The MENTOR Framework: Six Things To Improve your Feedback Sessions

I’ve founded two startups and co-founded another, exited my last one, invested on a handful, and founded an accelerator. That added considerable experience to shape myself into a better mentor, and I’ve been paying it forward for some time now:

  • I have been asked personal feedback on thousands of businesses, some repeatedly, usually on closed sessions or calls. Each took from 30 minutes to extensive 4-hour face-to-face discussions.
  • I’ve written 10,000+ emails with fact-based feedback on pitches, ideas, business models, or simply recommending useful literature on specific topics the founders needed to improve
  • I’ve worked closely (weekly meetings mostly) with dozens of startups and helped them grow, some reaching 8-figures revenue on less than five years

Those would be just numbers if I weren’t a conscious perfectionist. As a consequence, I’m insanely cautious about two things: a) achieving the best outcome I can, as every perfectionist desires, and b) doing more with less to quickly reach the intended results, thus reducing my exposure as a shameful perfectionist. To be more effective and efficient, I spend considerable time trying to find my work patterns and molding those into frameworks – always increasing reuse and reducing waste.

So I recently blueprinted a mentoring framework I’ve been using for years, and it occurred to me: the best way to improve it is to share it.


The framework uses six key concepts to be kept in mind before, during, and after feedback sessions, and together they form the acronym MENTOR:

  • the Mechanics of the business
  • the Empowerment of the founders
  • the Network a mentor can bring
  • the Trust that good teamwork can create
  • the Objectiveness of key arguments
  • the Results and mutual benefits

Let’s dive deeper into each one of them, and hope you memorize their meanings by the end of this post.


The M stands for Mechanics, so we can understand the engine that moves the company. The goal is to discuss refinements, validations, and improvements in the business model that may lead to a more scalable business.

I usually start with a business model canvas in my head, but the plot thickens depending on the startup’s stage. Here are the M questions to answer over time:

  • Are we satisfied with the present problem/solution fit? Is there room to improve?
  • What is the nearest business model we can validate to get initial revenue?
  • How can we improve the scalability of the business model?
  • What are the key drivers or metrics that must be tracked?
  • What are the riskiest assumptions we can validate first, for every model hypothesis we have?
  • How does all that propagate to the team, plan and operations?

Mentors who think out of the box often are the ones who suggest innovative and profitable business models. So mind the M.


The eagle teaches the eaglets to fly by staying close and letting them dive by themselves. Keep this metaphor in mind, because the E stands for Empowerment.

This is critical for early-stage startups: we must help the team find the answers, and not impose the answers we already have. The founders must feel confident they’re improving and learning together with the mentor – so they’ll solve the hard things by themselves.

Here’s what you must consider to empower the founders:

  • Am I imposing directions? I should not.
  • Did they get to the answer by themselves? If not, how can I run around the issue and help them figure it out?
  • Are they right and I’m wrong? Maybe I’m wrong!
  • How can I reach a middle ground that I’m confident will work?
  • What task or references can I list to help the founders decide on this?
  • Remember motivation works well to reinforce good behavior: “Nice!” or “You got it!” should come out naturally when the founders solve stuff out.

Is the mentor pushing too hard? The founders can drive the discussion towards those points and help the mentor realize opinions are too biased.


This may seem obvious to most mentors, but not many exercise the N (Network) the way they should. The expression “Expand your network” needs no further explanation, so here are the key reminders:

  • What founders do I know who have been through the same problem?
  • What investors do I know who can help them out without being judgemental?
  • What contacts do I have in their Customer Segments who can try the product, or help validate it?
  • Can I help them sell this as it is, right now? Who should I pitch?
  • If I’m not able to help them on this, do I know anyone who can?

Founders should always remember the networking factor on a mentoring session. Ask for contact info right away, not later.


There’s no I in MENTOR. The T stands for Trust to constantly remind us that mentor and founders must form a cohesive and dependable team, each one resembling the rock climber fixing the rope right above you.

There’s also an intrinsic ethics present in every feedback session: confidential operations details will be verbalized by the founders, and personal stories will be disclosed by the mentor. Hence, some conducts must be observed:

  • Ethics is massively important on mentoring
  • The mentor should not impose his experience nor act as an omniscient deity (mentors learn as well!)
  • The mentor should not impose respect, and should avoid comparisons between how mentor and founders performed
  • Mentors and founders must compromise to deadlines and pending tasks
  • Mentors and founders must be able to simply have fun, even while discussing the heaviest topics

In my experience, when founders and mentors build a bond that allows for straight-up teamwork and trust, feedback sessions and the overall relationship play a key role in the evolution of the business.


