Equity crowdfunding campaign — full-funnel performance analysis

Phase 1 — Testing the WatersWeeks 1–3private, insider list

Phase 2 — Public campaignWeeks 4–24~5 months, closed Week 24

Result~$490K raised~370 investors · 98.5% of goal

Data sources joined for this analysis Crowdfunding platform exports · GA4 · HubSpot CRM, email & marketing events · Monday.com execution boards · Amplemarket outbound · Meta Ads Manager · Instagram · TikTok · LinkedIn · YouTube Studio · Spotify for Creators · SharePoint · open-web research

Anonymised work sample. A real campaign analysis with identifying details removed. The company is referred to as Northwind; individuals appear by role rather than name; partners, press and podcast guests are described by category. Dates are expressed as campaign weeks, headline figures are rounded, and screenshots are blurred — the people in them were participants in the campaign, not in this portfolio.All percentages, ratios and multipliers are unchanged from the original analysis, so every calculation shown remains reproducible.

Total Raised

~$490K

vs $500K goal — 98.5%

Investors

~370

384 investments

Avg Investment

~$1,280

median $500

Marketing Spend

~$28K

vs $27,200 budget

Blended ROI

17.6×

~$490K raised / ~$28K spend

Cost per Investor

$75

~$28K / 371

Headline: the campaign raised ~$490K from ~370 investors on ~$28K of marketing spend — a 17.6× blended return and 98.5% of the $500K goal. Average check came in at ~$1,280 vs a $275 plan (4.7×), which is how a 371-investor campaign hit a target modelled on 2,000+ investors.

Funding trajectory

Goal attainment

98.5%

of the $500,000 Reg CF goal

Raised	~$490K
Goal	$500,000
Shortfall	$7,583
Investors	~370
vs investor goal (2,130)	17%
Avg check vs $275 plan	4.7×

Funnel — top to bottom

TOP OF FUNNEL

~600,000

all channels: 70% owned, 23% earned, 7% paid. Not unique people

PAGE TRAFFIC

12,174

GA4 sessions to the platform page

FOLLOWERS

654

peak the platform followers (soft intent)

INVESTMENTS

384

~370 unique investors · 3.2% of sessions

CAPITAL

~$490K

$40.45 raised per session

Where the money is traceable

Only 18.4% of dollars carry a UTM. The hatched slice is real money with no recoverable source.

Spend allocation

Podcast production was the single largest discretionary line at 28% of spend.

The four things to tell the exec team

The raise worked, on a shoestring. ~$490K on ~$28K — 17.6×. Cost per investor $75 against an average check of ~$1,280.
Fewer, bigger investors carried it. ~370 investors, not the 2,130 planned. The plan's assumption about retail volume was wrong; the assumption about willingness to write $500–$1,000+ checks was very right.
Paid media returned nothing. ~$7K all-in produced 37,400 impressions, 969 sessions and zero confirmed investments. Every other channel outperformed it.
The podcast worked as a content factory, not as a direct investment channel. 13 episodes generated 141 cutdowns and 106,083 short-form views — 62× the reach of the episodes themselves. Direct attributable investment from organic social was ~$3.1K.
Three free webinars were the single best thing we did. 62 registrations produced ~$120K — 24.3% of the raise — at 36.5% conversion, 12.6× the base rate. Almost no budget, almost no promotion, and no measurement until now.
Guest collaborations were our highest-reach tool, and we barely used them. One Instagram Collab with a 47,000-follower meteorologist produced the best asset of the campaign — 14,533 views, 10× our whole follower base. Borrowed audience across guests and partners totalled ~25–40× our owned Instagram reach.

Read this first. 81.6% of the raise (~$400K across 290 investments) has no recoverable source. the platform tags it "direct" — your own tracker calls that "a black hole of multi touchpoints." HubSpot cannot help: 97.9% of investors are labelled Offline / integration, meaning the platform sync created the record at the moment of investment, so no arrival channel was ever captured. 90% of investors did not exist in HubSpot before they invested. Everything below describes the 18.4% we can actually trace — treat it as a directional sample, not a channel P&L.

Traceable

~$90K

94 investments · 18.4% of raise

Untraceable

~$400K

290 investments · 81.6%

Largest traced driver

the platform's own list

~$30K · 33% of traced $

Paid media

from ~$7K all-in

Traceable investment by source

Source bucket	Investments	Attributed $	% of traced	% of raise	Share	Detail

The the platform community performed to plan. The strategy assumed 25% of investments would come from the platform's on-platform community, and that is broadly what happened: their platform audience accounts for 61 of 384 investments (15.9%) and ~$30K on traced data alone — a floor, given 81.6% of the raise is untraceable. Within the traced sample it is 65% of investments and 33% of dollars, which simply reflects that platform traffic is UTM-tagged by default while our own channels largely were not.

The standout mechanic was the dedicated email blast (wf_dedicated / a campaign code): 40 investments and ~$24K from a single send — more investments than any other identifiable source in the campaign. The read for next time is not that our marketing underperformed; it is that platform-audience access is a high-yield, low-effort lever worth negotiating harder for, and that one dedicated blast outperformed the newsletter, feed and closing reminders combined.

GA4 sessions by channel

GA4 channel efficiency

Channel	Sessions	Key events	Conv %	GA4 revenue

GA4 revenue sums above the true raise because it credits every session-channel touch; use for relative comparison only.

Paid media, in full. the paid-media agency sprint signed Week 13: $5,000 fixed fee + ~$2K actual media = ~$7K. Delivered 37,400 impressions at a $54.47 media CPM / $188.16 all-in CPM — against a $15 CPM plan assumption. Produced 969 GA4 sessions ($7.26/session all-in), 1 GA4 key event, and 0 the platform-confirmed investments across every facebook, google, google-ads and adwords UTM. The $40,000 contribution this experiment was modelled to produce did not materialise in any measurable form.

Meta campaigns

Campaign	Spend	Impressions	Reach	CPM
Sales campaign — ABO	$1,650.47	24,570	16,266	$67.17
Sales campaign — CB0	$370.92	7,703	4,867	$48.15
Engagement campaign — CBO_Testing	$15.62	5,127	5,125	$3.05
Total media	~$2K.01	37,400	—	$54.47
+ agency fee → all-in	~$7K.01	37,400	—	$188.16

Episodes

the podcast S1

Cutdowns

141

short-form assets

Cutdown views

106,083

YT Shorts + IG Reels + TikTok

Reach multiplier

62×

cutdowns vs episodes

Deep attention

251 hrs

long-form watch + listen

Cost

~$7.8K

28% of all spend

The argument that holds. The podcast was not a direct investment channel — it was the campaign's content supply chain. ~$7.8K bought 13 episodes that were cut into 141 short-form assets, and those cutdowns reached 62× more people than the episodes themselves. On the Monday board, 61 of 71 social items (86%) were podcast derivatives. Strip the podcast out and there is essentially no social program. At $55 per finished asset and a $72.50 CPM — versus $188.16 all-in for paid media — it was the most cost-efficient reach the campaign bought.

And the limit, stated plainly. Direct attributable investment from organic social across the whole campaign was ~$3.1K (6 investments) — 0.6% of the raise, against ~$7.8K of podcast spend. Episode-drop weeks show no reliable traffic lift: of the 9 episodes outside launch week, 4 saw a week-over-week increase and 5 saw a decrease. The two largest traffic spikes both have better explanations than the podcast — public launch (Week 4, 542 new users) and the platform's dedicated email blast (Week 12, 280 new users, the same week your own Week 12 memo credits to "the dedicated email"). Do not claim the podcast drove the raise. Claim it built the awareness layer and the asset library at a fraction of paid cost — that is what the data supports.

Long-form vs short-form — the flywheel

Log scale. Episodes create the depth; cutdowns create the reach. Neither number works without the other.

Efficiency vs paid

PODCAST CPM (all-in)

$72.50

PAID MEDIA CPM (all-in)

$188.16

Cost per finished asset	$55
Podcast share of spend	28%
Podcast share of IG views	67%
Podcast share of YT Shorts views	48%
Podcast share of TikTok views	10%
Social items that were cutdowns	86%

TikTok's share is low because its one viral hit (147K views) was original content, not a cutdown.

Episode by episode — ranked by total plays and views

How to read this. Total is the primary lens: Spotify plays plus YouTube views for the same episode, so an episode is judged on its whole audience rather than one platform. Completion rate is the quality lens — it tells you whether people who started actually stayed. The two together are what separate a good episode from a well-promoted one.

