What I Learned at Salesforce.com
medium.com
medium.com
1) Implementation absolutely requires assistance from an integration partner.
2) Developing on top of it is an exercise in patience and futility. Their dev platform & APIs are inconsistent and half-baked.
3) You can't just license a la carte. No matter what you think you need, you end up being quoted and convinced you need the kitchen sink instead.
4) It is prohibitively expensive compared to pure focus CRMs like SugarCRM, and doesn't have nearly the engineering polish (I can't believe I'm saying this) of things like Microsoft Dynamics.
5) If you aren't a multi-thousand user customer, Salesforce's sale folks don't take you seriously and it's hard to feel like you mean anything to them. I have never received such poor support, and this includes other companies famous for poor support (Google, Amazon, Microsoft, Oracle).
It's like they're still pretending to be a lean, mean, sexy startup when they're firmly in enterprise territory, yet they refuse to behave in ways appealing to enterprise customers. Moreover, they have spawned a whole slew of SAAS providers whose sole purpose is to interface other things into Salesforce (like 4Bridge from 4Thought, which connects Eloqua (lead generation on the web) to Salesforce. Sure, Salesforce has some huge customers -- at Dreamforce a few years ago, Cisco claimed to have something like 30,000 seats and Dell had 20,000 -- but that's not the entirety of the market. To me, Salesforce is the embodiment of the perception that many SAAS companies exist solely to support other SAAS companies, and that they bring little to no value to anyone.
That's because it's just as expensive for them to close a deal with and support your small-to-medium-sized org, but the revenue they make from you is an order of magnitude less than they'd earn from closing a Fortune 500.
This is the reason for the valley in between low-touch SaaS for small businesses (pick a monthly rate plan, pay with a credit card) and high-touch enterprise SaaS (face-to-face pitch and negotiations, sign service contracts, cut a purchase order)
In these kinds of situations, it often works out a lot better to deal with a middleman, like Blue Wolf (in the case of Salesforce) or Cloud Sherpas/Appirio in the case of Google Apps, etc).
Make something people want.
If they require assistance, have a half-baked API, sell you the kitchen sink, are prohibitively expensive and they are printing money selling sales software they are the definition of making something people want.
> it's platform is just not viable for many potential customers
It might not be what you want, but let's not lose the forest in the trees. To build a billion dollar company you have to sell a product that is not viable for many potential customers. Not all phones are Apple and not all sales software is Salesforce.
The size of a company doesn't allow or disallow it to be lean and mean, just as both a 7' and 5' tall athlete can be lean and mean on the court.
Life doesn't work in simple axioms like this. You're clearly not familiar with "enterprise software" if you think value = sales.
I'd wager that the vast majority of SalesForce users despise it. That means it's delivering less value than something that's efficient and gets out of their way. "SalesForce" is now listed as a skill on resumes, which means it is the opposite of usable!
So companies buy it, and you could argue that companies "want" it and get "value" out of it, but sometimes it's still not that rational. SalesForce has great salespeople, and it's a huge company. Sometimes just being the biggest will get you deals because it guarantees you'll be around and can offer enterprise-level support.
In fact, every time I talk to someone about salesforce and go "what, about that interface, huh? How bad is Salesforce!". They look at me like I'm a little loopy.
It's basically a lot better than the thing they used to have, which was nothing. And they can use it on their phone. And it hooks up to Outlook. And it does X, Y and Z other useful things that make their life easier.
The difference is that you and I can look at Salesforce and realize it's a mess of legacy and outdated interface ideas, that SOQL and Apex really fucking suck, so, so bad. But for users it's better than the Excel spreadsheet that kept getting out of synch, or the painfully clunky google sheet or whatever.
I mean, every time I look at salesforce it's a disgrace. Something as simple as basic filtering by related object values on Views is impossible. Or that the Notes hide the note behind a popup instead of showing the note in table view. Or that you can't change the layout of the entity homepages. Or that bizarre news feed interface.
And yet they've built a thing which tries to automatically get the twitter or linked in profile. It just screams of bad, bad, here-be-dragons core code which no-one dares touch.
