6 ms·
The Empty Promise of Data Moats
- charleyma 8y ago"Most discussions around data defensibility actually boil down to scale effects, a dynamic that fits a looser definition of network effects in which there is no direct interaction between nodes." Good distinction between scale vs network effects, not every company with scale has a network effect...
- PaulHoule 8y agoThe best analogy from "data is the new oil" is that a data breach or privacy event is like the Exxon Valdez.
- sdoering 8y agoWorking daily with clients who expect a lot from data. Especially when they quote that sentence, I tend to ask them what their car uses as fuel. Do they really fill their cars with raw oil or a product that was refined and turned into usable gasoline. Data might be the new oil (I doubt it). But you can't use it. You need to work with this raw material and turn it into an endproduct - depending on the use case.
- naveen99 8y agothe nice thing is that computer power is outpacing the growth in human population finally. An enterprise data startup can actually collect, store, and process some finite amount of data on all 7.5 billion people on the planet. Just find an interesting angle and be better at processing that data then your competitors.
- mindslight 8y agoAnd getting positive consent from each of us 7.5 billion people?
- ci5er 8y agoWhy would anyone bother doing that?
- abacadaba 8y agoan additional data point to use
- mindslight 8y agoThe comment I'm responding to contains unstated assumptions promoting the surveillance industry, at a time when many people are realizing they don't want to be part of it. We should be clear that this distinction exists, lest we keep giving a pass to misleading corporate statements in the spirit of Google's "don't be evil" that tricks naive developers into building more surveillance systems.
- ci5er 8y agoI was joking, but there is little in terms of life- /commerce- /click-flow (outside of maybe Europe, and maybe healthcare in the US) that prevents me from surveilling you any way I can. If I have a popular health app, I can surveil even more if you click "accept" to a ToS.
- maxxxxx 8y agoI would call this very scary. Soon a single entity can do full surveillance of the whole planet at reasonable cost. what could go wrong?
- johnsimer 8y agoYes, but we should also acknowledge the upside. Soon (or now) a single entity can benefit/innovate for the entire planet at a reasonable cost. What could go right?
- userbinator 8y agoThat's how a lot of dystopian sci-fi starts...
- maxxxxx 8y agoYou may as well call for a worldwide benevolent dictator. A lot of things could go right.
- wwweston 8y agoAs the saying goes, power makes people benevolent, absolutely power makes people absolutely benevolent, right?
- walterbell 8y ago> Generating synthetic data is another approach to catch up with incumbents housing large tracks of data. We know of a startup that produced synthetic data to train their systems in the enterprise automation space; as a result, a team with only a handful of engineers was able to bootstrap their minimum viable corpus. That team ultimately beat two massive incumbents relying on their existing data corpuses collected over decades at global scale, neither of which was well-suited for the problem at hand. Generated data is being reverse-engineered by machine learning? With both the generation and ML written by the same team?
- skybrian 8y agoGenerated data can be used to increase variety as a way to avoid overfitting. A simple example might be translating or rotating images. Not an expert, but I'm guessing it must be easier to do this than to improve how the machine learning generalizes from less data.
- hahajk 8y agoUsually when someone says “synthetic data” they mean CGI, not simply transformations of existing data. Using synthetic data is fraught (and presumptuous), as you are assuming you understand the problem domain 100% and are also extremely good at reproducing it. There’s a chance the model is using something specific to the CGI (and not the general reality) to produce its results. For winning a computer vision competition it’s probably ok but I’d be very careful about using synthetic data for systems I cared about.
- TeMPOraL 8y agoI thought "synthetic data" it's something that rarely shows in training image recognition, and is more like randomly generated user data (name, surname, etc.) or data generated from simulations of some processes?
- johnnycab 8y ago>I thought "synthetic data" it's something that rarely shows in training image recognition On the contrary, it is used to train models but it cannot adequately capture the long tail of weird events in the real world. Hence, it cannot be relied upon, as alluded to by the parent commenter. With reference to using data collected from a simulated environment vs real world ─ this subject was discussed at some length by Elon Musk and Andrej Karpathy, at the Tesla Autonomy Day event a few weeks ago.
