A $2 million EBITDA precision machining company and a $2 million EBITDA software company go to market in the same quarter. The machining company’s acquirers request financial statements, a customer list, and a plant tour. The software company’s acquirers request read access to the billing system, a cohort file going back five years, and an export from the product’s own usage logs. By the second management meeting, they understand the software company’s customer retention better than its owner does. Nothing else in the lower-middle market is inspected this way, and nothing else prices this way.
Everything in the preceding 18 chapters applies to software companies, online businesses, and remote-first technology services firms. This chapter covers what changes through that lens: the pricing logic, the acquirers who concentrate here, the KPIs added to the stack, the diligence probes run nowhere else, and what actually transfers when the company has no building.
Why Technology Prices Differently
Three structural facts separate technology from the rest of the lower-middle market, and every pricing difference in this chapter traces back to one of them.
Revenue durability is inspectable. In most industries, durability is inferred: tenure anecdotes, renewal history, reference calls. In subscription software and much of the online economy, it is observable. Contracts sit in a billing system that timestamps every renewal, upgrade, downgrade, and cancellation. Cohorts can be rebuilt from raw invoice data. Product telemetry shows which customers log in, which features they depend on, and which accounts went quiet two quarters before they churned. An acquirer does not have to take management’s word for the revenue; it can watch the revenue. That inspectability cuts both ways. Genuinely strong retention is provable, so it gets paid for in full. Weak retention is equally provable, and no narrative survives contact with the cohort file.
The economics scale. Software delivers the next customer at close to zero marginal cost, gross margins in the 70-85% range are typical in our experience, and growth does not consume trucks, inventory, or receivables the way physical businesses do. Acquirers underwrite that operating leverage explicitly: each incremental revenue dollar in a scalable model is worth more than the same dollar in a business that must buy assets to serve it.
The acquirer universe is the deepest in the market. Every one of the seven acquirer types profiled in Chapter 3 buys technology companies, and the sector adds dedicated sub-species on top, covered below. Just as important, a software or online company can be bought from anywhere: no acquirer needs to live within driving distance of the asset, so the realistic buyer list is national or global by default. Depth of demand is pricing power, and it is a large part of why Chapter 8’s sector table gave software the widest multiple range in the guide. This chapter unpacks that row.
Valuation in Practice: The ARR Question
Chapter 8 flagged the exception: growing software companies are typically priced on annual recurring revenue rather than EBITDA. The logic is worth stating precisely, because owners routinely apply it to companies it does not fit.
An ARR multiple is not a different religion; it is the same cash-flow math with a different assumption set. When a software company is growing fast, keeping its customers, and reinvesting every available dollar into more growth, its income statement is deliberately suppressed. The acquirer is buying the compounding, not the current profits, so it capitalizes the revenue and models the profit pool at maturity. ARR pricing applies when all three legs hold: real growth, real retention, and a gross margin that proves the profits will exist once the reinvestment stops.
Remove any leg and the logic collapses. Here is the trap that catches more software owners than any other: the profitable-but-slow company. A SaaS business at $4 million of ARR, growing 6% a year, throwing off 30% EBITDA margins, is a genuinely good company. It is not a growth asset, and acquirers will price it on its profits, like a services business, exactly as Chapter 8 said mature software trades. The owner who anchors to ARR-multiple headlines earned by companies growing five times faster declines strong offers, goes stale in the market, and eventually sells for less. If growth is not well into double digits with retention to match, expect an EBITDA conversation, and prepare for it rather than resenting it.
Within ARR pricing, three dials set the multiple. Growth rate is the trajectory the acquirer is buying. Net revenue retention is whether that trajectory compounds without sales effort: a company whose existing customers spend more every year grows in its sleep. Gross margin is how much of each revenue dollar is actually available to fund everything else; a 65% gross margin business and an 85% gross margin business at the same ARR are different assets wearing the same headline. Where a company lands in the wide range Chapter 8 quoted is mostly these three numbers.
Acquirers also run a screening heuristic worth knowing because it will be applied to you: the rule of 40, the convention that a healthy software company’s revenue growth rate plus its profit margin should sum to at least 40. A company growing 50% while burning 10% passes; so does one growing 10% at a 30% margin. It is common practice as a first screen, not a valuation formula, and below roughly $10 million of ARR it gets applied with judgment rather than as a cutoff. But an owner whose company sums to 15 should understand how the first page of the acquirer’s model reads before the first call.
Online businesses sit on the other side of the earnings-base line. E-commerce brands, marketplaces, and content or affiliate businesses are priced on SDE or EBITDA, not revenue, because their revenue does not renew by contract; it must be re-earned from traffic and repeat purchase. The defining discount here is platform dependence. A content business earning most of its traffic from one search engine, a brand selling mostly through one marketplace, an app living in one app store: in each case a third party’s algorithm or policy can reprice the business overnight, and acquirers price that fragility with lower multiples, heavier earnouts, or both. The question from Chapter 8’s e-commerce paragraph is the whole game: is this a brand, or a sales channel that currently works?
Illustrative example. Two SaaS companies, each at $5 million of ARR, go to market the same year. Company A grows 30% annually, with 115% net revenue retention, 93% gross revenue retention, an 80% gross margin, and a CAC payback of 12 months; it runs at breakeven because every dollar goes back into growth. Company B grows 7%, with 96% NRR, a 68% gross margin weighed down by hosting and hand-held onboarding, a CAC payback near 30 months, and a 25% EBITDA margin ($1.25M of profit). In a competitive process, Company A clears 5.5x ARR, roughly $27.5 million. Company B prices at 6.0x EBITDA, roughly $7.5 million, which is 1.5x its revenue. Same ARR, $20 million apart. Company B’s profit is real; what it lacks is compounding, and compounding is what ARR multiples pay for.
The Technology Acquirer Sub-Universe
Chapter 3’s seven acquirer types all show up for technology companies; four sub-species concentrate here and behave distinctly enough to profile.
Vertical-market software consolidators. Operating companies, some very large, built entirely on acquiring niche software businesses that serve a single industry: software for pharmacies, marinas, municipal courts, funeral homes. Their model is decentralized and permanent: keep the brand, keep the team, run the business for cash, and never sell it. Because they underwrite cash yield rather than an exit, they are disciplined on price, but they are also fast, experienced, and close at a very high rate, and they are the natural, honest bid for exactly the profitable-but-slow company described above. For an owner who wants continuity for the product and the team, the hold-forever buyer offers the family-office trade in software form: some price for a great deal of certainty and permanence.
Private equity technology platforms. Funds that buy a platform in a software category and then execute an agglomeration playbook: add-on acquisitions of neighboring products, cross-selling into the combined customer base, long-deferred price increases, consolidated infrastructure, and a disciplined push on net revenue retention. Chapter 8’s multiple-expansion arithmetic applies with extra force here because software multiples step more steeply with scale than services multiples do. For an owner below platform size, the add-on dynamics from Chapter 3 apply verbatim: your earnings may be marked at the platform’s multiple the day the deal closes, and the gap between standalone value and that number is negotiating territory.
Strategics buying capability and talent. Operating companies that acquire to fill a product gap or acquire an engineering team faster than they could hire one. These deals price on what the target does for the acquirer’s roadmap, not on the target’s standalone financials, which is why they can look irrational against any multiples table. They also carry the most structure: retention packages, milestone earnouts, and employment terms doing much of the work, because what is being bought is substantially the people.
Online-business aggregators. The honest history: in the years around 2021, a wave of acquirers raised abundant cheap capital to roll up e-commerce brands and marketplace sellers, and competed multiples in that niche to levels the businesses could not support. Integration proved harder than the spreadsheets promised, the cheap capital ended, and a painful hangover followed; some aggregators unwound entirely. The survivors still buy, but with discipline: lower multiples, more earnout, real operating diligence. The lesson generalizes: a hot acquirer category is a market condition, not a durable pricing floor. Sell into a wave if one arrives; never build a plan that requires one.
The KPIs Acquirers Add
The nine indicators in Chapter 9 all apply to technology companies. Acquirers simply add a layer with sharper instruments, because the data exists to support them.
Retention, benchmarked by model. Chapter 9 introduced GRR and NRR; technology acquirers apply expectations that differ by business model, and they benchmark against the right peer set, not a generic bar. The table below shows how retention expectations typically shift by model, based on FIH experience:
| Model | Typical GRR posture | Typical NRR posture | What acquirers read |
|---|---|---|---|
| Enterprise SaaS, annual contracts | High, commonly 90%+ | Commonly 105-120% | Expansion engine, offset by anchor-account risk |
| SMB SaaS, monthly plans | Lower; steady logo churn is structural | Commonly near or just above 100% | Whether the acquisition engine outruns the churn |
| Usage-based and API products | Depends on committed floors | Can run well above 100% in good years | Cyclical exposure: usage falls without a cancellation |
| Consumer subscription | Lowest of the four | Commonly below 100% | The acquisition machine itself is the asset |
Logo churn versus dollar churn. A company can post a healthy NRR while losing a fifth of its logos every year, because expansion inside a few large accounts masks an exodus of small ones. Acquirers decompose the two, and both matter: dollar retention is this year’s economics, logo retention is the durability of the engine. If your NRR story depends on your biggest three accounts expanding, expect that dependence to be found and priced.