> "If you aren't a multi-thousand user customer, Salesforce's sale folks don't take you seriously"
Thanks for the breakdon. I strongly agree with both above, which I think leaves plenty of opportunity for those purely focused on SMBs where companies like Hubspot have stated they will stay. Ever heard of a 'Salesforce Developer'? I put that in the same category as 'Websphere developer', 'Dynamics CRM developer' or other 'Huge Headache Engineer'. When the market needs a specialist for your tool (ie one who is familiar with the half-bakery) then how is one NOT leaving opportunity behind?
Thanks
Yes, and the good ones charge obscene amounts of money. A good niche if it interests you.
My enterprise domain knowledge is far more valuable than my technical skills, and Salesforce is just a more fluid vehicle for delivering this value compared to traditional on-premise solutions from Oracle, SAP, and IBM.
Sure, the upstart CRM vendors have better UIs, but they aren't the real competition.
Salesforce is executing a classic innovators dilemma strategy of acquiring these upstarts (desk.com) and allowing them to operate independently in the SMB space as an on-boarding ramp to the enterprise-scale solution.
Great idea (i guess) but Desk.com sucks pretty badly so we ended up with worse responses to customer issues...
How many other live applications can you think of that have evolved continuously for over a decade without becoming an unmaintainable mess?
I work with charities and SF offer an automatic 80% discount on everything which makes them a pretty good option. Why bother custom building a CRUD-type database when you can knock it together in Salesforce using (mostly) point and click?
On the downside: too much marketing fluff (Can't google anything to do with Salesforce coding without hitting loads of marketing bollocks), really bad names/constantly changing names for features (Salesforce1 - anyone know what that is yet?), really complicated licensing if you're trying to figure out how to do X+Y with cheapest licences. And the names of the licences keep changing ....
But technically, they must be doing something very right under the hood. And while I'm at it, I like their core API, its basically an ORM.
PS: Salesforce1 is their new platform that bridge web and mobile. It's not just one or two thing. It's everything. It's how they build Salesforce...
I absolutely hate dealing with vendors who use Salesforce.com. It's a red flag that support might be dismal. The only lesson to be taken from this is that you can build a business around a terrible product.
Among other duties, I am in charge of maintaining the Knowledge Base for our product. SF's Knowledge article capabilities, however, are abysmal. The default formatting is atrocious, barely above plaintext. SF will allow you to try to "jazz it up" by turning on HTML mode. However, due to the incredibly small character limit per article, trying to insert any more than a few paragraphs with proper HTML formatting will quickly exhaust that limit (the HTML tags themselves are counted as part of the character limit). Something as simple as creating a Table of Contents for a longer article hence becomes a nightmare in implementation.
To boot, those articles do not have static links that I can disseminate, meaning that when I'm trying to point people to a particular article, I have to tell them "Go to the Knowledge Base and search for 'X'." This is, frankly, embarrassing to put on something like a Newsletter.
In the effort to integrate two distinct factions in the business, they decided the best approach was not to find one holistic tool that works for both cases, nor unify the process across the company but to do the following crime:
Integrate an on-site established JIRA processes with cocked up Salesforce lack of processes with bidirectional escalation and updates!
So now the shitball of chaos and incompetence that is support and sales teams can directly lay turds into the development team's toolchain, not that they're any better in the majority. Likewise the other way, the developers can write internal swearing and technical information all over salesforce cases which can get published to the clients all the time...
And yes we're doing the same with the knowledgebase but we're pushing it and release notes from JIRA into it.
Now I'm no fan of JIRA the money-sink by any means, as a guy who supported it for a number of years, but neither platform is a good match for a well established company devoid of any reasonable processes and going through a rapid reinvention phase.
Ugh.
If you wanted to find out why SF.com worked, you'd need to write a large book. You'd need chapters on each key individual, each key decision, the environment they were operating in (how lucky they were), the risks, the interpersonal conflict and cooperation. There's at least a lifetime of work gone into SF, split across hundreds of people+.
For example, one high-profile guy I know isn't about to blog about how he made his money screwing over his cofounders, or his ruminations about how incompetent his employees think he is. :)
If that's what counts as aspirational around here, I'm in the wrong job.
Salesforce were pretty courageous back in '99 basing their entire strategy around SaaS.