Published	Episode	TOTAL plays + views	YouTube views	Spotify plays	Completion rate	Spotify listen hrs	YouTube watch hrs	Spotify 1st-7d vs its own normal
Week 5	How to Break Overconsumption: a sustainability author	270	239	31	50%	33.2	30.1	182% above
Week 12	Forecasting Our Future: a national TV meteorologist	237	226	11	29%	4.0	11.1	Normal
Week 14	From Robotics to Reuse: the chief product officer	216	195	21	25%	3.5	9.9	91% above
Week 3	Part 1 — The Why Behind Northwind	179	134	45	—	14.1	19.1	309% above
Week 21	Inside a Double Alpha Household: a high-profile founder-investor	152	135	17	50%	11.6	26.0	55% above
Week 16	Beyond the Plastic Debate: a materials scientist	134	130	4	—	—	18.5	64% below
Week 6	From Pediatrician to Founder: a pediatrician-founder	121	110	11	29%	5.2	4.1	Normal
Week 11	Investing in Supply Chain: the lead VC fund	114	104	10	—	4.2	12.8	9% below
Week 7	Why Our Drinking Water Isn't Safe: a water-NGO chief executive	100	91	9	20%	3.9	10.5	18% below
Week 9	Water Is Infrastructure: a water-nonprofit director	83	76	7	38%	6.7	3.4	36% below
Week 4	Part 2 — How We're Building Northwind	77	65	12	29%	7.8	11.7	9% above
11 episodes		1,683	1,505	178	—	94.2	157.2

What the “vs its own normal” column actually measures — and why to mostly ignore it. It is Spotify only. Spotify compares each episode’s first-seven-day plays against the typical first-seven-day plays of this show — not against YouTube, not against podcasts generally. Two problems make it close to useless here. First, the bases are tiny: “309% above normal” means 45 plays instead of ~11. Second, and more importantly, Spotify is only 11% of this podcast’s audience — YouTube carries 89% (1,505 of 1,683). The a materials scientist episode is the clearest illustration: Spotify flags it 64% below normal on 4 plays, while on YouTube it pulled 130 views and 18.5 watch hours and ranks 6th of 11 overall. Judge episodes on total plays plus views and on completion rate; treat the Spotify benchmark as a footnote.

What actually worked, on the two metrics that matter. a sustainability author (Week 5) is the only episode that wins on both — 270 total plays and views, a 50% completion rate and 33.2 listening hours, the highest on every measure. a high-profile founder-investor (Week 21) also hit 50% completion, and drew 26.0 YouTube watch hours from just 135 views, meaning the people who came stayed the longest of any episode. At the other end, a water-NGO chief executive (Week 7) pulled a respectable 100 total but only 20% completion — the weakest retention of the season, which suggests the audience came for the regional water topic and left mid-episode. Part 1 (Week 3) shows the launch effect clearly: 45 Spotify plays, the highest of any episode, off pure launch curiosity rather than sustained interest. The pattern holds — guest-led, issue-first episodes retain; company-narrative episodes attract and then lose people.

One caveat on the total. Spotify only publishes first-seven-day plays, while YouTube views are lifetime. Summing them understates Spotify, so the 11% Spotify share is a floor rather than a precise split. Also excluded from this ranking: the Welcome to the podcast trailer (1:19, 19 Spotify plays) and the “Would you invest $100” promo (4 plays), neither of which is a full episode.

The single clearest pattern in five months of content: issue-led content earns engagement, ask-led content earns reach — and the two almost never happen in the same post. Every top-engagement post reframes a problem the viewer already half-believes. Every direct "invest now" post pulled respectable views at a fraction of the engagement rate. The best-performing asset of the campaign did both, because a 47,000-follower meteorologist lent it her audience.

Best-performing creative across every channel

Every row links to the live post. Thumbnails are embedded stills captured from the posts themselves — where a still isn't embedded, the coloured tile still links out. Two of these turned out to be Instagram Collabs: a national TV meteorologist reel and the regional water reel both carry Instagram's “2 people” badge, which is documentary confirmation of the collaboration.

#	Asset	Channel, copy and link	Reach	Eng.	Why it worked
1		TikTok · Week 16 “260 cans of Diet Coke. That’s how many cans you’d waste a year for every little treat during the workday…” brand account/video/a campaign code	147,175 views	1.7% eng. rate	A concrete, countable number attached to a habit people recognise in themselves. No brand, no ask. 2,028 shares.
2		Instagram · Week 12 a national TV meteorologist Collab. “Climate change isn’t a future threat — it’s a present reality. And how we communicate it determines…”	14,533 views	6.7% eng. rate	Borrowed authority plus borrowed audience. Best reach and near-best engagement — the only asset that achieved both.
3	Instagram ▶	Instagram · Week 10 “That kick was craaaaaazy! Join us in repackaging the beverage industry together on @the platform”	4,656 views	1.4% eng. rate	Culture-jacking a viral moment. Big reach, weak engagement — the classic ask-led profile.
4	YouTube ▶	YouTube · — Short. “KADEYA V2 9x16” — untitled product film	4,160 views	— eng. rate	Pure product footage. Machines doing visibly clever things travel on their own.
5		Instagram · Week 7 regional water episode — also a Collab. “And if we want to protect them — and our drinking water — we have to remember why they got better…”	3,412 views	6.8% eng. rate	Highest engagement rate of any high-reach post, and 30 shares — the most of any Reel. The “2 people” badge confirms a regional water NGO collaborated.
6	Instagram ▶	Instagram · Week 20 “A lot of hard work has gone into building our first stations (from scratch!)…”	3,160 views	1.6% eng. rate	Build-in-public manufacturing content. Proof of progress at the moment investors were deciding.
7	Instagram ▶	Instagram · Week 3 “the podcast: A startup podcast by Northwind launching Monday 11/17. Follow us on YouTube, Spotify…”	2,717 views	8.3% eng. rate	Highest engagement rate on Instagram all campaign. Announcing something new to an owned audience is the most reliable engagement play there is.
8		LinkedIn · Week 3 Post. the podcast podcast launch announcement campaign code	4,920 views	11.4% eng. rate	Best-reaching LinkedIn post of the campaign at 11.4% engagement, with 5 reshares.
9		LinkedIn · Week 21 a high-profile founder-investor episode. “What happens when two founders build companies, and a family, under one roof?” campaign code	4,048 views	12.5% eng. rate	Highest engagement rate of any post on any channel. A human curiosity hook plus a recognisable name.
10		TikTok · Week 9 “We’ve made water invisible — and that’s the problem. 💧” brand account/video/a campaign code	1,309 views	20% eng. rate	The single highest engagement rate recorded anywhere. One sentence, one reframe, no ask.

Messaging that resonated

Angle	Example	Signal
The contrarian reframe	“Plastic isn’t a bad material, we just are using it wrong” · “What if we’re asking the wrong question about plastic?” · “We’ve made water invisible”	8.5–20% ER
A concrete, countable number	“260 cans of Diet Coke”	147K views
Health, made personal	“It’s About Your Kids’ Future” · “BPA-Free Isn’t Enough — A Pediatrician on Why Plastic Isn’t Safe for Kids”	top 5 YouTube
Justice framing	“Clean water isn’t just a climate issue — it’s a justice issue”	17% ER
Founders as people	“Two founders, one household” · “When you’re also married to an entrepreneur”	~1,100 views each, 12.5% ER on LI
Infrastructure, not packaging	“Water isn’t just essential — it’s infrastructure” · “Replacing the bottling plant”	2,130 impr · best outbound line

Messaging that didn’t

Angle	Example	Signal
The direct investment ask	“Invest in our community round on @the platform. We’ve already raised over $…”	0.42% ER
Milestone and progress updates	“$430K+ raised” · “how much we’ve raised” launch posts	4 skipped
Static graphic quotes	The “Graphic Quote” slot in the podcast template	8 of 11 skipped
Company-narrative episodes	“Part 2 — How We’re Building Northwind” · the materials-science episode	64% below normal
Promotional podcast episodes	“Would you invest $100 to eliminate plastic bottles?”	4 plays

The team voted with its feet: the formats that got skipped mid-campaign were the ask-led and static ones. Video cutdowns were never skipped.

Paid creative — by delivery

On links. Meta ad creative has no public URL — ads only exist inside Ads Manager, so there is nothing to link to directly. But four of the video ads were cut from assets that also ran organically on YouTube, so those rows link to the organic version of the same footage. The image ads and the VSL founder recuts were made for paid only and have no organic twin.

Creative	Format	Angle	Spend	Impr.	CPM	Results	Cost / result	Organic version
103-V4-02-4 — Recycling: Northwind	Image	Solution	$173.02	3,841	$45.05	0	—	no organic twin
101-V1-02 — “hello to sustainable hydration”	Video	Engagement	$12.36	3,734	$3.31	1,447	$0.0085	no organic twin
101-V1-01 — “Welcome to the podcast”	Video	Solution	$282.84	3,032	$93.28	1	$282.84	watch the organic cut →
103-V4-02-6 — Recycling: Northwind rejected	Image	Solution	$115.75	2,849	$40.63	1	$115.75	no organic twin
103-V4-02-3 — Recycling: Northwind	Image	Solution	$201.57	2,696	$74.77	0	—	no organic twin
101-V1-04 — “Startup truths from Northwind’s founders”	Video	Solution	$317.39	2,494	$127.26	0	—	watch the organic cut →
001-V1-W5-02 — Testing Northwind Ads	Video	Product	$102.35	2,105	$48.62	0	—	no organic twin
103-V4-02-0 — Recycling: Northwind	Image	Solution	$90.95	1,782	$51.04	0	—	no organic twin
002-V2 — Facebook post, Testing Northwind Ads	Video	Product	$126.03	1,762	$71.53	0	—	no organic twin
105-V1-01 — VSL Founder recut	Video	Solution	$100.55	1,704	$59.01	0	—	no organic twin
101-V1-02 — “Cheers to a Future Without Single-Use Bottles”	Video	Solution	$9.47	127	$74.57	0	—	watch the organic cut →
101-V1-03 — “The beverage supply chain is inefficient”	Video	Solution	$76.18	806	$94.52	0	—	watch the organic cut →

56 creative variants ran in total. “Results” are ThruPlays on the engagement campaign and landing-page conversions on the sales campaigns — not directly comparable, which is itself the finding below.

The real paid lesson isn’t the creative — it’s the objective. The same 101-series videos ran under both campaign objectives with wildly different economics. Optimised for ThruPlay, they bought impressions at a $1.37–$3.31 CPM and video views at a quarter of a cent. Optimised for landing-page conversion, the same concepts cost $45–$127 CPM and converted essentially nobody. Meta was perfectly willing to sell cheap attention and completely unwilling to sell cheap investors — because a Reg CF investment is a slow, high-consideration decision that a cold prospecting ad cannot close.

Two specifics worth naming: “Startup truths from Northwind’s founders” spent $317.39 — 16% of all media — for 2,494 impressions and zero results, the worst line item in the campaign. And “Welcome to the podcast” produced one result at $282.84. Three Recycling: Northwind image variants were also rejected by Meta, so budget flowed to whichever variants survived review rather than whichever performed.

the social lead’s three-month social review — what it adds

Source: Northwind: 3 Month Social Learnings, covering end-November to end-February on Instagram and TikTok — a subset of the full campaign window, and a qualitative read from the person producing the content. It agrees with the quantitative analysis in most places, sharpens it in two, and adds five things the exports could never have shown.

Where it sharpens the numbers

My read from the data	the social lead’s refinement
“Nine of the top ten assets are video; carousels never exceeded 474 reach without a collab.”	“Carousels and statics are outperforming Reels.” Both are true and the distinction matters: Reels win on reach, carousels win on engagement. Her example — Punch The Monkey — had high engagement and low views, which is invisible if you rank by reach alone.
“A concrete, countable number was the winning hook” (260 cans of Diet Coke, 147K views)	“Numbers with no return — reduce content featuring stats.” The sharper version: the 260-cans number worked because it was personal and tied to a trend (“3pm Diet Coke”), not because it was a statistic. Abstract sustainability stats underperformed. The relatability is doing the work, not the number.

Where it confirms them

Guest collabs and the podcast carousel. She flags a national TV meteorologist carousel as “the first time a podcast episode was broken down into a carousel and it saw great return” — independently arriving at the asset the reach data singles out as an outlier. She also notes they repeated it with the CPO and it underperformed, attributing that to a missing headline in the first frame.
Trends with a Northwind twist. Matches the contrarian-reframe pattern in the engagement data.
People over product. Her “Northwind employees ARE Northwind” is the same finding as “led with a person, not a product” — but she credits employees specifically, not just founders.
TikTok isn’t cracked. “TBH, we haven’t cracked the code here.” Consistent with TikTok delivering the most raw reach, the thinnest engagement and zero traced investment.

Five things the exports could not have told us

Finding	Why it matters
Trending language beats trending formats	Posts using trending language with “longer cultural staying power” outperformed posts copying trending formats. 3pm Diet Coke and Imagine the Propaganda are both language plays. A production instruction, not a metric.
Captions are now SEO, and hashtags aren’t	TikTok auto-generated a “plastic-free options” search off the 3pm Diet Coke caption and generated nothing for almost every other post. A searchable-caption vocabulary is a free, immediate distribution lever.
Native in-app text + a custom thumbnail	The two top performers both used in-app text with a designed thumbnail. A specific, replicable production recipe.
“Machines, not bottles” confusion is surfacing	A TikTok commenter asked whether Northwind had invented a bottle. A positioning risk on the core proposition, visible only in comments — worth fixing in caption language before the next raise.
Faceless content is not resonating	Text-only video underperformed, with one exception: Propaganda, which carried trending language. The reframe formula needs either a face or borrowed cultural language to land.

The most actionable thing in her document is a recommendation that was made and never acted on. the social lead explicitly flagged two proven organic winners for paid promotion — the CPO’ Kick (“I’d consider this one for paid. I think it might be worth it!”) and Living In China (“Consideration: Paid! This could be quite successful”). Neither was ever put behind spend. Instead the ~$7K went to Recycling: Northwind image variants, a Welcome to the podcast cut, Startup truths from the founders, and VSL founder recuts — none of which had organic traction, and three of which Meta rejected outright.

So the paid failure was partly a creative-selection failure, not just a targeting or objective failure. The campaign had a 147,000-view organic hit and several other proven assets sitting unused while budget ran against untested creative. The rule for next time is simple: only put spend behind creative that has already earned reach organically.

Her Start / Stop / Continue, worth carrying into the next campaign

START

Saveable, actionable inspiration content (Ways to Reduce Plastic in showed promise)
The “faceless + trending language” formula — statements, not information
Language clarifying Northwind is a circular machine, not another bottle
A culturally-relevant caption vocabulary for searchability
Supportive, non-promotional commenting on other accounts to earn algorithmic favour

STOP

Stats-and-numbers content — the audience’s understanding of them is surface-level
Faceless text-only video, unless it carries trending language

CONTINUE

Featuring employees in trends, BTS and company updates
Trends with a Northwind twist, favouring language over format
“Quick gets” — content understood at a glance
In-app text with a designed thumbnail; build Canva thumbnail templates

What the best content had in common

It made a familiar thing strange. Water is invisible. Plastic isn’t the villain. You’re throwing away 260 cans a year. Every top-engagement post reframed something the audience already half-knew — the campaign’s most repeatable formula. The nuance from the production side: it works when the reframe is personal and carried in trending language, not when it arrives as a statistic.
It borrowed someone else’s credibility. A meteorologist on climate, a pediatrician on plastic, a regional water advocate on water policy. Guest-led assets outperformed company-led assets on every platform, and the one true Collab outperformed everything.
It led with a person, not a product. Founder-marriage content pulled ~1,100 YouTube views a piece and the highest LinkedIn engagement rate of the campaign (12.5%). Nobody shares a vending machine; they share a couple arguing about sleep.
It didn’t ask for money. The direct-ask posts got views but 0.42–1.7% engagement, and four launch-week ask posts were abandoned mid-campaign. Reach and persuasion came from issue content; the ask worked in email and on webinars, where people had already opted in.
Short-form video for reach; carousels for engagement. Nine of the top ten assets by reach are video, and carousels never exceeded 474 reach without a Collab. But the production-side read is the other half of the picture: carousels and statics that are “quick gets” out-engage Reels even when they under-reach them. Static graphic quotes were the format that genuinely failed — abandoned 8 times out of 11.
Platform shapes the payoff. TikTok delivered raw reach with thin engagement (1.7% on the 147K hit). LinkedIn delivered small reach with the deepest engagement (11–12.5%) and the most traced investment. Instagram was the only channel that did both — and it is where the podcast cutdowns lived.

All channels, grouped owned / earned / paid

Channel	Assets	Reach / views	In total?	Followers Δ / group reach	Traced $	Note

"Traced $" is the platform UTM-confirmed investment only — it undercounts every channel roughly 5× given 81.6% of the raise is untraceable. Compare channels against each other, not against the total.

Top of funnel: owned / earned / paid

Owned ~420,000 (69.6%) · Earned ~138,000 (22.9%) · Paid ~45,000 (7.4%). The campaign ran overwhelmingly on channels we controlled and distribution we earned. See Data Quality for the build-up.

Follower growth

TikTok

Posts	98
Views	188,401
Reach	176,116
Likes	2,934
Shares	2,075
Followers	314 → 431
Top video	147,175

One video — "260 cans of Diet Coke" (Week 16) — did 147,175 views, 78% of all TikTok views. It was original content, not a podcast cutdown. Engagement rate was only 1.7%, so it delivered reach without depth, and no traced investment.

Instagram

Posts	110
Post views	111,507
Reels	100
Reel views	102,346
Stories	230
Story views	10,225
Followers	913 → 1,100

Reels are a subset of posts (86 of 110 posts are video) — do not add the two rows. De-duplicated feed total is 111,507.

The podcast's best home. 67% of IG views came from podcast-derived content. Top reel — a national TV meteorologist climate cutdown (Week 12) — did 14,533 views at 6.7% engagement, the strongest single organic asset of the campaign.

Posts	37
Impressions	34,416
Clicks	2,254
CTR	6.55%
Reactions	867
New followers	+405
Traced investment	$2,100

Smallest reach, best conversion. 34K impressions — a fraction of TikTok's 188K — but a 6.55% CTR and the highest traced investment of any organic social channel. Audience skews Senior (955) and a US metro (726). This is where the investors actually were.

YouTube

Total views	42,620
Watch time	309 hrs
Subscribers gained	+75
Impressions	37,595 (2.87% CTR)
Shorts views	38,507 90%
Long-form views	4,112 10%
Shorts watch time	130 hrs
Long-form watch time	179 hrs 58%
the podcast playlist	1,662 views · avg view 5:43

The flywheel in one channel. Shorts took 90% of views but long-form took 58% of watch time. Reach and depth came from different formats of the same content. Note 7,488 views (17.6%) came from YouTube advertising — paid video that produced no traced investment.

Spotify for Creators

Episodes	13
Followers	25
Consumption	97.6 hrs
Plays (first 7d, all eps)	201
Comments	7
Best episode	a sustainability author — 33.2 hrs
US share of audience	82.9%

Spotify never found an audience. 25 followers and 201 first-week plays across 13 episodes. As an audio destination it did not work. Its value was as a credential and a link — the Refashiond episode URL in outbound email — not as a distribution channel. Consider audio a byproduct next season, not a target.

the newsletter was a core pillar of the campaign and has been under-credited. Thirteen issues went out during the raise. It is 34.6% of all campaign email volume but pulls 36.3% of the opens, it produced the single highest-clicking email of the entire campaign (#8, 51 unique clicks), it was the primary promotional vehicle for the webinars — carrying 58% of all pre-webinar email clicks — and it is the only asset in the campaign HubSpot attributes any revenue to at all. Both attributed deals run through a newsletter touchpoint.

Issues sent

#6 – #17, in window

Delivered

14,642

34.6% of campaign email

Open rate

25.4%

vs 23.6% non-newsletter

Unique clicks

272

1.86% click rate

Pre-webinar clicks

58%

64 of 110

Attributed deals

2 of 2

100% carry a newsletter touch

Issue by issue

Issue	Sent	Delivered	Opens	Open %	Clicks	Click %	CTOR	Note
#6	Week 1	1,982	392	19.8%	33	1.66%	8.42%	TtW push
#7	Week 2	1,340	258	19.3%	10	0.75%	3.88%	pre-Webinar 1
#7c	Week 2	1,121	235	21.0%	15	1.34%	6.38%	date correction resend
#8	Week 3	1,938	353	18.2%	51	2.63%	14.45%	best email of the campaign — public launch
#9	Week 4	870	278	31.9%	26	2.99%	9.35%	launch week
#10	Week 6	816	243	29.8%	9	1.10%	3.70%	quiet week
#11	Week 6	838	251	30.0%	5	0.60%	1.99%	weakest issue
#12	Week 7	864	273	31.6%	10	1.16%	3.66%	pre-Webinar 2
#13	Week 12	845	213	25.2%	15	1.78%	7.04%	subject mislabelled "#12"
#14	Week 14	947	297	31.4%	22	2.32%	7.41%
#15	Week 15	991	310	31.3%	11	1.11%	3.55%	4 spam reports
#16	Week 20	1,046	299	28.6%	29	2.77%	9.70%	Webinar 3 — same morning
#17	Week 23	1,044	321	30.8%	36	3.45%	11.21%	best click rate — closing week
Total		14,642	3,723	25.4%	272	1.86%	7.31%

Clicks track milestones, not cadence

The four best-clicking issues all coincide with an event — public launch (#8), closing week (#17), the Testing the Waters push (#6) and webinar day (#16). The four worst were routine "quiet week" sends. Send the newsletter around moments, not on a calendar.

Engagement predicts investment

Cohort	People	Invested	Rate	vs base
All contacts	~10,800	283	2.62%	1.0×
Received ≥1 email	5,296	282	5.32%	2.0×
Clicked ≥1 email	1,078	120	11.13%	4.3×
Registered a webinar	96	16	16.67%	6.4×

The lift lives at the click, not the delivery. Getting someone to click any email doubled their investment probability again on top of merely reaching them.

the newsletter as webinar engine

Webinar	newsletter issue	newsletter clicks	Reminder clicks	Verdict
Webinar 1 — Week 3	#7 + #7c (Week 2)	25	21	newsletter led, narrowly
Webinar 2 — Week 8	#12 (Week 7)	10	25	the "time change" email did the work
Webinar 3 — Week 20	#16 (Week 20, same morning)	29	8	newsletter out-clicked the reminder 3.6:1
Total		64 (58%)	46 (42%)

Honest caveat on that 58%. HubSpot doesn't expose per-link click data, so this counts every link in each email. A newsletter issue has five sections and a dozen-plus links; a reminder has one. 58% is an upper bound and structurally flattering to the newsletter. Registration source isn't recorded anywhere, so the defensible statement is that the newsletter plausibly drove 30–50% of the 62 registrations — anything more precise would be invented.

The strongest single fact in the newsletter's favour. HubSpot attributed revenue to exactly two assets across the entire campaign: the newsletter #8 (email click → Ahuja Family, $13,000 deal) and the newsletter sign-up form (submission → Erika Gianni, $5,000 deal). That is 100% of HubSpot-attributed campaign revenue and 2 of 2 attributed deals, both carrying a newsletter touchpoint. Three caveats keep it honest: linear weighting means the credited slices are only $236 and $294; the attribution pipeline scores closed-won deals only, so ~$387,000 of the raise sits entirely outside it; and n=2 is a very thin dataset. But no other asset in the campaign earned any attribution at all.

Where it under-performed, so the exec read stays balanced. Per delivery the newsletter clicked 26% worse than the rest of the campaign's email (1.86% vs 2.53%) and converted opens into clicks 32% worse (7.31% CTOR vs 10.73%) — expected for a five-section newsletter competing against single-CTA sends, but real. It generated 37.8% of unsubscribes on 34.6% of sends. Its sign-up form has produced two contacts in total. And in a 12-investor sample, four of the five largest cheques were never on its list — the top of the cap table came through direct founder relationships, not the newsletter.

Three operational faults worth fixing: cadence drifted off Tuesday and became erratic — weekly through December, then a four-week gap, then bi-weekly, with only one issue in the final three-and-a-half weeks of the raise. Issue #13 shipped with the wrong subject line ("Issue #12"). And #16 went out the same morning as Webinar 3 rather than ahead of it, functioning as a day-of reminder instead of a lead-in.

The defensible framing

the newsletter was the campaign's retention layer, not its conversion layer. It held a third of the list engaged across a five-month raise and converted decisively at the two moments that mattered — public launch and closing week. It did not originate the money; the CTA sends, the webinars and direct founder relationships did. That is a real and under-credited contribution, and it is a stronger claim than "the newsletter drove the raise" because it survives scrutiny.

Emails delivered

40,380

47 sends

Open rate

24.1%

9,745 unique

Click rate

2.17%

CTOR 8.98%

Unsubscribes

347

0.86%

Traced investment

$25,000

16 investments

Rank

of all traced sources

Email was Northwind's best-performing owned channel. $25,000 in traced investment across 16 investments — second only to the platform's own list, and it cost effectively nothing. Open rates rose from 19.6% in Testing the Waters to 25.9% during the public campaign. The pattern is unambiguous: small, targeted investor sends massively outperformed broad blasts. "Post the platform Investment Flow #4" hit an 11.6% click rate; the rewards-request send hit 50% on 70 delivered. The worst performer was a broad "didn't open" resend — 1,638 delivered, 6.9% open, 19 unsubscribes.