Remind everyone in the room to be Objective: when someone says “I think” or “I like it better” in a feedback session, a fire alarm should go off. That speech balloon should be instantly replaced with “Warning – I’m stating an opinion that may in NO way relate to the inexorable truth. That said, [ insert the original remark ]”.

That doesn’t mean people can’t issue opinions, but instead prevents subjective decisions to influence the startup in a harmful way. Here are the tips:

  • Look up numbers and facts that corroborate your opinions
  • Define the metrics that will validate which path to follow
  • Propose assumptions that invalidate the present conclusions (they are mostly intuitive ones, and those should be minimized whenever possible)
  • UX is only validated by users, not anyone’s (much less a mentor’s) design skills
  • Shut up and listen. Rethink what you’re about to say, and rephrase it in a non-subjective way

So be objective, and trigger the fire alarm whenever an opinion may clutter your perception of reality and what must be done.


It all comes down to the R: Results. They provide greener pastures and better crops; they must be found, measured, and compared. As a consequence of great results, a mentor may become an advisor, investor, board member or earn sweat equity. I dare to suggest a recipe for R:

  1. Finish all mentoring sessions with “What have we learned today?”
  2. Then ask “What are the key issues that need further addressing?”
  3. Then ask “Who will take care of what?”
  4. Take notes of “Who – What – When” to make sure things get done
  5. Follow-up on those notes at the start of the next session
  6. Mark down the strongest results so you can review them later

Make sure results are tracked; see that the work and effort spent on mentoring sessions will not be lost into oblivion. Keep them either on a shared Evernote, spreadsheet, or even Slack channel, and you’ll be amazed of how much can be done with mentoring if you follow this framework.

Exercise all six concepts equally

First of all, remember that the best mentoring is long-term. Quick feedback sessions can make great use of this framework, but its real benefits are only tangible over frequent sessions.

I don’t believe I’m the best mentor many people have had. I’m more demanding than I should, way too blunt sometimes, but that bug is my best feature to help get things done. The specific issues I need work (and constantly try to exercise that) are E and T, and my highlights are M, O, and R. So I do recommend you try to know what your strengths are, and improve your weak spots.

It’s hard to go through all six concepts in MENTOR on a single mentoring session, but always keep them in mind. As your experience progresses (or as a founder, as your mentoring sessions flow) you start getting the whole picture and the modus operandi unconsciously falls into place. If you’re a bad mentor, you will get better by exercising this framework. If you’re already a good mentor, you will gradually notice your effectiveness increases as the time passes; and if you’re a founder, you may use it to make sure your mentoring sessions – good or bad – tilt towards what you need by getting the mentor into the right path.

Click the image below to download a quick reference card on the MENTOR framework. It is yours to use – let me know how it goes.

Acceleration is a process, not a business

Accelerators have once been called a breakthrough in venture capital… But not anymore. So what changed?

The acceleration market – the one where an accelerator tries to operate as a business on itself – has very little room for small players, since that market can’t comprise a fragmented offer. You can sell selfie-sticks and make a small fortune with the right distribution channels and pricing strategies, but top accelerators – usually attracting startups from around the world – have a huge advantage in making money when compared to smaller ones.

In a way, that’s counter-intuitive: small accelerators can provide more attention and more funding for all the great local startups, and intensify their domain over a particular geography. However, a massive effort from small players can’t compete with a handful of accelerators inside the huge flux capacitor that is the Bay area. So even when blueprinted to be F1 racing cars, small accelerators will often become rusted oldsmobiles kicking into gear for a living.

Scaling up is key

Think YCombinator, Techstars and 500startups. Can you list 2 other accelerators that really have similar global reach? I can’t.

By bootstrapping and running a small accelerator of my own, I learned that it’s hard to keep up the wins over time. As time progresses, competition increases and more accelerators become attractive for startups. Even if they’re not that good, the proliferation itself creates friction for applicants… so either you’re number one, or you’re destined to become second choice to every great founder considering an accelerator (and believe me, not many great founders do).

As a self-funded operation, we ran 6-month sprints over 4 years trying different business models – and those 4 years were a premise considering our investment budget. Our weaknesses were mainly our riskiest assumptions, not necessarily execution flaws (some may disagree):

  • We didn’t want to raise money to fund the operation itself until we were sure that there was at least a 2x return on the business. We never really validated that premise, so we decided not to burn other people’s money.
  • We contacted trusted investors, but they prefered to direct the money to create / increase their own funds. All of them asked us to feed their dealflow with the startups we chose (I would ask the same)
  • Our close friends who managed funds kept saying “you should raise your own fund” – but soon enough they were weeping on our shoulders asking “why oh why can’t I find decent investments?” or “when did I come up with the ridiculous idea of raising this fund?” (dealing with LPs is hard in an emerging economy, but that’s a whole post by itself)

We publicly declared in late 2012 we wouldn’t have batches anymore, precisely when our country was seeing dozens of accelerators flood the market. Obviously we quickly became the ones rowing alone against the stream, but from our perspective, we were just the tired pioneers collecting arrows in our backs.

Nonetheless, we love the portfolio we built and still have buy options on other startups; that’s way better than our original worst-case scenario. We’re still #1 on search results, get ~10 pitches everyday via our website, did our part to improve the national ecosystem, directly helped 2,000+ entrepreneurs, and we get all the recognition from that work. But it doesn’t pay off, and we’re looking for a different way of multiplying capital with what we learned.

Burning fuel in the wrong places

Accelerators suck as a lifestyle business. YC, Techstars and 500startups are proof that either you scale up and aim for global reach, or your destiny will be no different from a single angel investor: only a few good portfolio companies, a lot of effort, but very small odds for success. And even worse: as an accelerator, you’ll feel forced to burn more fuel… but it will keep burning up in all the wrong places. And that waste is unacceptable for a Lean VC.

To accelerate is to do more faster (thanks Brad and Dave for coining the expression). Depending on one’s will, an acceleration process can be plugged into any fund, angel, or even an accelerator… as long as the model composed of small bucks + small equity + a lot of sweat isn’t expected to break-even by itself.

Most accelerators face this dilemma somewhere in time, and often try to scale up their services by:

  • hiring designers, lawyers, developers, community managers, even PR agencies. That’s their first mistake: it doesn’t pay to have a full-working staff or simulate a startup to grow. I give 1-2 years until their investors start to question the efficienty of their internal ops.
  • They throw events, start education courses and infoproducts; hold exclusive VIP dinners and meetings, ask for sponsor funding and sometimes even pay to sponsor events themselves. That’s an intuitive search for revenue that will also burn out in around 1-2 years.
  • They try to go corporate, pitching enterprises to increase their relationship with startups and use an internal accelerator as means to openly innovate. Some will succeed, but 5x or 10x returns are improbable – so a coma may descend upon a corporate accelerator in 1-2 years.

Like Bill Maher, I don’t know this for a fact – I just know it’s true. This is why I believe an accelerator can resist from 3 to 6 years anywhere outside Silicon Valley, unless it reaches global scale somehow. There is no fixed acceleration model you can borrow to become successful… I myself assumed that from scratch back in 2008, but decided it was worth the experimentation.

Those are the reasons why I state my point: in spite of what most accelerators originally think, acceleration is a process, not a business model.

TechstarS is diversifying and it makes sense. YC now has Sam Altman on board and he’s adding his own twist of running things. 500startups is the one pivoting towards the sweet spot to become a strong contender, mixing a fund + accelerator + party thrower format (update: though the party has prooven excessive lately). They still need fix on cashflow issues (raising x investing) to ease the follow-on process and peak their efficiency on equity and returns. I also believe their internal structure is ever-changing, but they seem ok with that. Even so, I’ve been scratching my head over something… to scale up acceleration the same way AngelList did with fundraising.

Food for thought.

Here’s why you don’t get concrete feedback from VCs

Our digital channels receive an inbound volume of 10 startup pitches daily. That may be a small figure for most big funds, but add those to the startup we dig ourselves, and they trigger a considerable effort on our side: replying to those founders with a proper answer, and inserting a piece of advice when we feel like it.

