Ask HN: How to obtain a semi-accurate appraisal of a business entity?
Hello, all,
I have a question that we've been struggling with recently:
Do you know of any known, established official (or even completely unofficial) useful metric by which a brand or website, or a web application, or perhaps even something like (what I refer to as a) 'business entity' (such as an LLC or Sub-S Corp, for instance) can be appraised of it's worth?
For instance, say that when you re-brand a business entity's website by investing several thousand dollars into graphic design services, or otherwise "upgrade" a site by investing an equivalent amount of money or sweat equity, what are some ideas for establishing ballpark figures for:
A. How much the re-branding initiative costs overall,
B. Approximately what you consider the business (or website) to be 'worth' before the project,
and,
C. What the business (or website) is subsequently valued at after the investment in rebranding.
My firm is currently dabbling in a very low-tech method by which we can fairly and accurately establish reasonable estimates of site/brand/business valuations, and we're having a tough time because many of the projects we are involved with are much smaller than the established market journals recognize. (i.e. I'm pretty sure that TechCrunch won't run an article on the last website we built, because it was a pretty cookie-cutter site for a large national law firm, and subsequently of little to no interest, even though we were adequately compensated in our opinion)
We feel that although our sites aren't standard "startup" fare, (or at all even slightly interesting to the hardcore developers of HN) the sites in question hold considerable value for their owners, even if they are not massively popular like TwitBookSpaceGoogPokerStarzWhatEver is; they generally get steady streams of high-quality, discerning visitors that are sincerely interested in the content, they don't have a lot of overhead in terms of hosting or server hardware, they have solid, recognizable logos, typography, & content, consistent and acceptably "catchy" brandnames and/or domain names, and aren't shabby on Web 2.0 functionality like tag clouds, social networking widgets, & regularly updated content that auto-refreshes the site when the owners get too busy to post original work.
(Some of our other projects are actually pretty cool, like Perl/Catalyst-based user-submission driven news sites similar to Reddit or HN, but assisted in content gathering and refreshing by a bunch of Perl/Plagger modules, but I digress...)
So, anyway: I hope this isn't something that's been covered extensively in the past (if it was, I missed it, sorry) but I think these are interesting questions as not all web developers aspire to build the next Big Thing.
Do you have any advice?
What do you do, when a business person who owns a website asks you to place a value on the website, or, (to make it even more complicated), do you get asked to appraise, say, businesses who own not just one domain, but rather, collect 'portfolios' of websites and operate them under the corporate umbrella provided by the primary (typically incorporated) business entity?
Maybe the best way to phrase a question of this nature would be to say, "How does Y Combinator make semi-accurate, ballpark-ish guesstimates when evaluating the value of the smaller projects they fund?"
(which also then leads to the question, "Um, I'm not trying to compete with Y Combinator here or anything, but, uh, if a hypothetical company was trying to compete with PG, uh, what's a good equation to use to evaluate sunk costs and stuff?"
Don't laugh - I'm really curious about this stuff.) Last time I did this (10ish years ago) the standard was a product called "eVal" that was basically a huge Excel plug-in. It took FOREVER to enter all the data/variables, but it gave you a pretty good valuation idea. A quick Google search shows there are many similar tools out there today. For what you are describing, a modified/simplified template could probably be developed in a week. I did a little more research and there were indeed quite a few resources to be found that I had initially overlooked when I last pursued it. And, coincidentally, before I posted this I spent a long while working with a Google Docs spreadsheet to see whether or not I might be able to hack together some sort of rough guesstimation tool or primitive convolutedValuationCalculator or something, which turned out to be a little overly optimistic in my case due to my somewhat limited spreadsheet chops. (Which is not to say that I didn't learn a lot from playing with the data. Also, it is impressive to me that Google Docs has a spreadsheet application that is actually pretty usable and responsive, I was kind of expecting the worst case scenario but ended up making 3-4 additional spreadsheets for a couple other projects.) However, while fun, my efforts still didn't feel like they were yielding results that could be held up to even minor scrutiny by one of the alternate solutions hawked by most of the top Google results. You know how there are some areas of software where Open Source has for all intents and purposes been able to provide options that mirror the feature sets and overall functionality of their proprietary analogs? Well, unfortunately, to me it seems that Business Analytics is still something of an Open Source ghetto, with few viable options to be found, and those scarce resources and projects that do exist don't seem to have the features I'd most benefit from. Wrapping up, though, I was able to come up with a few very primitive views and models for this project that were slightly helpful; however, as I mentioned before, I am no spreadsheet shaman by any stretch of the imagination, and although I'd love to take the time to practice those skills, it is honestly not a viable option this month and we need software with this functionality quickly. My feeble efforts did "work", (well, sort of), which was interesting, but I lack confidence in the way I approached the problem and overall my efforts still felt like I was doing a lot of voodoo-powered guesswork, which is probably accurate, and I think that even if I practiced for a few days, I'd probably not be able to handle incorporating the hardcore number crunching and quantitative analysis features necessary to make a project successful. I know in the banking industry there are some standard valuation methods for this sort of thing, although I don't remember much from my Principles of Banking classes now that it's twelve years later. It would be really nice, however, to be able to build something with a strong core foundation, and I suppose the only way to succeed with an approach like that is to work with industry-accepted formulas & methodologies. I forgot to specify another particularly important point, as well: specifically, my main concern (and an area that was conspicuously unrewarding in terms of search results) as I mentioned before, my ideal "goal" would be to find an algorithm or set of algorithms that are (if not universally) accepted by the banking industry, the business valuation industry, intellectual property appraisal services, trademark appraisers, etc. It is important to me that these valuation methods are transparent, as I'd like to develop some Open Source software that utilized them, and I'd like to be confident in the methodologies and formulas before wasting any time writing the software rulesets. But I'm not even sure that such metrics exist. Is the valuation software market as fragmented and wildly variable as it appears? I will probably never know. :-) I did find a few books on Amazon that seemed to contain this type of information, but holy mackerel, I cannot currently justify shelling out $250.00 for a single title! I had no idea that this type of information was still entrenched in proprietary solutions, or even if the theoretical texts are available, they are still prohibitively expenses for my particular situation. Long-term, I think I'm going to attempt to collect a decent sampling of the formulas and algorithms involved, because I think Open Source projects could eventually be made to rival the best of the proprietary options. But short-term, I do have an immediate need for something that can provide decent valuations without being too laborious or convoluted. You're not aware of any Open Source projects that handle this type of thing, by any chance? I did spend a little time Googling for such a project, but it didn't result in anything worth pursuing, unless my search queries were not using the correct verbiage, which is possible. I understand that business analytics in general is one area where the Open Source software equivalents are still not compelling alternatives to the proprietary offerings, and I am trying to explore the reasons why this is so. I understand that the legal expertise is an expensive problem domain, and probably represents a large barrier to entry, which is also probably true of the MBA & higher-level education involved in enhancing the rulesets and programmatic routines, and also that analytics projects may not be appealing to Open Source programmers necessarily due to the fact that they are not very glamorous compared to certain other projects. And quantitative analysis of the data sets is also quite clearly a highly specialized and difficult realm, I'm sure, which is to say nothing about the subjective stuff that might be involved - I'm thinking stuff like "user group studies" to measure brand impact and loyalty trend information, which is probably overthinking it of course, but any additional thoughts into what might would yield insights and statistically solid data points might be useful. However, maybe that's overthinking the problem, which I tend to do sometimes. I guess I'd be happy with a super simple scoring method that provided a little feedback about "consumer's overall brand satisfaction score", or something. Back to the old drawing board for now - thanks for the response and information, brk. - Mark