Email by phase

Top sends by unique clicks

the newsletter #8 (Week 3)	51
the platform 2 weeks left (Week 20)	40
Warming #1 — Initial CTA / VIP (Week 1)	38
the newsletter #17 (Week 23)	36
1st station + 1 week left (Week 22)	36
the platform Rewards Request #1 (Week 16)	35
the platform Closed (Week 23)	35

the newsletter #8 is also the only email HubSpot could attribute revenue to ($236) — a reminder of how thin platform attribution was.

Amplemarket investor outbound

Timing caveat. All four sequences were created February–a campaign month — the final nine weeks. There was no Amplemarket outreach during Testing the Waters or the first 11 weeks of the public campaign; that period ran on manual 1:1 outreach (~150 contacts logged Week 1 – Week 3). Meetings booked was never tracked in Amplemarket, so no conversion-to-investment link exists.

Sequence	Contacted	Emails	Open	Reply	Positive reply	LI accept
LinkedIn-only — Variant B edits	74	0	—	—	20.27%	43.3%
Investor Outreach — Variant B edits	232	572	64.59%	0.89%	8.62%	23.4%
Investor Outreach — Ali edits	219	608	58.11%	1.82%	7.31%	31.8%
Consumer Angel — LinkedIn only	188	0	—	—	1.60%	18.2%
Total	713	1,180	61.3%	1.36%	—	24.9%

The winning formula was small, late, LinkedIn-first and InMail-enabled. The LinkedIn-only Variant B sequence hit a 20.27% positive-reply rate on just 74 leads — 2.4× the email sequences and 12.7× the Consumer Angel sequence. Its differentiators: connect before pitching, InMail carrying the full infrastructure pitch, and one proof point the others lacked — "first commercial stations shipping to customers this month."

Messaging: what the data says

"Replacing the bottling plant" beat "Angel Investment Opportunity" on opens (64.6% vs 58.1%) and positive replies (8.62% vs 7.31%). Infrastructure framing outperformed generic investment framing.
But the vaguer subject won more conversations — Ali's version got 2× the raw reply rate (1.82% vs 0.89%). One generated volume, the other generated quality.
Audience mattered more than copy. The same LinkedIn-only mechanic scored 20.27% on investors and 1.60% on consumer angels — a 12.7× gap from targeting alone.
Email earned the open and died on the ask. 58–65% opens against 0.89–1.82% replies. The hook works; the conversion request needs rebuilding.
The reasons investors gave (from 200+ angels, per the final sequence email): mission alignment on microplastics and PFAS; belief in the team; category-defining infrastructure rather than packaging; personal connection — several said it was their first-ever startup investment; and backing female and climate founders.

This is the most important finding in the analysis. Three investor webinars — near-zero incremental cost, just Zoom and team time — produced ~$120K across 19 investors. That is 24.3% of the entire raise from 52 people, or 0.48% of the contact database. Webinar registrants invested at 12.6× the base rate and wrote cheques 4.9× the campaign average. Nobody was measuring this while it was happening.

Registrations

across 3 webinars

Attendees

46.8% show rate

Investors

of 52 matched in CRM

Conversion

36.5%

vs 2.89% base rate

Capital

~$120K

24.3% of the raise

Avg cheque

$6,292

4.9× campaign average

Webinar by webinar

Webinar	Date	Reg.	Attended	Show rate	CRM cohort	Invested	Conversion	Capital
Webinar 1 — Testing the Waters	Week 3	17	11	64.7%	13	2	15.4%	$21,000
Webinar 2	Week 8	19	8	42.1%	15	5	33.3%	$53,200
Webinar 3 — the platform investor	Week 20	26	10	38.5%	24	12	50.0%	$45,350
Total		62	29	46.8%	52	19	36.5%	~$120K

Source: HubSpot MARKETING_EVENT objects (Zoom-synced) cross-referenced against the platform_investment_amount per contact. Cohorts built from the Zoom webinar ID in the webinar-registration field.

Attending mattered — roughly double the conversion

	People	Invested	Conversion	Capital
Attended	22	11	50.0%	$88,900
No-show	30	8	26.7%	$30,650

Attendees were 42% of the cohort but delivered 74% of its capital. Even no-shows converted at 9× the base rate — registering alone signals intent.

Capital by channel — nothing else is close

Put this in perspective. Paid media: 969 sessions, ~$7K spent, zero investments. Organic social: ~377,000 views, ~$3.1K traced. Webinars: 62 registrations, effectively no cost, ~$120K. The highest-leverage activity in the raise was the one with the smallest audience and no budget.

The obvious miss: we barely promoted them. Only Webinar 1 got a dedicated invitation email ("Webinar Tomorrow", Week 2). Webinar 2 got a logistics "Time Change" notice; Webinar 3 got only a same-day "1 Hour" reminder. Registrations came from the newsletter newsletter and the platform page by default. Three webinars, 62 total registrations, off a list of 1,000+ deliverable contacts — each converting at 12.6× base rate. The clearest growth lever for the next raise is running more of these and actually inviting people. Note too that the post-webinar recording emails consistently beat the reminders (22–27% opens, 1.5–2.7% clicks) — the recording is the strongest asset in the sequence and was under-used.

Caveats — read before quoting these numbers. This is correlation, not causation. High-intent people self-select into investor webinars, so some of this capital would have arrived anyway. Three specific limits: (1) cohorts come from the webinar-registration field, which stores only a contact's most recent webinar, so Webinars 1 and 2 are under-counted (7 contacts registered twice) and their true conversion is likely a few points lower — Webinar 3, at 92% cohort coverage, is the most reliable. (2) the platform_investment_date is a last-interaction timestamp, not a first-purchase date, so timing is directional only; 2 of Webinar 2's 5 investors — including its $50,000 cheque — appear to have invested before the call. (3) Webinar 3 sat inside the 7 April deadline ramp alongside two "closing soon" emails, so its lift is confounded. To harden these numbers you need the Zoom attendee CSV per event and a real the platform_first_investment_date field.

Total live audience across the campaign

Format	Count	Audience	Confidence
Investor webinars	3	29 attendees (62 registered)	Measured
In-person events & panels	9	~648	1 measured, 8 estimated
Total live audience	12	~677

Only one in-person event has a recorded figure — a legislators caucus Plastics Learning Summit tour on Week 3, logged in an update thread as "25-30 people", state legislators. Everything else is estimated from event format.

Total activities

~50

PR, events, partners, guesting

Media placements

across the full window

Est. earned reach

~138,400

22.9% of all campaign reach

Events

+ 3 skipped

Guest podcasts

aired in window, of 10 pursued

Traced investment

~$3.8K

a Reg CF research platform + a paid investor newsletter

Activity for this campaign is split across two Monday boards — an archived “ - Marketing Activity Calendar” covering the launch period and a separate board — with nothing bridging them. Everything below is consolidated from both, cross-checked against the public record. Reach is recorded on only 6 of 37 launch-period items and on none of the items; figures marked est are modelled, and the methodology is set out at the foot of this tab.