I can remember 5 or 6 very successful startups to which I gave extensive feedback in the past. Because of that feedback, they approached me on every roadshow, even when the round was larger than I could afford. I lost count of how many board seats I’ve been offered on hockey-stick revenue startups simply because of previous feedback I have given (even those sent over email).

I do believe in careful and thoughtful feedback, even when you’re not interested in investing. In other words, I believe feedback is part of a VC’s job description. When we do that, 50% to 60% of the founders come back with something like this:

First of all, thanks for getting back to us. The other funds we contacted haven’t even replied with an automated message after weeks.

In regard to your suggestions, (…)

We’ll get in touch when it’s all done!

Startup founders are a key customer for a VC fund. Without them, we have nowhere to deploy the capital we manage; without their work, the process of transforming venture capital into profit is… nonexistent. If they value feedback that much, we should consider it carefully.

According to entrepreneurs, most VCs are pedant and self-centered individuals. What do the founders need to understand about the lack of feedback, and how can VCs improve?

VCs are a team, not a person

Even small VC funds have a 3+ people team managing the dealflow and the relationship with entrepreneurs. This includes interns to process superficial contact, analysts to handle the dealflow, and senior analysts or entrepreneurs in residence (EIR) to further analyze potential investments.

When a founder starts talking to a fund, a single point of contact may become two to three different people over time – and even before the VC has sufficient information to provide feedback. So it is expected that early-stage startups, the most typical dealflow candidate, think they’re not getting enough feedback from Pete or Kimberly or whoever. After all, that person is just the front-end of a complex process, and those who manage that process aren’t focused on all startup stages – only those specifically applicable to their investment thesis. The remaining founders simply don’t qualify for real feedback; the fund believes a polite automated message will suffice.

It’s not what you know, it’s who you know

Every startup submitted to a VC fund enters a pipe. The same way a startup processes its customer funnel, VCs need to process their startup dealflow. Each startup is prioritized within this pipe, and receives more or less attention according to its entry channel:

  1. startups that have been personally recommended by a Limited Partner, General Partner, or Associate Partner
  2. the ones recommended by trusted sources, who in turn are close to LPs or GPs
  3. the ones harvested by a Senior Fund Analyst, who met the founders and has a favorable opinion
  4. the ones recommended by a portfolio founder (from one of the startups already invested by the fund)
  5. those coming from any source at all, but promptly showing extremely consistent metrics
  6. those coming from a partnership with a startup competition, accelerator, or earlier-stage funds
  7. those cold-called by an intern or analyst (some funds have dozens of people cold-calling startups)
  8. those submitted via the fund website or any other digital inbound channel

As you can see, the quality and promptness of the VC feedback is proportional to the entry channel. The first channels above have a higher score, so startups coming from them will receive higher attention when evaluated, prompted or getting final feedback.

The further you go into the investment funnel, the more different people you’ll get to meet. To make that worse, different people will have fragmented opinions and will have to converge to achieve a cohesive feedback. With all those levels of attention, it’s hard to get concrete feedback from a VC on the earlier funnel stages.

Please say yes! Please say yes!

VCs (almost) never say “no”. That happens because most VCs believe they’re able to determine whether a startup will or not be successful in the future – or at least, they think they’re skilled in the art of evaluating the timing and attractiveness of good businesses. If they pass on a company, they may be seen as stupid or visionary… It all depends on how big that startup becomes in the future.

The further a startup gets into the dealflow funnel, easier it is to get a “no, thanks” answer. So at some point in the future, VCs will say no… But usually, they postpone the next follow-up indefinitely – which is the equivalent of a “no, thanks” answer.

On a funny note, here’s what investor Josh Kopelman got recently after passing on a startup:

Scaling feedback

VCs are not in the business of giving feedback – they’re in the business of multiplying money (mostly other people’s) by exiting an investment. Giving feedback to every founder that approaches a fund costs money and time to execute a process they haven’t initially budgeted for. This is why concrete feedback to startups from the top of the dealflow funnel is so underrated and overlooked in the VC world, considering the massive number of entrepreneurs entering the pipes of several funds at the same time.

On the other hand, proper feedback can drastically change a startup’s future, and also how that particular VC is perceived by the founders. Good feedback shows them how to improve: it helps startups to do more faster, and understanding how attractive they are (or not) to a particular VC’s investment thesis.