Consider that one of the jobs most frequently posted on HN is for ZeroCater which seems to love the tagline that they are "help[ing] feed the world." In fact they exclusively deliver food truck meals to startups in SF & the Valley.
Nothing wrong with that, but "feed the world" is beyond hyperbole.
The "making the world a better place" statements of many startups is embarrassing. Silicon Valley parodied this trend best: https://www.youtube.com/watch?v=IXuFrtmOYKg
Personally, it keeps me up to date on the dozen or so people I really care about, but don't have the time to call/email every day. When I speak to those people, we don't have to waste time "catching up".
For others, I'm sure it has many other life-changing, world-changing benefits. Maybe it's not ending wars or feeding the poor, but it can be a great scientific resource, emotional comfort, and organizing platform.
They're also now working on delivering internet to lots of people, preventing suicide, and probably other stuff I don't know about.
So don't judge ZeroCater's endgame when they're not even close.
(All of this is coming from someone who hates frivolous Silicon Valley culture and would normally love to pile on. But in this case I really can't.)
It's hype, it's marketing bullshit, it's cliche, it's premature, and IMO it speaks to a lack of grounding and a focus on the wrong things.
One of the people I respect most in this world has worked for SAP since the bubble and the only reason he's stayed is because he has yet to "find a bigger lever".
It took the resolve of Marc, Parker, Dave, and Frank to stay true to their mission — to upend the software industry. They heard what their customers were asking for, but delivered something better. They refused to build an “on-premise” version of their product. And if they had, it would have prevented them from building a platform for SaaS applications.
Why is that? I can install also applications for the Android and iOS Platforms "on premise".
I would suggest people to view Salesforce from the perspective of the business users (or check this comment as well: https://news.ycombinator.com/item?id=9184356)
The place I work for at the moment uses Salesforce and we're paying lots of money to them (company with 100 people employees, balanced between engineering, support, sales/bdr, marketing, hr). I've never heard our Sales/SE/Support complained about Salesforce. In fact, they continue to find better tools/vendors inside Appexchange that can help them to be more productive closing Sales.
Our software integrate with Salesforce to some extend w.r.t to tracking customer acquisition, license, and retention. I can't complain. Salesforce is better than any other Enterprise software I've dealt with before.
I had a chance to talk to the CTO of Salesforce1 yesterday while he was demo-ing the Salesforce1 platform on his iPhone. He showed me how Sys-Admins (sys-admins of Salesforce themselves) can write a small "module" that can be deployed and visible inside the Salesforce1 mobile app immediately.
That's probably one of the few "enterprise" mobile apps I've seen so far where you can allow your customer to develop a small-modular-app to be deployed within the _same_ Salesforce1 mobile app but only visible to your Organization (or to certain group/role within your Organization).
Our customers purchased our solution (we are an APM company) to monitor their Salesforce instances because they need a 3rd-party to ensure Salesforce does not break their SLA. (We're talking about Fortune 500 company that signed $10m contract with Salesforce; Salesforce will be held accountable if they break SLA).
I asked him how Salesforce test Salesforce itself (the whole thing, that big ball of mud if you want to call it that way). He went off and explain how they rigorously test the whole system in details. That includes testing the mobile app... and testing the "custom app" within Salesforce ecosystems.
So I got curious and asked him further: how do you test those custom apps written by the customers? His replied was as follows:
1. When the customer sign up, they agree to let Salesforce run test-automation of their custom app
2. Salesforce tries their best to educate "Salesforce Developer" to write automation test that will be executed as part of Salesforce update/deploy cycle
3. If the customer/vendor does not do that (or write tests that don't test properly), it's their fault.
That's insane! I've never seen an enterprise software company go that far!
During the event, I get the impression that Salesforce cares a lot of automation, automation testing, and to ensure "things don't break". Can't say the same thing with Facebook even though they're developing lots of "cool" cutting edge tech. But then again, Salesforce customers force Salesforce to not break things or else refund...
TLDR: Salesforce as an enterprise company is pretty good compare the rest. HN community is definitely not the target market so it's natural to see negative comments.
Typical Salesforce propaganda. I stopped reading there.