All media placements and mentions, Week 1 – Week 23

Date	Outlet / account	What	Type	Reach	Note
Week 1	Founders & Empanadas	the chief product officer guest appearance	Guest podcast	~300 est
Week 4	a Reg CF research platform	Deal page, ratings and financials	Investor research	~3,000 est	drove ~$3.7K / 3 investments — best-converting placement of the campaign
Week 4	the lead VC fund	Portfolio page, Q4 cohort	Investor	~400 est	lead investor's own site
Week 4	a lead-fund partner (the lead VC fund)	LinkedIn post	Advocacy	~800 est
Week 4	a raise-listing aggregator	Automated Reg CF raise listing	Aggregator	~300 est
Week 4	a metro business daily	“a US metro water-vending startup Northwind raises $1M” — a business-daily reporter	Tier-one press	~12,000 est	the only tier-one placement of the raise
Week 4	a regional tech newsletter	Newsletter — CBJ article featured	Newsletter	~7,500 est
Week 4	the platform newsletter	“4 New Startups This Week” feature	Platform	10,000	recorded
Week 4	a regional VC podcast	the CEO guest appearance + LinkedIn post	Guest podcast	~1,700 est
Week 4	a high-profile founder-investor (the founder-investor newsletter)	“Invest in the CEO” — explicit call to invest	Advocacy	~9,000 est	Bonobos founder; 6 likes on-platform
Week 4	a lead-fund co-founder (the lead VC fund)	LinkedIn post	Advocacy	~1,600 est
Week 4	a community advocate	LinkedIn post	Advocacy	~500 est
Week 4	a news aggregator	CBJ syndication	Syndication	~500 est
Week 4	an investor newsletter	“the platform Wednesday” feature	Newsletter	~1,750 est
~Week 5	a B2B data vendor	Auto-generated funding page	Aggregator	~200 est	contains factual errors; feeds a B2B data API
Week 5	a sustainability newsletter	Newsletter feature	Newsletter	~1,050 est
Week 5	a regional tech newsletter	Newsletter — second placement	Newsletter	~7,500 est
Week 6	a regional broadcast segment	Segment with Tony	Broadcast	1,700	recorded
Week 6	a values-led business podcast	the CEO guest appearance	Guest podcast	191	recorded; host noted “we've not kept stats”
Week 6	a startup-community newsletter	Newsletter	Newsletter	1,000	recorded
Week 7	a hard-tech podcast	“Investing in disruptive hard tech” — the CEO	Guest podcast	~400 est
Week 7	Impact Investor White Paper	Inclusion + LinkedIn post	Report	2,000 + ~800	LinkedIn reach recorded
Week 10	a climate-tech newsletter	Social posts	Partner social	~3,000 est
Week 11	a finance newsletter	Newsletter	Newsletter	~1,500 est
Week 11	the lead VC fund	LinkedIn post on the episode	Advocacy	~2,000 est
Week 12	a climate-tech newsletter	Thought-leadership guide citing the podcast as its worked example — unprompted	Earned citation	~12,000 est	15 likes — most engagement of any a climate-tech newsletter item
Week 15	a climate-tech newsletter	the CEO guest essay, “Unlocking the Capital Stack”	Bylined	~12,000 est	generated a live financing inbound in the comments
Week 18	a high-profile founder-investor (the founder-investor newsletter)	“The Double Alpha Household” — full Northwind endorsement	Advocacy	~3,000 est	17 likes
Week 19	a climate-tech newsletter: Off the Record	48-minute interview; a full chapter on “Why Choose Crowdfunding”	Guest podcast	~2,500 est	8 likes
Week 19	the newsletter host	LinkedIn video promoting the episode	Advocacy	~2,500 est
Week 21	a sustainability newsletter	Newsletter with the platform CTA	Newsletter	~1,050 est
Week 22	a high-profile founder-investor (the founder-investor newsletter)	Full newsletter issue on his the podcast episode — embedded video, the platform CTA	Advocacy	~3,000 est	~7,000 subscribers; the episode's LinkedIn post hit 4,048 impressions at 12.5% ER
in-window	a paid investor newsletter	Paid placement	Paid newsletter	~5,000 est	drove $100 / 1 investment

PR pipeline

Item	Date	Status	Reach
a metro business daily — a business-daily reporter	Week 4	Done	~12,000 est
a regional tech newsletter newsletter — CBJ article	Week 4	Done	~7,500 est
the platform newsletter feature	Week 4	Done	10,000
an investor newsletter — the platform Wednesday	Week 4	Done	~1,750 est
a sustainability newsletter newsletter	Week 5	Done	~1,050 est
a regional tech newsletter newsletter — second placement	Week 5	Done	~7,500 est
a regional broadcast segment	Week 6	Done	1,700
a startup-community newsletter newsletter	Week 6	Done	1,000
Impact Investor White Paper + LinkedIn post	Week 7	Done	2,000 + ~800
a finance newsletter — newsletter	Week 11	Done	~1,500 est
a climate-tech newsletter guest Substack	Week 14	Done	~12,000 est
a sustainability newsletter — the platform featured	Week 21	Done	~1,050 est
a climate-tech newsletter social posts	Week 10	Working	~3,000 est
a national-press contact — Forbes article	—	Working	—
a circular-economy programme website feature	—	Working	—
JLL video release	—	Working	—
a case-study author BizCafe case study	—	Later	35,000 planned
a climate accelerator × a partner org content	—	Stuck	—
NBC / Today Show — tied to a national TV meteorologist episode	—	Rejected	—
Go Fund Yourself	—	Skipped	—
a climate accelerator newsletter	—	Skipped	—

12 of 21 landed — a 57% strike rate, better than the 23% the board alone suggested. The launch period carried it: 9 placements in five weeks from Week 4. The highest-value target, Forbes via a national-press contact, never closed, and the one item with a real reach projection attached (35,000) was deferred.

Events

Event	Format	Date	Status	Audience
an innovation event a retailer innovation day	In person	Week 1	Skipped	—
1871 Tech For Good Innovation Summit	In person	Week 2	Skipped	—
Webinar 1 — Testing the Waters	Virtual	Week 3	Done	11 of 17 reg.
a legislators caucus Plastics Learning Summit tour — state legislators	In person	Week 3	Done	28
the CPO @ a founders network, NYC	In person	Week 4	Done	~60 est
MZ speaking, a business school	In person	Week 4	Done	~45 est
a pitch showcase	Virtual	Week 4	Done	~80 est
a regional showcase — MZ panel	In person	Week 6	Done	~75 est
an industry event Manufacturing & Supply Chain Day	In person	Week 6	Skipped	—
an impact-investing conference Conference	In person	Week 7	Done	~120 est
a partner event	In person	Week 7	Done	~40 est
Webinar 2	Virtual	Week 8	Done	8 of 19 reg.
Independent Schools Alumni, a regional tech hub	In person	Week 17	Skipped	—
a policy advocacy day, the national capital	In person	Week 18	Done	~150 est
Webinar 3 — the platform investor	Virtual	Week 20	Done	10 of 26 reg.
a plastics-advocacy nonprofit webinar with Useful & a kids-tableware brand	Virtual	Week 22	Later	—
a partner event event	In person	Week 21	Done	~50 est

Total live audience ~677. Only the three webinars and a legislators caucus tour have real figures; everything else is estimated from format. The three webinars are the only events whose investment outcome is measurable — and they produced ~$120K. See the Webinars & Live tab.

Partnerships & co-marketing

Partner	Activity	Status
a climate-tech newsletter	Unprompted case study (Week 12) + guest essay + social + podcast	4 touchpoints
the lead VC fund	Portfolio page, 2 LinkedIn posts, episode — most-reshared post of the season (7)	Active
a regional water NGO	Episode (Week 7) + hosted the CEO for a policy advocacy day (Week 18)	Active
Current (a water-nonprofit director)	Episode (Week 9) — 2,130 LinkedIn impressions, an outlier	Active
a kids-tableware brand (a pediatrician-founder)	Episode (Week 6) + Instagram collaboration	Active
a sustainability newsletter	Two newsletter features (Week 5, Week 21)	Active
a legislators caucus	Plastics Learning Summit tour for state legislators (Week 3)	Done
a startup-community newsletter	Newsletter (Week 6)	Done
an impact-investing network	Case study posted to the platform page (Week 13)	Done
a circular-economy programme	Website feature	Pending
a plastics-advocacy nonprofit + Useful + a kids-tableware brand	Co-hosted webinar	Later
a climate accelerator + a partner org	Joint content — the CEO recorded, partner deprioritised	Stuck

a climate-tech newsletter was the standout, and warmer than anyone logged — they cited the podcast as a case study unprompted on Week 12, three weeks before the CEO's own essay, and that unprompted piece out-engaged both of Northwind's own placements. Worth investing in ahead of the next raise.

Podcast guesting — Northwind on other shows

Show	Who	Date	Status
Founders & Empanadas	the CPO	Week 1	Aired
a regional VC podcast	the CEO	Week 4	Aired
a values-led business podcast	the CEO	Week 6	Aired
a hard-tech podcast	the CEO	Week 7	Aired
a climate-tech newsletter: Off the Record	the CEO	Week 19	Aired
a climate podcast	the CEO	Week 21	Scheduled
a water-sector podcast — its host	the CEO	Week 22	Working
a sustainability podcast — its host	the CEO	rec. Feb, aired Jun	Post-window
a legacy podcast	—	—	Rescheduling
a packaging podcast	—	—	Unscheduled

5 of 10 aired inside the campaign, four of them in the launch period. Guesting has a long lead time — a sustainability podcast episode was recorded in February and aired in June, two months after the raise closed. Book guest slots three to four months before the window opens.

Guest amplification and third-party advocacy

Guest collaborations were the single most effective reach mechanism in the campaign, delivering roughly 25–40× Northwind's entire owned Instagram audience in borrowed attention. One control worth noting: Meta ads didn't start until Week 17, so every December and January outlier below is 100% organic.

