Hacker News new | comments | show | ask | jobs | submit login

This detection process doesn't detect toxic customers, it detects corporate customers at large companies.

It's fine for a small startup to cater to small startups, but the big companies have big budgets, and eventually, you'll be making 80% of your money off of them, so learning how to deal with them can be helpful.

1. Big companies often have purchasing departments actually do the purchase. They are trained to expect discounts and the people in the purchasing department know a lot more about asking for discounts than they know about software, because that is their specialized role in the organization. If you politely tell them that you have one price for everyone, they'll still purchase, because the purchasing department ususally doesn't have the power to stop the purchase.

2. Those 80-question checklists usually come out of the following, typical corporate process:

* A team of people identifies a need for software

* The team meets to agree on everything they need

* The junior person on the team is tasked with evaluating 12 possible products to see which one is best

* That person makes up a spreadsheet and sends it to each of the vendors hoping that they will do his homework for him

* The vendors who have decent presales support or sales teams fill out the spreadsheets by marking everything as "Yes" or "Yes with a footnote" and get the deal.

This also explains the "multiple questions that can be answered from a website" -- it's a sign of a person who has been put in charge of evaluating multiple products, not a sign of a toxic customer.

3. Multiple contacts through multiple channels are usually the sign of multiple interested parties at the client site. You can't sell to big companies without touching multiple people. One of a salesperson's most important jobs is helping the customer themselves get organized and make a purchase. A good salesperson helps the person who wants your software navigate their own corporate purchasing politics.

Summary: while it's fine to turn away truly toxic customers, and you are welcome to decide that you'd rather sell to the starving startup founders on Y-combinator who would rather spend 2 hours scouring your website than deal with a salesperson, the corporate customers turn out to be remarkably price-insensitive, once they make a purchase they will keep paying you maintenance for years long after the product is not even in use, and they're just as likely to leave you alone as the small guys, but they do have "multiple stakeholders" and if you want to sell to them you need a process that matches their reality.




Star that bit about helping customers get organized (and relatedly, providing ammo for your internal champion, like PPT which he can take right into the meeting where he asks for budget to purchase you): it is some of the most important inside sales advice you'll ever hear.


Although I generally agree with your comment, I disagree that the process only detects large corporate customers. In fact, I've seen plenty of small customers who for whatever reason act like a large corporate customer, but instead of bringing 80% of revenue end up bringing nothing more than support overhead.

These are also the guys who either slander you online or at least threaten to do so. These are also the entitled customers who tend to think you have infinite resources to throw at them. In my experience, large corporate customers were more understanding to having to draw the line somewhere than small irate ones.


My impression of the original article is that it suffers from overrationalization. The author knows, from empirical observation, that toxic customers exist. And he feels -- possibly accurately, but possibly due to the awesome power of hindsight -- that these customers should be detectable in advance.

And that may be true. But if so, it's not a scientific process. The warning signs do factor into it, but in the end it's a matter of gut feeling. (Perhaps because it's not necessarily the case that a given customer is universally toxic, it's that they don't fit you, your company, or your product.)

But it's always embarrassing for an engineer to write a blog post like "how to detect toxic customers by going back and forth in conversation, trying various rhetorical strategies that you make up on the fly, and occasionally closing your eyes and trying to use the Force". So instead the author tried to invent a nice, rational checklist. It has the virtue of looking scientific. The only downside is that it may not actually work very well. It's not what the author is actually using, either. You can't do sales, especially corporate sales, with an algorithm unless you're willing to leave money on the table, maybe a lot of money.


Gut feeling is the only real way of separating things out.

Asking huge numbers of questions can be a sign of a customer support nightmare or a person/company/entity doing their homework. Following up can be a sign of high expectations or them being keen to make a decision asap. Asking for a discount is sometimes a buying signal. Conversely, whilst rudeness isn't a good thing, some of the customers that will be a really bad fit are the ones who are really making the effort to be polite and charming.

The one almost universal red flag would be constant inappropriate reference to price. Asking for discounts is to be expected. Asking if there are cheaper, slimmed-down alternatives, or freebies for volume purchases is reasonable. Constantly referencing the price whilst raising unrelated issues or requests is a sign that something's amiss. Whenever you hear something like "I'm surprised to hear you haven't added xyz considering you're charging £xxx" or "so before I starting paying £xxx, would you...?" it's a pretty good sign that they're either rationalisig a decision to not spend the money with you or have an inflated sense of how far up your priority queue their spending pushes you. Then again, you can suspend that rule for businesses that are likely to pay enough to push themselves up your priority queue too..


I think it's a little bit disengenious to post this, because http://www.joelonsoftware.com/articles/CamelsandRubberDuckie... covers this just as well, but hints in the opposite direction, viz:

"There's no software priced between $1000 and $75,000. I'll tell you why. The minute you charge more than $1000 you need to get serious corporate signoffs. You need a line item in their budget. You need purchasing managers and CEO approval and competitive bids and paperwork. So you need to send a salesperson out to the customer to do PowerPoint, with his airfare, golf course memberships, and $19.95 porn movies at the Ritz Carlton. And with all this, the cost of making one successful sale is going to average about $50,000. If you're sending salespeople out to customers and charging less than $75,000, you're losing money."

In this case, the software is lower than $1000 - so he needs to detect and avoid time-guzzling corporate buyers, even if they are likely to buy something.

Or, to setup corporate pricing. But you can't recommend the 80 question answer approach at this pricepoint.