The only way VCs could scale up their feedback throughput would be radically changing the way they approach their interests, manage the dealflow, and how they delegate internal decisions. For VCs, feedback is waste – but for a Lean VC, feedback is an asset to differentiate…

The secret lies in the balance. We’re still trying to find the answer, and keep giving feedback to all applications we receive.

Can Venture Capital be Lean?

The Lean Startup creed is now deeply intertwined in every conversation about building innovative and scalable businesses. Lean is not the only way to build innovation, but the concept has somehow become an obligatory topic when those businesses are built and refined.

It’s been years since I tried to tilt this discussion towards Venture Capital – mainly, how we should (if ever) consider a lean approach to the VC business. When I first compiled my initial thoughts and drafted this post back in 2012, I sent it to 5 VC friends I respect. One replied with “this is a wasp nest you don’t mess with”, and the other sent some feedback: “The overall ideas are nice and you should try this – but don’t count on me…” Thanks for boosting the morale, buddy.

After a long pause with this whole thing dwelling in my head, here’s that draft my VC friends couldn’t care less about.

The VC business

This is a factual and brief description of how most VC funds work:

  1. The General Partners spend a couple of years raising the fund
  2. The Limited Partners’ capital is partially used to pay for the fund team’s salaries, costs, and expenses
  3. The fund team captures, analyzes, and processes dealflow to pick specific startups as potential investments
  4. Analysts are assigned to dig deeper into a couple of startups from the dealflow pipe
  5. A negotiation process takes place and the fund managers decide whether to invest or not
  6. The money is deployed and the startup is evaluated over its investment period
  7. The startup founders use the money as they see fit, under some control of the fund and/or formal Board
  8. The startup will either die, stabilize or grow. Success is whenever the fund sells its shares, yield returns to the LPs and takes a cut of the profit.

If Lean is about reducing waste, that means solving inefficiencies, constantly measuring performance, reassessing the operation and optimizing internal processes. So here’s an experiment for you. Pick any VC you’re close with, and analyze how they perform those 8 macro steps. You will find waste in most of them, and here are some examples:

  • Depending on how GPs approach LPs (and who they approach) they may take much longer than planned to raise the fund. They often need to close the fund without reaching their intended watermark.
  • Salaries, costs, and expenses are poorly budgeted – either too low or high. They may hire young, cheap and inexperienced people, or experienced but expensive analysts.
  • They have no concrete way of measuring performance when managing the dealflow. VC is an art, they say.
  • Analysts may miss incredibly good opportunities, or invest too much time on false positives.
  • Inexperienced VCs take waaaay longer to negotiate termsheets, and sometimes forget investing is a fair game.
  • There is no pragmatic approach on how to quantify startup maturity or measure its evolution.
  • Frequently, the money is deployed in large chunks on immature startups – and VCs mostly pay for entrepreneurs to learn and make mistakes.
  • VCs are used to burning money. They know only a couple investments will really work – the others will die or zombiefy, and that’s just the VC business.

So here I am, stating my case: the average VC fund is a pile of waste. And though VCs criticize entrepreneurs wasting time and money, the VC fund itself doesn’t minimize waste on its own operation.

On both cases, it’s other people’s money to spend. But if you wear angel investor shoes (I do), you start thinking about how you’d use that money in a smarter way (and the most fiduciary way of all): as if it was your own.

The Lean approach

I’ll spare you a long explanation on the Lean principles, because Wikipedia is there for your amusement. In respect to Lean and venture capital, there’s a couple of examples worth mentioning.

Dave Mcclure may be somewhat explicit or uncomfortable at times, but he has approached this before:

What we’re doing is a quantitative experiment at scale to do quick cycle-time product development, and then increase capital for things that are working. (…) We can do lots of fast experiments that are cheap, they may fail at high rates, but we’ll be able to determine success quickly. (…) This is sort of putting Internet entrepreneurship into the Henry Ford era: assembly-line startup kind of efficiency. We’re able to do this at a very, very large scale.

Even so, I believe there’s more to that than “lots of little bets, doubling-down on a few” and the product / market / revenue investment thesis. In fact, Dave does run a noisy itinerant party and did create a good heuristic to reducing waste – but minimizing waste on venture capital is an important discussion.

