whether they'd pay $9 standalone, my guess is it's more likely to land bundled into their existing workflow ("I added this to my $300-500 audit service") than as a separate tool they have to remember to open.
The one feature I'd bet moves the needle most: letting them white-label it with their own logo/colors before sending to their client. Freelancers seem to care a lot about looking polished to their own clients, sometimes even more than the underlying data.
The white-label point has now come up twice in comments and that's enough signal for me. Going to build logo upload as the next feature.
Your framing about "replacing copy-pasting screenshots" rather than "replacing judgment" also clarifies the positioning significantly. That's exactly what it does.
Really solid execution for 10 days, especially starting from non-technical background. The part about losing 3 days on Reddit rabbit holes is relatable - that's the founder tax.
On your questions about pricing: $9 feels too low if you're competing on freelancer speed. They're paying for the time saved, yes, but also for confidence in the report quality. What if you positioned this more around "client-ready audit" vs just speed? That's worth more. The "one feature" you might explore - would SMMs pay more if clients could see their specific improvement actions and then track progress over 30 days?
The 30-day tracking idea is interesting — that essentially turns a one-time tool
into a recurring relationship. Going to think about that seriously.
One thought: freelancers may not be paying for a PDF—they're paying for confidence in the conversation that follows.
If the report helps them explain why an account needs certain changes and makes clients more likely to say yes, it becomes much more valuable than simply generating an audit faster.
"Confidence in the conversation that follows" — that reframe changes how I should be marketing this entirely. Thank you.
I'm glad it resonated.
Your reply made me think about one consequence of positioning the product that way. I'd be happy to share it, but I'd rather explain it with your product in mind than as a generic observation.
What's the best email to reach you on?
Appreciate that — [email protected]. Would love to hear your thinking.
Thanks! I’ve just sent it over.
Looking forward to hearing your thoughts whenever you have a chance.
@aryan_sinh Hey - haven't seen anything land in my inbox yet, might have gone to spam or a typo somewhere. Can you try again at [email protected]? Would genuinely love to hear the thought.
Perfect—I've just resent it, so it should be in your inbox now.
If you don't see it in the next few minutes, could you also check your Promotions and Spam folders? Sometimes first emails end up there.
Looking forward to hearing your thoughts whenever you've had a chance to read it.