- sandGorgon 8y agohow does this reconcile with stuff like this ? https://factordaily.com/indian-data-labellers-powering-the-global-ai-race/ https://factordaily.com/indian-data-labellers-powering-the-g... is this very contextual to business space (like enterprise startups) which A16z clearly mentions ?
- naveen99 8y agoThe good thing is there are infinite functions, not just infinite data. So even if you limit yourself to finite data, you can do interesting things.
- evrydayhustling 8y agoAmen to this. Enterprises are aware their data is valuable and increasingly have top of the line devops and pipeline tools to manage it. Pendulum is swinging towards data living with vertically integrated brands rather than horizontal services.
- jandrewrogers 8y agoMost of the leverage is in the efficiency at exploiting data, and not just in technical terms but also operational economics. This is far more important than the data itself for a fundamental reason that has not been broadly internalized by the industry at large, and which makes it difficult to build a data moat. Almost any proprietary data model based on proprietary data can be reconstructed with sufficient fidelity from unrelated external data sources to be competitive with that proprietary data model, at the cost of being somewhat more expensive to engineer all things being equal. I've demonstrated this many times in practice. As a corollary, a company with sufficiently efficient end-to-end data infrastructure could, in principle, commoditize all proprietary data model companies. This company does not exist, yet, and it would require some very specialized talent but every necessary ingredient exists. This is a realizable endgame due to the reality that virtually all data companies, for good and practical reasons, are strongly incentivized to build on infrastructures that are literally orders of magnitude less efficient than is possible in principle. A company that was purpose-built on exceptional end-to-end data infrastructure engineering could capture much of the revenue in these markets in a surprisingly short time by commoditizing the data model and arbitraging efficiency.
- huac 8y ago> A company that was purpose-built on exceptional end-to-end data infrastructure engineering could capture much of the revenue in these markets in a surprisingly short time by commoditizing the data model and arbitraging efficiency. do you have any examples?
- untilHellbanned 8y agoI was with you at the beginning. Can you ELI5?
- dswalter 8y agoLet's say you have a company that tries to make online checkout more efficient by predicting which payment the user will prefer, and research shows that if the user is first presented with a payment method they use, they are more likely to buy. If I subscribe to your API and get the outputs of your predictions, since I can see the kinds of inputs that you tell me are associated with preferring PayPal, I can approximate your PayPal-preference model. This works for basically any machine learning model-as-service. So if I invest heavily in approximating and then serving models more at lower cost than anyone else, that might be a viable businese.
- motohagiography 8y agoI'm really glad a16z put this issue to rest because it's the kind of problem that causes insane wheel spin at a certain kind of company. The "Data Acquisition Cost", and "Incremental data value," made me laugh because I had to solve that problem before. The most interesting related work on this is a paper covered on HN at some point (http://blog.mldb.ai/blog/posts/2016/01/ml-meets-economics/ http://blog.mldb.ai/blog/posts/2016/01/ml-meets-economics/) about economics' indifference curves and ML's ROC curves. The big mistake I've seen data companies make is they approach their market based on customer vertical, assuming that say a Bank will be like other banks, and a health care provider will be like other health care providers, and this is the fatal error. The bank with the same sensitivity to fp/fn/tp/tn rates will have more in common with the health care company with that same sensitivity than it will with other banks. Basic problem with any data product is the customer's ROC curve, or where they economically benefit from using your data service. Different customers have different sensitivities to false positive/negative, true postive/negative rates - and the customer categories themselves are defined by this sensitivity. e.g. What they have in common is not their vertical but their risk appetite. I have a blog post 3/4 written on this specific topic. That sensitivity is specifically an artifact of the customers growth stage as a company, which determines their risk appetite and the economics of the asymmetrical value that the effective ROC curve of your data product describes. (see above link). This is the fundamental problem for an ML/AI company, where they will go bankrupt trying to find their 2nd or 3rd insurance company customer because they think the value of their product is because their next customer is in the same vertical - not because they have the same sensitivity to fp/fn/tp/tn. Slight aside, it's so important that an investor like these can weigh on in these issues and other technical economic factors, because IMO when I listen to every technical person I know, the #1 cause of internal suffering at companies is caused by people trying to bullshit their investors, and blog posts like these just wipe away a big source of that temptation.