LinkedIn Content Strategy for a Vertical AI Platform

01 · The brief

A category-defining product with no consistent channel strategy.

The platform builds Vertical AI purpose-built for banking and financial services, spanning collateral management, prime brokerage, structured products, and direct lending. The target audience is senior: Front Office, Middle Office, and Operations leads at global tier-1 institutions who are time-poor and highly selective about the content they engage with.

When I joined, LinkedIn was the main outbound channel but had no strategy behind it. There was no editorial structure, no consistent cadence, and no single person responsible for it. I was brought in to own that from the ground up.

The bulk of day-to-day output was client-facing and confidential: pitch decks, bespoke product presentations, and meeting materials produced for client calls and demos. LinkedIn sat above that as the brand-building layer, reaching the same decision-makers through thought leadership and product positioning.

Target audience: Head of Collateral Management, VP Operations, Front Office Technology leads, and Chief Risk Officers at institutions including HSBC, Nomura, Barclays, J.P. Morgan, Citi, Bank of America, and DRW.

HSBC

Nomura

Barclays

J.P. Morgan

Citi

DRW

02 · Content strategy

I gave each product vertical its own weekly slot. Forty-two posts. Ten months. One consistent editorial voice.

Every post had a job: build the "Vertical AI vs. general-purpose GenAI" positioning and reinforce one of the four product lines. Rotating through each vertical on a monthly basis meant the feed read as a coherent product story rather than a series of disconnected announcements.

Week 1

Direct Lending

Origination, servicing and compliance workflows, connecting AI capability to everyday pain points for lending teams.

Week 2

Collateral Management

Post-trade efficiency and regulatory compliance, tied closely to SRP Europe and the Collateral Management Summit.

Week 3

Structured Products

NLP-driven term-sheet automation, positioning Vertical AI in a $7tn market where unstructured data is the main bottleneck.

Week 4

Prime Brokerage

Front-to-back automation narratives bridging Front Office and Operations for buy- and sell-side firms.

03 · Process

End-to-end content production: research, copy, design, scheduling and reporting.

Full ownership across every stage of the content lifecycle, from audience research to post-campaign analytics. Quality over quantity: every post was planned, reviewed, and tied to either a product pillar or a live event.

Step 01

Audience research & competitive analysis

Used LinkedIn Sales Navigator to map BFSI decision-makers, track competitor content, and identify the angles and formats that drove engagement in the space before writing a single post.

Step 02

Content calendar & editorial planning

Built a rolling monthly calendar in Hootsuite, replacing manual Excel tracking, with each pillar on a fixed day and event-aligned content layered in so there was never a gap or a missed opportunity.

Step 03

Copywriting & visual design

Wrote every post in a direct, B2B-appropriate voice with clear stat call-outs and a "book a demo" CTA. Designed all LinkedIn banners and event graphics in Canva and Figma to keep the feed visually consistent.

Step 04

Client-facing collateral & sales decks

Beyond LinkedIn, the majority of output was confidential B2B collateral: product pitch decks, bespoke meeting materials, and PowerPoint presentations built for daily client calls, demos, and sales conversations.

Step 05

Stakeholder review & cross-team amplification

Set up a structured approval workflow in Hootsuite with the Director of AI Products and Sales Partner. Co-ordinated the wider London team to amplify event content across personal profiles.

Step 06

Analytics & performance reporting

Tracked impressions, clicks, reactions, and shares in LinkedIn Analytics monthly. Used data to prioritise event-led content and rework underperforming formats.

04 · Marketing collateral

Product brochures for each vertical: written, designed, and delivered.

Alongside the social strategy, I produced a four-piece print and digital brochure library, one per product line, giving the sales team credible, on-brand collateral to back every pitch and conference appearance.

Brochure Direct Lending

Brochure Collateral Management

Brochure Structured Products

Brochure Data-Aware AI

05 · Performance

Impressions grew 159% from campaign launch to peak month.

Monthly reach increased steadily as the content cadence was established, with the strongest results in Q4 driven by event activations and a more targeted posting approach. November was the peak month, supported by the Collateral Management Summit and year-end outreach activity.

Vertical AI Platform · Organic impressions by month (Mar – Dec 2024)

MarAprMayJun JulAugSepOct NovDec

Burgundy bars mark Q4, when event-led content and a refined posting cadence drove the strongest results. November peaked at 5,696 impressions, the highest month of the campaign. Total: 33,913 impressions across 10 months.

06 · Results

42 posts. 33.9K impressions. 159% impression growth from launch to peak.

Figures pulled directly from LinkedIn's organic analytics for the company page. A quality-over-quantity approach with roughly one post per week, each tied to a product pillar or live event, and tracked monthly against engagement and reach targets.

Posts published

Mar – Dec, ~1 per week

33.9K

Organic impressions

From my posts alone

3,198

Link clicks driven

59% of all company clicks

11.1%

Avg. engagement rate

Per post across the campaign

680

Reactions

44% of all company reactions

51.7%

Best single-post CTR

Collateral Mgmt Summit recap

159%

Impression growth

Mar to Nov peak month

807

Avg. impressions / post

Across the 10-month campaign

07 · What I learned

What worked. What I'd change.

01
Story beats stat.

The best-performing post was a short "we're heading to NYC" announcement, not a data-heavy thought-leadership piece. Event-led narrative consistently outperformed pure stat call-outs. Lead with the moment; add the number second.
02
A hard CTA changes the outcome.

The Collateral Summit recap drove 51.7% engagement with a single "book a demo" link. Next time, every post gets a CTA, not just event recaps.
03
Comments are the weak link.

Low comment volume across the campaign. Ending posts with a genuine question and actively replying is the fix: comments carry more algorithmic weight than reactions on LinkedIn.
04
Cadence gaps cost momentum.

Posting dips directly correlate with impression drops. I'd batch-produce content ahead of busy periods so the feed never goes cold between events.

LinkedIn content strategy that put Vertical AI on the BFSI map.