Borrowed reach

~25–40K

impressions from guests & partners

vs owned IG audience

~25×

against ~1,010 followers

LinkedIn reshares

third-party, across the season

Addressable guest audience

~109K

only ~25% realised

a national TV meteorologist Instagram Collab — the best asset of the campaign

Date	Format	Views	Reach	Engagement	vs baseline
Week 12	Reel	14,533	10,418	6.7% · 700	22× reel median · 10.3× follower base
Week 12	Reel	2,254	2,068	1.2% · 25	4.4× reel median
Week 13	Carousel	5,507	3,840	1.2% · 49	20× carousel median · 8.1× next-best carousel

The carousel is the proof. Instagram gives carousels almost no algorithmic distribution to non-followers — the other 18 carousels have a median reach of 193 and a maximum of 474. The 28 January thank-you carousel reached 3,840 organically with no ad spend. Nothing organic does that to a carousel except the post appearing in a second, much larger feed. @the meteorologist has ~47,000 followers against Northwind's ~1,010 — a 46:1 ratio. The low engagement rate on the carousel and second reel is the signature of a large borrowed audience with weak affinity, which is exactly what a Collab produces.

a kids-tableware brand

Posts tagging @the kids-tableware brand in December sit in normal organic range — best reach 1,928 against a no-guest reel that reached 1,885. The likely explanation is that the Collab ran the other way: a kids-tableware brand owned the post and Northwind was the invited collaborator, so the metrics live in their account and never appear in the Sprout export.

~45,000 followers — a comparable prize to the meteorologist. Verify by checking their feed.

a high-profile founder-investor

Week 22 — full the founder-investor newsletter issue on his episode	~7,000 subs
Week 18 — Northwind endorsement post	17 likes
Week 21 — episode's LinkedIn post	4,048 · 12.5% ER

Worth stating plainly for external audiences: a high-profile founder-investor is the CEO's spouse and the second-largest investor, and discloses this himself. Genuine amplification, but not third-party endorsement.

a regional water NGO

The Week 7 a water-NGO chief executive reel did 3,412 views / 2,693 reach at 6.8% engagement — 5.7× the reel median and 2.7× the then-follower base, fully organic, with high engagement as well as high reach. Consistent with a Collab or a reshare by @the water NGO.

LinkedIn — 41 third-party reshares

Episode post	Impressions	Reshares
Launch teaser (Week 3)	4,920	5
Ep6 — a water-nonprofit director / Current (Week 9)	2,130	5
Ep7 — the lead VC fund (Week 11)	1,307	7
Ep11 — a high-profile founder-investor (Week 21)	4,048	3
All other episodes	474–1,752	2–3 each

Current and the lead VC fund are clear impression and reshare outliers against a 500–900 baseline, pointing to both organisations amplifying. LinkedIn's company export does not name resharers.

Third-party advocacy roster

Advocate	Platform	Date
a national TV meteorologist	Instagram Collab	18–Week 13
a kids-tableware brand (a pediatrician-founder)	Instagram Collab	Week 9
a regional water NGO	Instagram	Week 7
a high-profile founder-investor	Substack ×2	Week 18, Week 22
a lead-fund partner (the lead VC fund)	LinkedIn	Week 4
a lead-fund co-founder (the lead VC fund)	LinkedIn	Week 4
a community advocate	LinkedIn	Week 4
the lead VC fund	LinkedIn	Week 11
the newsletter host (a climate-tech newsletter)	LinkedIn	Week 19
41 unnamed LinkedIn resharers	LinkedIn	Week 4 – Week 22

The lever that worked was never scaled. The planned "100+ micro and mid-tier influencers posting in the first 2–4 weeks" did not run — the Influencer label is used by 0 of 195 items on the campaign board, and no paid activation of that size appears anywhere in the public record. But the handful of guest collaborations that did run outperformed everything else per unit of effort. Total addressable guest audience across the meteorologist, a kids-tableware brand and a high-profile founder-investor alone was ~109,000, of which roughly 25% was realised. Systematise the collab; drop the volume play. Make a Collab invite and a Story-reshare ask a standard, tracked deliverable for every guest.

Estimated reach and methodology

Press & newsletters

~93,150

27 placements

Guest podcast reach

~2,591

5 shows aired

Live audience

~677

12 events incl. 3 webinars

the platform platform

~42,000

estimated, band 17–67K

How these estimates were built. Only 14% of the ~105,700 earned impressions is measured — the six Reach values recorded on the archived board. The other 86% is estimated: newsletters as subscriber count × typical open rate, press from outlet size and story type, LinkedIn posts at roughly 8% of following, events from format. Where a platform showed engagement signals (a climate-tech newsletter's 7–15 likes, a high-profile founder-investor's 6) the estimate was sized down. Treat these as order-of-magnitude, not reportable metrics — a ±50% band is realistic. The useful conclusion is the ratio: earned media added roughly 28% on top of the ~377,000 owned-social reach, so about 22% of all campaign reach came from earned — a real contribution, achieved with no budget.

What the public record does not show

Absence	Why it matters
No trade press at all — nothing in the vending, beverage and packaging trade press; no mention at the the industry trade show	The biggest gap. The trade press that serves Northwind’s actual buyers — vending, beverage and packaging operators — never covered the company or the raise, so the audience most likely to become customers never saw it.
No customer announcements from a federal agency, a named enterprise customer, a named enterprise customer, a named enterprise customer, a named enterprise customer or a named enterprise customer	Every customer name traces back to Northwind-controlled copy, never to the counterparty's own channel. Co-announcements are free credibility.
No a US metro press beyond CBJ — nothing from Crain's, a metro news outlet, a regional tech newsletter editorial, a local outlet, a local station or the Tribune	Local press is the cheapest reach available to a US metro hardware company, and two named targets never landed.
No national tech or climate press — no a tech outlet, Forbes, Axios, a climate outlet or a sustainability outlet	Consistent with organic social and guest collabs being the only real awareness engines.
No awards, accelerators or pitch wins in the window	The entire award trail predates Oct. Milestone news is a reliable press hook and there was none to use.
Zero forum footprint — no Reddit, Hacker News or X discussion; @HelloNorthwind is dormant but still linked publicly	No organic word-of-mouth layer formed. Also a hygiene fix: stop advertising a dead account.

What cannot be verified externally. Instagram Stories are checkable only by Northwind — they expire in 24 hours and are archived nowhere public, and the export covers only Northwind's own 230 stories, not guests'. Since Story reshares are the most common form of podcast-guest amplification, guest reshares are entirely plausible and simply cannot be evidenced from outside; settle it from the IG story-mention notifications or DM archive. Also unverifiable: the 41 LinkedIn resharers are unnamed in the company export, guest Instagram feeds cannot be browsed directly, and there are no public podcast-aggregator share signals.

Three quick wins. Screenshot the Collab attribution line on the 18 and 28 January posts to convert the meteorologist from strong inference to documentary proof. Pull a per-domain referral report from GA4 — it would show traffic arriving from the kids-tableware brand.com, greatlakes.org and substack.com directly, and is the highest-value follow-up available. And ask the other lead-fund co-founder to add a metro business daily story to the lead VC fund's portfolio page, which currently omits it.

Before this goes to the exec team, know where it is soft. The raise total, investor count, spend and all platform metrics are solid. Channel attribution is not, and four channels were never measured at all. Anyone who pushes on "so which channel drove the raise?" should get the honest answer: we can account for 18.4% of it.