CAC payback. The number of months of gross profit it takes to recover the cost of acquiring a customer. As a rule of thumb acquirers commonly apply, payback under 12 months is excellent, 12-24 months is fundable, and beyond that the growth is being bought rather than earned, which quietly reprices the growth dial. Paired with retention, it answers the only question that matters about a growth engine: does a dollar in produce more than a dollar out, and how fast?
Cohort quality. Not just the retention rate, but the shape: cohorts that decay for a few months and then flatten are annuities; cohorts that decay steadily toward zero are a treadmill. Acquirers plot every cohort you have ever acquired, and the flattening curve is among the most valuable exhibits a software company can produce.
Concentration in new clothes. Chapter 9 treated customer concentration; technology adds four single points of failure that behave identically. One traffic source, where a search or social algorithm update can reprice a content business in a week. One app store, which takes its share and can change the rules. One cloud marketplace or channel partner standing between you and your customers. And one anchor customer on a master agreement, where the whale that is 35% of ARR holds a termination-for-convenience clause. Each is priced the way a 40% customer is priced: through the multiple, through structure, or through a quietly smaller buyer universe.
Key point. In technology, the platform you built on can be your largest concentration risk. Acquirers read dependence on a single algorithm, app store, marketplace, or master agreement exactly the way they read an anchor customer: someone else’s decision can reprice the business, so the risk gets priced, structured onto the owner, or both. Diversification of traffic, channels, and contracts is multiple protection, not marketing strategy.
Diligence: What Gets Added to the Eight Workstreams
Chapter 11’s eight workstreams all run in a technology deal. What changes is that the technology and cyber workstream stops being a supporting act, becomes a center of gravity, and is staffed with specialists. Six probes matter most.
Code and architecture review. Outside engineers read the codebase: architecture that can scale versus one that needs rewriting, dependency health, test coverage, technical debt, and whether anyone other than one person can deploy a release. The finding is rarely “the code is bad.” It is a capital bill: the rewrite or remediation the next owner must fund, priced exactly like the deferred capex Chapter 11 described on the shop floor.
IP chain of title and open source. Diligence confirms that every person who ever wrote code, employee or contractor, signed an intellectual property assignment. The freelancer from 2016 who built the billing module and never signed anything is the technology version of Chapter 11’s contractor finding, and it holds up closings until cured. Alongside it runs an open-source audit, commonly automated: every dependency scanned, every license classified. Permissive licenses are routine; copyleft licenses linked into the product the wrong way can create obligations to release source code, and acquirers treat undisclosed copyleft in the core product as a serious finding. Run the scan on yourself before market, and disclose what you find.
Security posture. As typical practice, formal security attestations such as SOC 2 have been creeping down-market for years: enterprise customers demand them from ever-smaller vendors, and acquirers increasingly expect at least a credible security program, penetration-test history, honest incident disclosure, and sane access controls. The absence of an attestation is rarely fatal at small scale, but its presence shortens diligence, widens the buyer universe, and reads as management maturity.
Cloud cost discipline. Hosting sits inside cost of goods sold, which means gross margin claims get tested against the cloud bills. A company reporting an 80% gross margin while its infrastructure spend grows faster than revenue is presenting a problem it has not priced. Conversely, a documented cost-per-customer trend, improving over time, is the kind of evidence that makes a margin claim underwritable.
Data and privacy. What data the product collects, under what consents, stored where, and subject to which privacy regimes. A large database of consumer data gathered without clean consent is a liability wearing an asset’s clothes, and email lists, tracking pixels, and analytics stacks all get reviewed. Online businesses whose value rests on audience data should expect this probe to run deep.
Key-engineer concentration. Chapter 9’s vacation test, applied one commit deeper: if the lead engineer disappeared for four weeks, could the team ship a release, restore a backup, rotate the credentials? A system only one person can operate is owner dependence in a hoodie, and acquirers price it the same way, then bind that person with retention terms, covered below.
The Remote-First Transfer
Strip away the building and ask what an acquirer of a remote company actually receives: a codebase, a customer base, a set of systems, a body of documentation, and a group of people connected to the company by employment agreements and a chat workspace. Every fear an acquirer has about remote-first businesses reduces to one sentence: what if the company is really just two founders’ heads, and the heads leave?
