The big industrial automation and SCADA companies all provide some level of application engineering/system integration services. But, it's only practical to do the integration work in-house for large (millions of dollars in revenue annually or tens of millions of dollars in revenue one time) customers. Managing thousands of local teams of integrators all over the world to take on smaller jobs would be prohibitively expensive.
For smaller, "one-off" jobs, the integration work is done by distributors or by 3rd party integrators. Distributors will often do the integration for free and cover the integration cost and their profit from the discount the manufacturer gives them from list price. Most integrators do fixed-price bids for work, and may also make money from equipment markups. In both cases, there is a lot of incentive to do the minimum possible, especially since the projects tend to be poorly specified.
This is all made worse by the fact that the customers tend to be technically unsophisticated. That makes it hard for them to effectively manage projects, and hard for them to make informed judgements when selecting suppliers. Suppliers are usually picked based on personal relationships with the sales team (manufacturer's or distributor's) and the in-house engineer's familiarity with a given supplier.
Finally, the whole industrial automation industry isn't terribly glamorous. The typical problems being solved on any given job have been solved thousands of times before. The technology is often old and clunky (the most common language is called ladder logic... look it up, it's good for a laugh). Being successful requires a mixture of software, electrical engineering, mechanical engineering, and sales skills. Since most distributors and integrators live almost hand-to-mouth, sales skills tend to be emphasized, even among the engineers. The engineers who are good at sales find they can make more money doing sales. The ones who aren't salesy find there isn't much room for advancement and move on. I'm over-generalizing, but the overall trends don't encourage high-quality software engineering.
For smaller, "one-off" jobs, the integration work is done by distributors or by 3rd party integrators. Distributors will often do the integration for free and cover the integration cost and their profit from the discount the manufacturer gives them from list price. Most integrators do fixed-price bids for work, and may also make money from equipment markups. In both cases, there is a lot of incentive to do the minimum possible, especially since the projects tend to be poorly specified.
This is all made worse by the fact that the customers tend to be technically unsophisticated. That makes it hard for them to effectively manage projects, and hard for them to make informed judgements when selecting suppliers. Suppliers are usually picked based on personal relationships with the sales team (manufacturer's or distributor's) and the in-house engineer's familiarity with a given supplier.
Finally, the whole industrial automation industry isn't terribly glamorous. The typical problems being solved on any given job have been solved thousands of times before. The technology is often old and clunky (the most common language is called ladder logic... look it up, it's good for a laugh). Being successful requires a mixture of software, electrical engineering, mechanical engineering, and sales skills. Since most distributors and integrators live almost hand-to-mouth, sales skills tend to be emphasized, even among the engineers. The engineers who are good at sales find they can make more money doing sales. The ones who aren't salesy find there isn't much room for advancement and move on. I'm over-generalizing, but the overall trends don't encourage high-quality software engineering.