Confidence by metric

Metric	Value	Confidence	Basis / caveat
Total raised	~$490K	High	the platform dashboard, campaign closed Week 23
Investors / investments	371 / 384	High	the platform dashboard
Marketing spend	~$28K	Medium	Two versions of the tracker disagree by $1,580 (~$28K SharePoint vs $29,525 OneDrive)
Top of funnel, all channels	~600,000	Medium	Owned (~420,000) and paid (~45,000) measured from platform exports; the entire earned group (~138,000) is modelled. Sums non-equivalent units, not unique people
the platform platform reach	~42,000	Low — modelled	Backed out from 61 confirmed investments at assumed conversion rates. Band 17,000–67,000
Podcast reach multiplier	62×	High	YouTube duration split + caption/keyword classification of 141 cutdowns
Traced investment by channel	~$90K	Medium	the platform UTM export — real, but covers only 18.4% of dollars
Channel share of total raise	—	Low	Cannot be computed. 81.6% of dollars have no recoverable source
Paid media contribution	$0	High	Zero across every paid UTM, corroborated by GA4 (1 key event)
Events / partnerships / podcast guesting	19 activities	Medium	Consolidated from the archived board and the board; no reach or attendance figures on most items
a national TV meteorologist IG Collab	~16,800 reach	High	Carousel reaching 20× its median with zero ad spend has no other organic explanation
a kids-tableware brand IG Collab	—	Unresolved	Not visible in Northwind's export; likely owned by a kids-tableware brand's account, so metrics sit there
Guest Story reshares	—	Unverifiable	Stories expire in 24h and are archived nowhere public. Only Northwind can confirm
Mass influencer programme	0	High	The planned "100+ influencers" never ran, in any system or the public record
the newsletter performance	13 issues	High	Full per-issue HubSpot metrics
newsletter share of webinar registrations	30–50%	Inferred	No registration-source field exists; click share is an upper bound (newsletter has ~12× the links)
Earned media activity	28 placements	High	Verified across the archived board, the board and the public record
Webinar registrations / attendance	62 / 29	High	HubSpot MARKETING_EVENT objects, Zoom-synced
Webinar investment conversion	~$120K	Medium	Per-contact amounts verified individually; cohorts under-count webinars 1–2, timing directional only
Estimated earned reach	~105,700	Low — modelled	86% estimated from outlet/event type; ±50% band. Order-of-magnitude only
Earned media reach / EMV	—	None	Reach and EMV columns filled on 1 of 41 activity-board items
HubSpot channel attribution	—	Unusable	2 touchpoints / $530 recorded for the entire campaign; 97.9% of investors tagged Offline/integration

How the ~600,000 top-of-funnel figure is built

Every channel that put a Northwind impression in front of a person during the campaign window, grouped owned / earned / paid. Rows marked counts in the channel table are summed; subsets, borrowed reach and no-data rows are excluded.

Group / channel	Metric as exported	Value	What it counts
OWNED
TikTok	Total Video Views	188,401	a view registers at ~0 seconds
Instagram	Views	121,732	Posts 111,507 + Stories 10,225. Reels nested inside Posts, excluded
YouTube (organic)	Views	35,132	42,620 total less 7,488 served by ads, now in the paid group
LinkedIn (owned page)	Impressions	34,416	impressions, not views. Sponsored component is zero
Spotify	Plays	201	first-7-day plays only
Email / newsletter	Emails delivered	40,380	an impression opportunity, not a view
Investor webinars	Attendees	29	measured; 62 registered
Owned subtotal		~420,000	69.6% — all measured
EARNED
Press, newsletters & third-party posts	Modelled reach	93,150	86% estimated from outlet size and open rates
the platform platform	Modelled reach	42,000	fully estimated — model below
Guest podcasts	Modelled listens	2,591	5 appearances; 1 recorded, 4 estimated
Speaking events	Attendance	648	1 recorded, 8 estimated from format
Earned subtotal		~138,000	22.9% — almost entirely modelled
PAID
Meta (Facebook + Instagram)	Impressions	37,400	measured, from Ads Manager
YouTube ads	Views	7,488	measured, from the YouTube traffic-source report
Google Search	—	0	182 GA4 sessions but no impression export, so uncounted
Paid subtotal		~45,000	7.4% — measured
TOTAL TOP OF FUNNEL		~600,000	29% of the 2.1M plan

How the platform platform estimate was built. the platform never shared send or feed reach, so this is backed out from confirmed conversions against the one figure they did give us: the Week 4 “4 New Startups This Week” newsletter feature reached 10,000. Applying plausible conversion rates per surface — dedicated blast 40 investments at 0.20% (~20,000), nine newsletter sends 15 investments at 0.10% (~15,000), feed and activity digest 3 investments at 0.05% (~6,000), closing and transactional reminders 3 investments at 0.30% (~1,000) — implies ~42,000 impressions at a 0.15% blended conversion. Realistic band 17,000–67,000. It is the least certain number on this tab, and the fix is simply to ask the platform for send counts next time.

The one de-duplication actually performed. Instagram exports Posts and Reels as separate files, but Reels are a subset of Posts — 86 of the 100 reels share a timestamp with a post, and the Posts export contains exactly 86 rows of media type video. Summing both files would have double-counted 102,346 views and produced a false 480,000-ish total. Only the Posts figure is used. Instagram Stories (230 items, 10,225 views) are excluded as a separate surface; including them would give ~387,000.

What is deliberately excluded, and why. the newsletter (14,642) and CTA sends (25,738) are subsets of the email row. Guest collabs and advocacy (~30,000) is borrowed reach arriving through the Instagram, LinkedIn and press rows — a national TV meteorologist collab’s 22,294 views are already inside Instagram’s 121,732, and a high-profile founder-investor’s Substack is already inside the press row. Instagram Reels (102,346) is nested inside the Posts export. Substack, partnerships and Google Search contribute zero because their reach was never captured. Summing every row naively would give roughly 780,000 — about 30% too high.

What this figure is not. It is not a count of unique people, and not de-duplicated across channels — no platform exposes identity to any other, so someone who saw the same podcast cutdown on TikTok, Instagram and YouTube is counted three times. It is also not de-duplicated within a platform: TikTok recorded 1.07 views per person reached and Instagram 1.39, so even the single-platform figures exceed the people behind them (TikTok reach 176,116 vs 188,401 views; Instagram reach 80,498 vs 111,507 views). And it sums non-equivalent units — LinkedIn impressions sit alongside TikTok and Instagram views. And 23% of it — the entire earned group — is modelled rather than measured. Owned and paid are hard numbers from platform exports; earned is an estimate. Treat ~600,000 as directional, useful for comparing groups against each other and against the 2.1M plan, not as an audience size.

Paid is now separated properly. The 7,488 YouTube views served by ads have been moved out of the YouTube row into the paid group alongside Meta’s 37,400 impressions, so owned reach is genuinely organic and paid is complete for the two platforms with exports. One gap remains: Google Search shows 182 GA4 sessions but has no impression export, so its reach is uncounted. Note also that no separate Google Ads spend figure exists — the ~$2K media line in the budget tracker is Meta only, so the YouTube ad views likely sit inside the $5,000 the paid-media agency fee.

Reconciliation — why numbers differ between sources

Source	Raise stated	Why it differs
the platform dashboard (use this)	~$490K	Final, campaign closed Week 23
HubSpot the platform_investment_amount	~$450K	312 contacts; sync missed ~59 investors
HubSpot the platform_status = invested	~$387K	283 contacts; status field disagrees with amount field on ~29 records
Budget tracker "total raised"	~$345K	Snapshot ~Week 18
Budget tracker monthly sum	~$330K	Oct–Jan only; Feb and Mar columns were never filled in
Week 12 memo	~$321K	As of Week 13
Outbound copy	$330K–$360K	Live figures quoted at various send dates

Your tracker understates the final raise by $147,417 because it stops at January. Nearly a third of the money arrived in Feb–Apr, including a sharp final-week surge.

Fix these five things before the next raise

Run many more investor webinars, and actually promote them. They converted at 12.6× base rate and produced 24.3% of the raise from 62 registrations — off a single real invitation email across three events. Highest-confidence growth lever available.
Tag everything. Every cutdown, every post, every bio link, every placement gets a unique UTM before it publishes. This one change converts 81.6% dark into something measurable.
Keep PR, events and partnerships on one board that spans the whole campaign, and fill the reach columns. Activity is currently split across an archived board and a separate one with nothing bridging them, and reach is recorded on only 6 of 37 launch-period items. And book podcast guesting 3–4 months ahead — most of what was pursued aired after the raise closed.
Systematise guest collaborations — this is the real influencer play. One Instagram Collab out-reached everything else in the campaign. Only ~25% of the addressable guest audience was realised. Make a Collab invite and a Story-reshare ask a standard, tracked deliverable for every guest.
Either run the mass influencer channel or stop planning it. "100+ influencers posting in the first 2–4 weeks" was a stated pillar that never happened and that nothing was tracking. A named backlog of relevant manufacturing creators was identified and never contacted.
Model investments as deals in HubSpot. They were never created as deals, which is why HubSpot's attribution engine scored $530 against a ~$490K raise.
Keep the tracker current to the campaign close. Stopping in January is why the internal number was $147K light.
Negotiate platform-audience access up front. the platform's own list was the single largest traceable driver at ~$30K. Treat that as a lever to bargain for, not a bonus.

Northwind × the platform campaign retrospective · built after close · raise closed Week 23
Sources: the platform dashboard & conversions-by-traffic-source export · GA4 (Northwind the platform Analytics) · HubSpot portal a campaign code · Monday boards a campaign code / a campaign code · Amplemarket (4 sequences) · Meta Ads Manager · Instagram · TikTok · LinkedIn Page analytics · YouTube Studio · Spotify for Creators · SharePoint Marketing/Campaigns//the platform