AngelList, founded by Naval Ravikant, is my second example. Naval and his team implemented and distributed a beautiful business model for venture capital. AngelList can be seen as a much leaner VC than 500startups: they operate venture capital as a platform, so that both small and big VCs can rely on them to enjoy a less bumpy ride, worry less about handling picking a portfolio and more about important things such as the differentiation of their investment strategies.

AngelList’s tools and processes dramatically boost the dynamics of the VC market, easing the money flow for entrepreneurs and investors of various sizes. That’s an incredible achievement, and the world is a better place because of them… But not all investors can or will use the software approach, and may still want to do something about the VC waste.

Lean VC

Bringing those two examples to light (and believe me or not, there are not many other real-world examples) is my way of convincing you that there’s more to Lean VC than has been historically discussed.

My proposition is that Lean Venture Capital should be defined as a systematic and pragmatic approach to minimize waste on the key processes that really impact the investor life. Here’s a quick list:

  1. Getting dealflow
  2. Processing dealflow
  3. Negotiating and closing deals
  4. Assessing the investment portfolio
  5. Making liquidity-related decisions

If those processes are addressed in careful detail, one can model a different way of operating an investment portfolio – be it a fund, corporate, or personal one – and maintain one’s personality without having to mimic 500startups or AngelList. I’m not talking incremental changes such as cutting costs or reducing management fees (a lot of funds already do that to become less fat, but not lean). I mean sistematically trying to scale up a venture capital operation.

If you like what you read, don’t miss the next post: how to better treat dealflow and give proper feedback to entrepreneurs.

Fifty shades of bubble

Talks about a startup bubble have been all over the press for the past decade. Josh Kopelman has an interesting take on that: “Even a broken clock is right two times a day. By proclaiming a bubble every year, everyone can say they called it.” Even though Fred Wilson said ‘burn baby burn’, the public market has done a good job so far correcting whatever inflated bets by private investors.

Here’s my take to end whatever bubbly assumptions you might have with a simple explanation… Let me know what you think in the end.

Understanding shades

For the sake of simplicity, let’s take a look at the 8 possible outcomes of a single venture capital deal using two criteria:

  • Y axis: the amount of invested capital, from low (close to zero) to high (dozens or hundreds of millions)
  • X axis: the amount of returned capital, from a very big loss (a terribly capital-consuming deal) to a big win (10X or more compared to the deployed capital)

See, the graph is wider than tall to keep the proportions between deployed capital and returned capital. A high return should be considerably larger than the invested amount.

My biggest wins as a VC will be the deals with highest proportional returns, located in the green area. I can be forgiven for any other blue or orange section (even when I invested small amounts and the losses were huge: either I own a very small stake or my write-off is long past). Only the red area indicates I’ve been a bad, bad VC: I picked a deal that consumed a lot of capital (and maybe doubled-down to cover losses) but the company is a black hole of money, due to bad unsupervised execution or my bad VC handling.

This is why VCs manage portfolios, and not single deals : to make sure that risk of failure is distributed over a string of investments. But when your hear talks about a bubble popping up everywhere, this gradient graph can be a good framework for understanding the market.


 Imagine we gather all investments in the market today, and throw them into a deep bag. That would hold several thousand invested startups from all the VC funds at the present moment — in other words, today’s VC market as a whole.

All the best performing funds would be located inside the green area, some slightly dragged to the blue section because of the few poorest performing startups in their portfolio. The orange section is sometimes unwanted, but hey, it happens a lot.

But bubble? Really? A bubble is when the market as a whole — or a very high percentage of it — is systematically bleeding VC money, and that would be if a considerable portion of the VC deals were located inside the red area. That was the public Internet market in 2001, but it is not what we’re going through presently.

Yes, the consecutive warnings from every respected VC — including this article with Marc’s thoughts — may be the initial signs of a systematic increase of dark-orange deals, i.e., signs that the resulting performance of the average portfolio more to the left of the Y-axis than it should. But we’d need a very big portion of the market performing badly to cry wolf…

Now remember this…

We’re not in a bubble. Yet.

What we’re going through is that some widely known deals have been consistently showing signs of entering the dark-orange section: peak-high valuations and excessive burn rates, and may be tilting the market towards the red blooded hell. When those specific deals are examined individually, yes, they are somewhat scary – but we’re far from a pandemic disease.

Even though, beware: all hope abandon, ye VC who enter the darker shades of bubble.

Page 2 of 212