That fear is rational, and it is also solvable, which is the part owners underweight. Three assets carry the transfer.
The team, retained through change of control. A distributed team is hired from everywhere, which means it can be hired away from everywhere; a change of control is precisely when strong engineers update their options. Acquirers will ask who the five irreplaceable people are, how their compensation sits against the remote market, and what binds them through a transition. An owner should have honest answers before going to market, not during a management meeting.
Documentation as an asset. Runbooks, architecture notes, onboarding guides, decision logs, recorded walkthroughs. Well-run remote companies have a structural advantage here: asynchronous work forces things to be written down, so the operating knowledge already lives in documents rather than hallways. That same corpus is what makes knowledge transfer work after close: a transition that would take months of shoulder-to-shoulder time in an office business runs on recorded video and written process instead.
Systems instead of presence. A remote company that functions has already passed, daily, the exact test an acquirer applies to every business: does it run on written process and measurable output rather than on someone physically watching? This is why, in our experience, a well-documented remote company often transfers more cleanly than an office-bound one where knowledge moves by proximity, and why remote operation widens the buyer universe: an acquirer three time zones away can integrate the company without relocating a single person.
The two futures diverge on one variable. A remote company with real documentation and distributed knowledge is among the most transferable assets in the lower-middle market. A remote company where everything lives in two founders’ heads is among the least, because there is no building, no local workforce, and no institutional memory to catch what the founders drop on their way out.
The advisor’s view. Remotely run software, technology, and online companies are the market FIH works in every day, and the pattern we see is remarkably consistent: owners underestimate how much of their value sits in systems and documentation rather than in the product itself. When we prepare a remote company for sale, much of the work is making the invisible operating system visible: who does what, what is written where, and what would keep running if any two people left tomorrow. The companies that can show that clearly get priced as machines. The ones that cannot get priced as founder risk.
Structure: How Technology Deals Get Papered
Three structural patterns show up in technology deals often enough to plan for.
More rollover, more earnout, by design. As typical practice, high-growth technology deals carry contingent and rolled consideration more often than the broader market, for a structural reason: an acquirer paying an ARR multiple is paying today for a forecast, and structure is how the forecast risk gets shared. Earnouts here are commonly keyed to ARR growth or retention rather than EBITDA, which at least ties the contingency to metrics the owner’s remaining team can influence and a billing system can measure. Owners of genuine growth assets should walk in expecting a rollover conversation, and treat it as Chapter 3 framed the platform deal: a second bet, to be underwritten like one.
Retention packages for key engineers. Where diligence found key-engineer concentration, the purchase agreement’s ecosystem will address it: stay bonuses vesting over 12-24 months, acquirer equity or options, and employment terms negotiated in parallel with the deal. Raise the subject early. Whether those packages are funded by the acquirer on top of the price or carved from the consideration is a negotiation, and it goes better before exclusivity than after.
IP reps and open-source disclosure. Technology purchase agreements carry expanded intellectual property representations: that the company owns its code, that every contributor assigned rights, that open-source usage is fully disclosed on a schedule, that the company has complied with privacy law, and that security incidents have been disclosed. The open-source disclosure schedule deserves particular care, because the automated scan the acquirer ran becomes the checklist against which your disclosure is judged. Chapter 13 covers reps, disclosure schedules, and indemnity mechanics in detail; the technology-specific point is simply that these reps do more work here than in any other sector.
The Bottom Line
- Technology prices differently for structural reasons: revenue durability is inspectable in the billing data, the economics scale, and the acquirer universe is the deepest and most geography-independent in the lower-middle market.
- ARR multiples belong to companies with real growth, retention, and gross margin; profitable slow-growing software prices on EBITDA, and owners anchored to the wrong base go stale in the market.
- Growth, net revenue retention, and gross margin are the three dials of software pricing, and the rule of 40 is the screen acquirers commonly run before any of them.
- Technology concentration wears new clothes: one algorithm, one app store, one marketplace, one master agreement, each priced like an anchor customer.
- Diligence adds code review, IP chain of title, open-source audits, security posture, and cloud cost discipline, and the technical vacation test is the owner-dependence test one commit deeper.
- A documented remote-first company can be more transferable than an office business; an undocumented one is two founders’ heads with a customer list attached.
If you own a software, online, or remote-first company and want a candid read on how it would price through this lens, FIH offers a confidential conversation at fih.com.