This is a valid point. However, our product (small business invoicing software) and price point ($300) mean that we don't sell to enterprise customers. We don't make a product for this size business, nor do we have the price point to support the sales effort required.

We receive maybe one inquiry a month from a large customer and they actually tend to be some of the most reasonable prospects we come across. Very rarely do they show disrespect, ask for discounts, contact us over and over again, etc...

Though it's not critical to the discussion, the person I mentioned in the post was not a corporate customer. They were a very small (less than 5 employee) firm who happened to be exhibiting these toxic/large company traits.


Spolsky's comment is way overrated. He's only addressing the 80 questions bit. In that case you just need to add enterprise pricing to cover that sales cost. I'm guessing for a $9990 license (keeping it within a single manager's pre-authorized budget) you will feel quite happy filling in the little Excel cells for a couple hours.


You've really only covered two points - why corporate customers won't RTFW and why they ask for discounts. If you are dealing with a corporation you might get contacted multiple times through multiple channels, but probably not from the same person in a short period of time.

Disrespectful and abrupt communications are still a warning sign for larger companies as are unrealistic expectations (at least for the example given).

The article specifically mentions that "Any one of these warning signs is not a big deal, but stack 2 or 3 on top of each other and (depending on their severity) you have yourself a red flag." A corporation might only set off 1 or 2, and if you keep in mind the points you made I don't think it will generate too many false positives.


The tactless tosser in purchasing that quite possibly has been told to get some pricing options by his management by the end of the day really shouldn't cloud your judgement of the entire institution.

Even in the example blog, the problem person was eventually bypassed and the sale went ahead.


That was the key event for me.

If one finds oneself in this situation is there any general advice for trying to get to talk to someone else if you contacted by a jerk from what might otherwise be a good prospect?


Having been on both side of the enterprise sales process I think this is a remarkably insightful post. A lot of this goes to explain why enterprise software often needs to be quite expensive - you have to pay for a lot of time and effort from sales and technical pre-sales people hand-holding potential customers and helping them through their software purchasing process.

[Edit: I didn't notice who wrote that comment before replying!]


Actually, there might be a market for some kind of tool to help enterprises gather software requirements, create shortlist of candidate vendors and allow the vendors to reply online.


Microsoft Excel owns that niche.

If you're going to aim for the same niche, it'll likely involve a web service for creating RFQs and also for building up a database of both the requests as well as building up the feature loci of the various queries; you'll need to differentiate your service from the aforementioned Excel and from Google Docs and various of the testing services. Acquiring and then maintaining sufficient feature data (whether via web services or scraping) is probably going to be the tough part.

Look around for folks already dealing in electronic purchase, e-procurment, and particularly at SAP.


Well, software for generalized decision making already exists so I can certainly see a market for more specialized versions.


> This detection process doesn't detect toxic customers, it detects corporate customers at large companies.

For a small business with fixed pricing, a corporate customer at a large company may well be a toxic customer. All the points you made hold true, but they also all increase the cost of the sale for the small business. If the product pricing does not cover the "high-maintenance-ness" of a large corporate which is expecting a corporate service, then why isn't this customer toxic from the point of view of the small business selling the product?

OTOH, there are often purchasers at large companies who understand that they're buying a commodity product, can approve the purchase (based on the much lower price) and don't have such high expectations.

It's all about being aware of expectations and declining the sale when they don't match, which is exactly what Rob identified here.


Great comment. I've noticed that the procurement guys typically need to show their bosses they've gotten the company a discount. So now I just price the "standard" price column artificially high, show a 20 or 30% discount to get to the real price, and everyone is happy.


I would add - as a buyer (and possibly someone who's hell to a sales guy) - I usually ask to be put in touch with a sales engineer, and if he can't handle me, I very nicely explain I'd like to talk to the developers to get a feel for their product - because the devil is often in the details.

EDIT: Also worth noting - there is value in recognizing a customer you can't handle (whether or not you should be able to isn't relevant) - and then following up later to find out how things worked out for them so you can find out if you screwed up an opportunity.


> It's fine for a small startup to cater to small startups, but the big companies have big budgets, and eventually, you'll be making 80% of your money off of them, so learning how to deal with them can be helpful.

Perhaps, if you are running a B2B operation.

If you are a small business running a B2C operation, however, then the original article seems pretty much on the money to me.


Similar features-checklist spreadsheets can arrive from industry analysts seeking your assistance in building their whitepaper products and various other of their comparison products, too.

Seeing the other side of those analyst annual reviews can be quite the revelation.

You can infer interesting details from what check-offs they are including, what they're omitting, from the nomenclature used, and from the check-off combinations used for specific features and capabilities.


> Big companies often have purchasing departments actually do the purchase. They are trained to expect discounts and the people in the purchasing department know a lot more about asking for discounts than they know about software, because that is their specialized role in the organization.

My wife actually works as a senior purchaser for quite a big (non-US) oil company, and it was funny when she had to negotiate the fees with the web-design company that had been asked to implement an internal application. She didn't know anything about the technical side of it, as you well pointed out (even though I helped her out with is), but the sad part was that even her company's online department (the ones that should have known better) were quite on a different planet.


Thanks for this comment Joel. This is why I love hacker news.

I was reading the post nodding my head in agreement and then I come here and the first post provides a perfect counter point to the post and explains why the blog post may not in fact be correct.


Well, this is the first piece of actual experience I've gotten from the web in a while, and I didn't even have to screw anything up! I have a feeling it will come in handy in the future, thanks Joel.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: