Why we built our own deal platform
TL;DR — We were spending more time gluing tools together than running deals. The stack built for enterprise PE doesn't fit lower-middle-market economics, and the stack built for solo searchers doesn't scale past your first close. Arendel is what we built for ourselves at Legacy Capital, and now runs every deal we do.
Six tools per deal
The first time I actually counted, we were up to six.
A CRM to track the pipeline. A shared drive for the diligence files. A separate VDR when a seller's counsel demanded one. A spreadsheet for the model. Email threads for the LOI back-and-forth. A separate note-taking app where our IC memos lived. Slack on top of all of it for the internal running commentary.
Every handoff between those tools lost something. The buy-side broker sent the CIM as a link that expired in seven days. Our analyst re-keyed the financials into the model. The seller's counsel redlined the LOI as a Word doc and emailed it back with tracked changes we then had to reconcile against our own version. When we onboarded a new team member three months into a deal, catching them up meant walking them through six systems.
The tools weren't bad. The problem was that none of them knew about the others. A deal is one thing. The platform was six.
The economics of enterprise PE software don't work in LMM
I'd sit through demos of the platforms built for the big shops. They were good. They also priced at $50K, $80K, sometimes $150K a year per seat once you added the modules, and they were designed around the assumption that you had a diligence team of ten and a portfolio-monitoring team of another dozen.
Lower-middle-market shops don't run that way. A typical LMM sponsor is three to eight people. We're closing four to eight platform deals a year, plus add-ons. We don't have a dedicated ops team to configure the workflow engine. We don't have a data science group building integrations. We need the tool to work on Monday morning, out of the box, for people who have real deals to run and no appetite for admin.
Meanwhile, the tools built for solo searchers and independent sponsors run in the opposite direction — they're basically CRMs with a diligence checklist bolted on. That's fine until you have two people on the same deal, or you close and now you actually need to track quality of earnings, working capital true-up, and post-close integration in something more structured than a Google Doc.
There was a hole in the middle, and the middle is where most of the LMM lives.
The $3K PDF
The moment I decided to build was over a GF Data invoice.
If you're in this segment, you know GF Data. It's the standard benchmark set for LMM multiples — a PDF report that comes out quarterly, aggregating deal data contributed by member firms, and it costs around three thousand dollars a year to subscribe. It's a genuinely useful dataset. It's also a PDF.
I was staring at the invoice and thinking: we already have the underlying data. Every deal we've closed, every deal our peer network has closed, every LOI that ever crossed our desk with a multiple attached. The data exists. Someone just has to build the software to aggregate it, anonymize it, and let you slice it live — by industry, by geo, by deal size — instead of shipping it as a static PDF once a quarter.
That's not a hard problem. That's a weekend of database schema and a UI. The reason it hadn't happened wasn't technical difficulty. It was that GF Data had a fifteen-year head start on the distribution and the trust, and nobody wanted to compete on their turf.
But if we were already going to build a deal platform for ourselves — and every deal that closes on that platform contributes structured data — then the benchmark database isn't a competing product. It's a feature. The dataset builds itself.
That was the moment the business case clicked. Not "we could build a slightly better deal-management tool." That's a hard sell against incumbents with a decade of feature depth. Instead: we build the deal platform, and the benchmark database emerges as a byproduct that's fundamentally better than a quarterly PDF because it's live and drillable.
What we actually built
Arendel is what happens when you take the six-tool stack and collapse it into one.
Pipeline, CIM ingestion, diligence workspace, model attachments, LOI drafting with an AI clause library, IC memo generation, e-signature, closing checklist, working capital true-up, post-close tracking. One system. One source of truth. One place a new team member logs in on day one and can see everything without you having to walk them through anything.
We priced it for LMM economics. Not $80K a year. Not per-seat gouging. Flat monthly pricing that a three-person shop can afford on their first deal, that scales as the team grows, that never surprises you with a "workflow engine module" upsell six months in. You can see the pricing — it's the price we wished we'd been offered when we were shopping.
And we built the benchmark layer natively. Every deal that closes on the platform contributes anonymized data to the shared dataset. Every subscriber can drill into that dataset live — filter by industry, deal size, geo, structure — instead of waiting for the next quarterly PDF. The why-arendel page walks through the specifics.
The dogfooding rule
We use it. That sounds obvious, but it's the single most important discipline for keeping the product honest.
Every Legacy Capital deal runs through Arendel end to end. Every LOI, every IC memo, every closing binder. When something's broken or slow or awkward, we feel it first, and the person who feels it has commit access. There is no product manager writing tickets for a feature nobody on the engineering team has ever needed. The people building it are the people using it on live deals with real money at stake.
That's the standard I want to hold this thing to. The moment we stop using it ourselves, or we start building features for hypothetical customers instead of the deal we're closing this week, the product will start to rot. So we're not doing that.
What this means for you
If you're running an LMM shop and your stack looks like ours used to — six tools, three passwords, one shared drive where nobody can find anything — take a look at what we built. We built it for us. It turns out other shops our size have the same problem.
If you're a broker or intermediary, we have a dedicated flow for pushing your CIMs and teasers into buyer pipelines without the seven-day expiring link routine.
If you want to see who else is running on it, our customers page has the shortlist.
Either way — we're not trying to be the platform for everyone. We're trying to be the platform for the segment we actually live in. That segment has been underserved for a long time, and we think we know why. Now we're doing something about it.